-------
                                                            Chapter 4
   Plant Uptake.  Organic and inorganic contaminants are taken up into the
 plant from the soil.  Inorganic nutrients and organic contaminants are then
 either stored or metabolized within the plant through phytodegradation. In
 phytoextraction, metals or radionuclides are removed from the soil and accu-
 mulated in the aboveground portion of the plant. This part of the vegetation
 is harvested and disposed or processed for metal reclamation. Reviews and
 discussions of phytoextraction are provided by Chaney (1983); Cunningham,
 Berti, and Huang (1995b); Salt et al. (1995); and Cornish et al. (1995). The
 processes involved in phytoextraction are illustrated in Figure 4.17.
   All plants accumulate metals necessary for growth, such as Fe, Mn, Zn,
 Cu, Mg, Mo, and possibly Ni. Some plants accumulate metals that are not
 necessary for growth, such as Cd, Cr, Pb, Co, Ag, Se, and Hg (Salt et al.
 1995). Metallophytes are plants that can grow only in metal-rich soils
 (Baker, Brooks, and Reeves 1988).
   Metallophyte plants have long been sought as an indicator of ore in min-
 eral prospecting (Kumar et al. 1995). Herbarium specimens have been ana-
 lyzed for metal content, and as far back as 1924 high levels of lead (1300
 J4g/g) were detected in a plant (Armeria maritima van halleri) (Reeves and
 Brooks 1983). Hyperaccumulators are metallophytes that are capable of
 accumulating an exceptionally high level of a metal, reaching a specified
 concentration within the dried plant tissue, typically the leaf (Brooks et al.
 1977; Baker 1995).  The concentration criteria of hyperaccumulators for
 metals in leaf dry matter are 100 jog/g (0.01%) for Cd; 1,000 pg/g (0.1%) for
 Co, Cu, Pb, and Ni; and 10,000 |0g/g  (1.0%) for Mn and Zn (Reeves and
 Brooks 1983; Baker 1995). Hyperaccumulatoirs have also been defined as
 those plants containing a toxic element at concentrations 100 times greater
 than that in other plants growing in the same soil (Cornish et al. 1.995). Ad-
 ditional metals, metalloids, non-metals, and radionuclides that have been
 found in hyperaccumulators include Cr and Hg; Se  and As; B; and 90Sr,
 137Cs, 239Pu, and 238-234U respectively (Cornish et al. 1995; Kumar et al.  1995;
 Salt et al. 1995).                                                ;
   Examples of hyperaccumulators include Thlaspi caerulescens (alpine
pennycress) for Ni and Zn (Brown et al. 1994), Thlaspi rotundifolium for Pb
 (Kumar et al. 1995), Alyssum wulfenianum for Ni (Reeves and Brooks
 1983), and Brassica juncea (Indian mustard) for Pb, Cr, Cd, Ni, 90Sr, B, and
Se (Salt et al. 1995). T. rotundifolium grown in mine tailings and river grav-
els contaminated by lead-zinc mines  contained up to 8,200 (Og/g lead
(0.82%) and 17,300 |Jg/g zinc (1.73%) in the dried leaves, the highest ever
reported for any flowering plant (Reeves and Brooks 1983). The extent of
hyperaccumulation is defined by the phytoextraction coefficient, which is a

                                4.101

-------
Soil Treatment Systems
                                     Figure 4.17
        Processes Involved in Phytoremediation of Contaminated Soil
  A. Phytoextraction Processes
 B. Phytodegradation Processes
I"
:
?
A
».

1
• — > Harvesting
I
Translocation
to Shoots
I
	 Root Uptake •*—

Thermal, or
t Chemical Processing
Reclamation or
Disposal
Soil Amendments
	 Increase Availability
of Contaminant
to Root Uptake
	 Harvesting     •*-
 or Reincorporation
                                                                Sequestration/
                                                                 Metabolism

j
t
-Jp~ Transit
Vl.^— ' '
Degradation Vr
in Rhizosphere "
k. Incorporation
tcation
1
Metabolism
T Seques
1 i
                                 Desorption of
                                 Contaminant from Soil
                                                                  Root Update
                                                             >. Microbial Degradation
C. Phytostabilization Processes
Plant Cover Chosen to
Reduce Erosion, Human
Contact, and Decrease &%-^
Environmental Impact t^Jl/^
Incorporation of Soil ~^lv^
Amendments that Sequester ' 1
Contaminant into Soil Matrix
and Aid in Plant Growth
1
1
Leaching and Plant Uptake Reduced
^ Plants Chosen for Poor
Contaminant Translocation
- t
Sequestration in Roots
t
Nonsequestered Contaminant
Source: Cunningham, Berti, and Huang 1995b
                                       4.102

-------
                                                             Chapter 4
quantitative measurement of the degree of concentration of the metal in the
plant defined as the ratio: |jg metal/g dry weight tissue to the jjg rnetal/g dry
weight soil. Phytoextraction coefficients determined experimentally for B.
juncea were 58 (Cr6*), 52 (Cd2+), 31 (Ni2+), 17 (Zn2+), 7 (Cu2+), 1.7 (Pb2*),
and 0.1 (Cr3*) (Kumar et al. 1995).
   Few plant species are hyperaccumulators. One tabulation of
hyperaccumulators indicated more than 300 taxonomic groups (taxa) for Ni,
26 for Co, 24 for Cu, 18 for Zn, 8 for Mn, 5 for Pb, and 1 for Cd. The most
significant plant families containing hyperaccumulators include the
Brassicaceae, Euphorbiaceae, Asteraceae, Lamiaceae, and
Scrophulariaceae (Baker 1995); plants within each family accumulate a
variety of metals.                                                 ;
   Indicator plants do not accumulate metals to the same extent as
hyperaccumulators.  Instead, they are plants whose metal concentrations,
acquired through passive uptake, reflect the soil concentration (Baker 1995).
Silene vulgaris (bladder campion) is an example of an indicator plant. Indi-
cator plants do not effectively concentrate metals for phytoextraction but can
tolerate high concentrations of.metals and therefore can be used to stabilize
soils at sites prone to erosion (Brown et al. 1994).
   Phytoextraction is complex, but Salt et al. (1995) have provided some
explanations of the mechanisms. For root uptake, the plant solubilizes
bound metals in the soil by secreting metal-chelating molecules
(phytosiderophores) or metal-chelating proteins related to metalloithioneins
or phytochelatins.  Plants also solubilize bound metals by using specific
plasma membrane-bound metal reductases or by acidifying the soil through
proton release. Soluble metal ions are transported into the root cells via
specific or generic transmembrane ion carriers or channels. In the transloca-
tion of metals from the roots to the shoots, the metal ions move from root
cells through intracellular pathways across the Casparian strip into the xy-
lem.  Once in the xylem, they are transpired, assisted by metal-chelate com-
plexes.  Metals may also be transported in the phloem (Salt et al. 1995).
Despite these translocation mechanisms, the roots generally contain higher
metal concentrations than the shoots. An upper limit to the metal concentra-
tion within the root is known to occur. For example, root uptake of lead by
plants grown in hydroponic solution reached a maximum concentration and
did not increase further as the lead concentration of the solution increased
(Kumar et al. 1995).
   Plants resist heavy metals by limiting cellular uptake, detoxifying the
heavy metals, or developing heavy-metal-resistant metabolisms. The, most
likely resistance mechanism is detoxification by chelation and accumulation

                                4.103

-------
Soil Treatment Systems
within the vacuoles by compartmentalizatiorji or precipitation (Salt et al.
1995). Metal exclusion is by reduction of metal uptake, restriction of
root-to-shoot transport, and accumulation in roots and shoots (Baker 1995).
It is not clear if a plant's tolerance to one metal will induce tolerance to an-
other metal (Reeves arid Brooks 1983). Other mechanisms may prevent
adverse effects of metals on plants.  Lead, which is generally toxic to plants,
is usually not accumulated in plants, possibly due to precipitation of lead as
sulfate at the plant roots (Reeves and Brooks 1983).
   Plants can also take up radionuclides froni soil. Corn, sunflower, and tall
fescue took up ^Rn and 226Ra (Lewis and MacDonell 1990); "Tc was ab-
sorbed by barley (Sheppard and Evenden 1985); and 244Cm and 23XPu have
additionally been identified as being accumulated in trees (Entry et al. 1997).
Removal of "7Cs and ^Sr from contaminated soil near the Chernobyl
nuclear reactor in Ukraine was accomplished using B.juncea (Adler 1§96).
Uranium and thorium have also been extracted from contaminated soil near
uranium mines by harvesting high-yield plants (Koganov and Dawson 1995).
Phytoremediation of radionuclides may be appropriate for 137Cs and  ^Sr, as
there is  significant evidence of the accumulation of these radionuclides in
vegetation (Entry et al. 1997). There is less evidence for the feasibility of
phytoremediation of other radionuclides, and few field trials have been con-
ducted compared to phytoremediation of metals or organics.
   In addition to metals, metalloids, radionuclides, and nutrients, plants can
take up  organic compounds from solution. At least 70 organic chemicals
representing many classes of compounds were taken up and accumulated by
88 species of plants and trees (Paterson et al. 1990), although the mecha-
nisms and locations of uptake varied. The degree of uptake varies as a func-
tion of plant species and the properties of the chemical compound, such as
polar vs. non-polar, solubility, and kow. Low solubility compounds are less
available for transpiration and compounds with intermediate log kow  are more
likely to be translocated within the plant. Contaminants such as BTEX,
TCE, and chlorinated benzenes can be taken up by vegetation and
phytovolatilized. Plant uptake of some contaminants may be uncertain. For
example, a study of soil spiked with PCP indicated that after 155 days in the
presence of grass, 36% of the PCP was associated with the plant (21% in
roots, 15% in shoots)(Qiu et al. 1994). Other studies, however, found mini-
mal uptake of PCP by several plants (Bellin and O'Connor 1990) or insig-
nificant uptake of PAHs by grasses or alfalfa (Schwab and Banks 1994).
   Phytovolatilization and hydrologic control depend on the uptake of con-
taminated water by the plant. In root uptake, water that may contain organic
chemicals is absorbed from the soil solution into the outer tissue of the root.

                                4.104

-------
                                                             Chapter 4
 The chemical moves through the epidermis to and through the Casparian
 strip, and then through the endodermis, where it may be sorbed, bound, or
 metabolized. Generally, root uptake and bioconcentration within the root is
 inversely proportional to water solubility or directly proportional to octanol/
 water partition coefficient.  Highly hydrophobic compounds tend to accumu-
 late in roots. The chemicals or metabolites that do pass through the endoder-
 mis and reach the xylem are then transported in the transpiration stream or
 sap. These compounds may react with or partition into plant tissue, be de-
 graded, or be released to the atmosphere through stomatal pores via
 phytovolatilization (Paterson et al. 1990; Shimp et al. 1993).
 Phytovolatilization of TCE has occurred in poplar trees;  the TCE released by
 transpiration may break down quickly in sunlight (Adler 1996).  Plants can
 also transpire and volatilize selenium, first converting it from the more toxic
 selenate form into the less toxic dimethylselenide gas (Adler 1996; Banuelos
 et al. 1997). Phytovolatilization of mercury has also been studied; the more
 toxic forms of mercury are converted to elemental mercury and released as a
 gas (Watanabe 1997).
   A root concentration factor for non-ionized compounds taken up by plants
 with non-woody stems is defined as the concentration in roots divided by the
 concentration in the external solution. Concentration factors have been simi-
 larly defined for the transpiration stream and the stem. Correlation equations
 have been developed to relate the octanol/water partition coefficient to these
 concentration factors. The bioconcentration factor (defined on dry weight
 basis) is the ratio of the concentration in aboveground plant parts to the con-
 centration in the soil (Paterson et al. 1990).
   The primary phytoremediation processes for organic compounds that are
 taken up by plants are phytodegradation (metabolism within the plant) and
 phytovolatilization (release of the volatile compound or metabolized prod-
 uct) through the stomatal pores. Phytoextraction, or plant harvesting, of
 organic compounds is unlikely and has not generally been investigated. The
 harvesting of roots would be infeasible to remediate  highly hydrophobic
 compounds, such as PCBs and DDT that are concentrated in the roots due to
 their log k^, being too high  to allow translocation within  the plant.
   Stimulation of Biodegradation. Plant roots may stimulate contaminant
biodegradation by increasing the microbial populations and activity in the
rhizosphere; the roots contribute exudates to the soil that serve as microbial
substrates.  These exudates include sugars, amino acids, organic acids, fatty
acids, sterols, growth factors, nucleotides, flavanones, enzymes, and many
other compounds (Shimp et al. 1993; Schnoor et al. 1995). Leachate from
roots of specific plants contained compounds that stimulated the growth of
                                4.105

-------
Soil Treatment Systems
PCB-degrading bacteria (Fletcher, Donnelly, and Hegde 1995). The pro-
cesses involved in rhizodegradation are shown in the phytodegradation por-
tion of Figure 4.17. Rhizodegradation is applicable to organic contaminants
that remain in the soil without being taken up by plants (including relatively
strongly-sorbed or hydrophobic compounds), but can also apply to contami-
nants prior to plant uptake.
   Alfalfa plants and associated microorganisms promoted biodegrada-
tion rather than volatilization of organic substances from groundwater
(Davis et al. 1994). Some enhancement of rhizosphere biodegradation
has been found using various grasses in soils contaminated with PAHs
(April! and Sims 1990; Schwab and Banks 1994), in rhizosphere soil
contaminated with the pesticides atrazine and metolachlor (Anderson
and Coats 1995), and in TCE-contaminated soil (Anderson and Walton
1995). Radiolabelled PCP was used to examine mineralization in sys-
tems  planted with crested wheatgrass (Agropyron desertorum). After
155 days, 22% of the PCP was mineralized in the planted system, but
only 6% in unplanted controls. PCP mineralization rates in soil were
23.1 mg /kg in 20 weeks in the planted system, and 6.6 mg/kg soil in 20
weeks in the unplanted controls (Ferro, Sims, and Bugbee 1994). Min-,
eralization of pyrene  also occurred sooner in systems planted with
crested wheatgrass than in unvegetated controls (Ferro et al. 1994).
    ' ., " .'''•••• I         ' .. .•  ...   I'!  *' ..''.'.)        . '            I
   Modification of Soil Environment. Phytostabil.ization is the use of
metal-tolerant plants to reduce erosion of metal-contaminated soil or to re-
duce leaching  of metals by converting them from a soluble to an insoluble
oxidation state (Salt et al. 1995). For example, the leaching of metals from
soil was reduced by greater than 98% using B^juncea (Raskin et al. 1994).
Plant root exudates may change the soil pH, which can impact the dissocia-
tion of organic compounds or affect metal solubility and mobility.
Phytostabilization also refers to the incorporation of organic compounds into
the plant lignin (Cunningham, Berti, and Huang 1995b). Phytostabilization
processes are summarized in Figure 4.17.
                               •  !  ;:., '•'.     I     ,    ." ''      .     I "<:'
   The term phytodegradation can also refer to the degradation of soil con-
taminants through the  chemical action of enzymes released by a plant
(Cunningham, Berti, and Huang 1995b). This form of phytodegradation is
not dependent on microbial activity, arid since plants can grow in the absence
of soil microorganisms, it may possibly occur in sterile soils. The extent of
this effect will depend on the transport of the enzymes to the contamination
site. If the enzymes are not strongly sorbed, they may be transported
through the soil so that degradation can occur in locations away from the
actual plant root.  However, by analogy to the rhizosphere effect, this effect
                                4,106

-------
                                                             Chapter 4
 may be limited to near-root areas. Phytodegradation processes are also
 shown in Figure 4.17
   Indirect Effects on Soil. The presence of a plant that does not directly
 degrade or remove a contaminant may still be advantageous for othe^ reme-
 dial activities, assuming the plant can survive the contamination. Plants
 contribute organic matter that improves the tilth and water-holding capacity
 of the soil. The openings formed by plant roots can act as distribution chan-
 nels into the subsurface for nutrients, water, or oxygen applied at the surface
 (Shimp et al. 1993).  Thus, infiltration of nutrients, water, or oxygen may be
 aided by the presence of appropriate plants over a large area.

 4.5.1.2 Water
   Contaminant Immobilization or Removal. Plants may be used in natural
 systems (wetlands), engineered systems (constructed wetlands), or ex-situ
 systems (reactors) to remediate groundwater, surface water, or wastewater
 through degradation or accumulation and removal of the contaminants.
 Rhizofiltration is the use of plants to withdraw metals from aqueous solu-
 tions (Raskin et al. 1994; Dushenkov et al. 1995). Mechanisms that plants
 use to remove toxic metals  or radionuclides such as Pb, Cd, Zn, Cr, Mn,
 U,and MSr from solutions include the nonbiological process of surface sorp-
 tion (due to chelation, ion exchange, and specific adsorption) and biological
 processes (intracellular uptake, vacuolar deposition, and translocation to the
 shoots, in addition to and root-mediated precipitation) (Salt et aL 1995). For
 example, lead was removed via tissue absorption and by root-mediated pre-
 cipitation of insoluble lead phosphate (Raskin et al. 1994; Salt el: al. 1995).
 A bioaccumulation coefficient is used to describe the extent of rhizofiltration
                a
 and is defined as the ratio of the metal in the plant Qug/g dry weight) to the
 initial solution concentration of the metal (Salt et al. 1995).
   Terrestrial or aquatic plants can be used in irhizofiltration of metals and
 radionuclides. Hydroponically-grown sunflower roots reduced water con-
 centrations of Cr6*, Mn, Cd, Ni, Cu, U, Pb, Zn, and '"Sr (Salt et al. 1995).
 Roots of plants grown hydroponically concentrated Cr,  Cr6*, Cd, Ni, Zn, and
 Cu from wastewater, with lead having the highest bioaccumulation coeffi-
 cient and zinc the lowest (Raskin et al. 1994). The aquatic plants, water
 hyacinth (Eichhornia crassipes), pennyworth (Hydrocotyle umbellata),
 duckweed (Lemna minor), and water velvet (Azolla pinnata) took up Pb, Cu,
 Cd, Fe, and Zn from solution (Salt et al. 1995).
   A few field trials of phytoremediation for radionuclides have been pon-
ducted.  Rhizofiltration of 137Cs and phytoextraction of ^Sr were tested using
                                4.107

-------
Soil Treatment Systems
a raft of sunflowers in a contaminated pond near the Chernobyl site.  The
immobilization of the ™Cs in the roots and the translocation of the ^Sr into
the shoots indicated that these phytoremediation methods may be feasible
for remediation of water contaminated by radionuclides (Adler 1996). Sun-
flowers were also used in a successful trial of remediation of uranium-con-
taminated water at a D0£ uranium-processing plant in Ashtabula, Ohio, with
further tests planned for uranium-contaminated water at the Oak Ridge Na-
tional Laboratory (Adler 1996).
   Wetland plants can be used in engineered oir constructed beds to take up
or degrade contaminants. Aquatic macrophytes were used to remove chro-
mium in contaminated wastewater (Vajpayee et al. 1995). Accumulation of
arsenic and zinc was also observed in the rhizosphere of wetland plants
(Otte, Kearns, and Doyle 1995). Various aquatic macrophytes were evalu-
ated for the removal of nitrogen and phosphorus (Reddy and De Busk 1985).
   Water contaminated by organic compounds can also be treated using veg-
etation.  The aquatic plant parrot feather possesses the enzyme that enhances
degradation of wastewater containing TNT. Parrot feather decreased TNT
from 128 ppm to 10 ppm in flooded mesocosms of soil (Schnoor et al.
1995). Enzymes derived from horseradish roots were used in ex-situ reac-
tors to treat water contaminated with chlorinated phenolic compounds (Dec
and Bollag 1994).
   Hydrologic Control  Contaminated groundwater and vadose zone water
movement can be contained and controlled using plants and trees. Vegeta-
tive caps can be used for landfills, sediments, sludges, or contaminated soils
as a pollution  prevention method. The uptake of water by vegetation can
prevent infiltration of rainwater or vadose zone water into the contaminated
media and can prevent leacheate formation. Vegetative caps can also be a
remediation mechanism as well as a containment mechanism. The vegeta-
tion in the cap may remediate the contaminants through rhizodegradation,
phytodegradation, or phytovolatilization. Hydrologic control can also be
used as a treatment method in conjunction with phytovolatilization.  For
example, contaminated groundwater or leachate can be pumped and applied
to plants or trees for irrigation (Gatliff 1994; Schnoor et al. 1995). Cotton-
wood  and hybrid poplar trees were  used at seven sites in the East and Mid-
west to extract and treat shallow groundwater contaminated with heavy met-
als, nutrients,  or pesticides in conjunction with' pump-and-treat  systems. The
trees apparently contributed to water table drawdown by pumping 190 to
 1330 L/day (50  to 350 gal/day) per tree: Tree growth was greater and nitro-
gen uptake was faster the closer the root ball was planted to the shallow
groundwater (Gatliff 1994). Although plant root growth may be restricted to
                                 4.108

-------
                                                             Chapter 4
 above the water table, the roots can still influence contaminants in the
 groundwater.  Barley roots were confined to the oxidized zone of the soil,
 which occurred about 10 cm (4 in.) above the water table interface, but could
 still absorb Fe, Tc, U, and P that diffused upward from the capillary fringe
 (Sheppard and Evenden 1985).

 4.5.1.3  Additional Applications
   Plants can be used to reclaim degraded mining land and stabilize soil for
 land application of wastes.  Uptake of contaminants by crop plants and the
 potential adverse effects on the food chain were of concern in managing land
 treatment sites and the motivation for research into uptake mechanisms
 (Chancy 1983).
   Plants useful for the restoration of wetlands that provide wildlife habitat
 can also remediate contaminated water in the: wetlands.  A natural salt marsh
 contaminated with heating oil was restored using Spartina altemiflora
 (smooth cordgrass).  The most abundant bacteria in the wetland, were those
 that could metabolize naphthalene as a sole carbon source.  The plant pro-
 vided oxygen to its root zone, so the addition of these plants, along with
 fertilizers, could have stimulated aerobic biodegradation of the heating oil in
 the root zone (Bergen and Levandowsky 1994).
   Plants can be used as bioassays to assess soil or groundwater contamina-
 tion or the effectiveness of other remedial activities. Phytotoxicity testing
 was  used to evaluate the degree of bioremediation of a contaminated soil
 (Baud-Grasset, Baud-Grasset, and Safferman 1993).  Sandhu et al. (1991)
 used an on-site plant (Tradescantia) assay to test for levels of As, Cr, and Cu
 in soil; changes in the micronucleus of the plant correlated with the contami-
 nant levels in the soil. Data from tree rings were used to determine the di-
 rection and velocity of a chloride plume in groundwater near a landfill
 (Vroblesky and Yanosky 1990) and to correlate with concentrations of nickel
 in a groundwater plume near a landfill and stainless steel plant (Yanosky and
 Vroblesky 1992).
   Plants can be useful in hazardous waste site assessment as indicators of
 site conditions (i.e., the presence and/or concentrations of soil contaminants)
 since the visual appearance of the vegetation may reveal overall plant health
 and associated ecological stresses.  Remote sensing and historical air photos
can be used in a more elaborate site assessment of vegetation (Nicholson and
 Safaya 1993).  However, correlating plant appearance or health with soil
contaminant levels may be difficult or limited to areas of severe contamina-
tion or abrupt boundaries since subtle differences in vegetation health or
appearance may be due to factors unrelated to the soil contamination.

                                 4.109

-------
Soil Treatment Systems
4.5.1.4 Site, Soil, and Waste Characteristics
   Characteristics of the site, soil, and waste that have to be considered when
evaluating me applicability of phytoremediation include:
        •  site characteristics, such as climatic conditions (temperature,
          precipitation, and length of growing season), open space
          'availability and exposure to sunlight, and disturbance from
          site activities;
                                  •j  	|"        ,             • '(•
        •  soil characteristics including chemical characteristics (pH, nutri-
          ent content, organic carbon, acidity, salinity, etc.) and physical
          characteristics (water content, air-filled porosity, hydraulic con-
          ductivity, structure, texture, bulk density, tilth, etc.); and
                  I' .      •  "" i|. ,. ;,_ ' . ui., !" ; i' '  1 - , | ':'  ',  ,' "' ,  ' ' '»	, , , . ,. . -    I! i', _ -
        •  waste characteristics, sucli as contaminant concentrations, physi-
          cal form, contaminant types (amenable to phytoremediation), and
          contamination depth (not greater than the root zone).
   Proper site and soil conditions are critical to plant growth. For example,
water-logging resulted in  shallow roots and poor growth of buffalograss
(Buchloe dactyloides) and warm season prairie grasses in phytoremediation
test plots established at an industrial site in soils contaminated with rela-
tivelylow levels of PAHs and PCP (Qiu l995a).
   Some plants can tolerate high concentrations of organics and nutrients
better than soil microorganisms (Schnoor et al. 1995).  Toxicity screening of
different organisms indicated that plants (as measured by seed germination
tests) were less sensitive to heavy metal contamination than were bacteria
(Miller et al.  1985). This suggests that if microbial degradation of PAHs or
PCP fails due to metal toxicity, such as in cases where Cr, Cu, and As
wood-preservative solutions are also present, phytodegradation due to secre-
tion of plant-formed, enzymes might succeed. Modification of the waste by
amendments may be required to meet the requirements of growing plants.
Contaminants potentially amenable to phytoremediation include toxic metals
and radionuclides; pesticides (Anderson and Coats 1995); TNT, nutrients,
and BTEX (Schnoor et al. 1995); the chlorinated solvent TCE (Anderson
and Walton 1995); PAHs  (Schwab and Banks 1994); PCP (Aprill and Sims
1990); and PCBs (Fletcher, Donnelly, and Hegde 1995). Phytoremediation
of wastes with mixed contaminants has not generally been addressed in re-
search; however, there is some evidence that plants can tolerate mixed or-
ganic and metal contamination (Schnoor et al. 1995).
                                4.110

-------
                                                              Chapter 4
 4.5.1.5 Pilot Testing
   To determine the applicability of the technology, pilot testing is performed to
 select the plants, determine their growth potential in the contaminated soil under
 the site conditions, and to assess contaminant removal or degradation. Prelimi-
 nary screening of candidate plants can save time and money by determining
 contaminant concentrations that are phytotoxic or inhibitory to geirmination of
 the plant seeds. Prior to field testing, seeds can be placed into the contaminated
 soil in the laboratory to determine the level of contamination that is phytotoxic
 or inhibitory. One biotreatability study was designed to investigate the potential
 for grasses to enhance biodegradation in soil from a Superfund site contami-
 nated with PCP, PAHs, dioxins, As, and Cr. Grass germination and establish-
 ment failed in this study due to elevated concentrations of PCP (SBP Technolo-
 gies, Inc. 1994). In experiments investigating a vegetative cover for soil/waste
 mixtures, incomplete seed germination and grass kill were seen in germination
 studies with three native grasses in soil contaminated with petroleum wastes
 (Overcash et al. 1985). In contrast, laboratory germination studies using soil
 contaminated with PCP did not find any significant detrimental effects on seed
 germination, with buffalograss seed being tolerant to >300 mg/kg PCP (Qiu et
 al. 1994).
   Greenhouse trials can provide further preliminary information on the
 growth patterns of the plants and can be used to screen candidate plants.
 Outdoor test plots should be used to determine the growth characteristics of
 the plant in the contaminated soil under actual site conditions. An estimate
 of the contaminant removal or degradation rate can be developed using mea-
 sured concentrations in the plant or soil. For example, based on biomass
 measurements and metal concentration in the plant, a tested cullivar of B.
juncea could theoretically remove 630 kg Pbyha (562 Ib Pb/acre) if the
 shoots were harvested (Kumar et al. 1995).

 4.5.2  Process Design Principles
   Because phytoremediation is an emerging technology, there is limited
 experience with field applications and the principles of design are not well
 established. However, there are several design considerations that are impor-
 tant in field implementation for soil remediation. The process design prin-
 ciples described here are primarily for soil remediation, although many also
 apply to other contaminated  matrices.
                                 4.111

-------
Soil Treatment Systems
4.5,2.1 Plant Selection
   The selection of useful plants is based on the particular method or goal of
the phytoremediation effort and on criteria! such as the plant's ability to take
up and translocate or degrade contaminants, its growth rate and biomass
production, root characteristics, stress tolerance, and cultural demands. The
cultivation of the plant should not be labor- or cost-intensive.
• .              •           ,  „"•    • • ] ;•••;•   •	.•£••; ,    ,  '. ••:. ' -     '  ••.
   The optimum plant for rhizofiltration and phytostabilization will not
translocate metals from the roots to the shoots because the aboveground
portion of such plants  has the potential for food chain accumulation of met-
als by animals or humans. In addition, the optimum plant for rhizofiltration
will have rapidly-growing roots and the ability to remove heavy metals (Salt
etal. 1995).
   On the other hand, the optimum plant for phytoextraction would tolerate,
translocate, and accumulate high concentrations of heavy metals in the
harvestable, aboveground portion of the plant, in its shoots and leaves. The
aboveground portion of the plant is the desired location of metals since this
biomass will be collected and disposed or treated before food chain accumu-
lation can occur. The  ideal plant for phytoextraction also has a rapid growth
rate and a large biomass (Salt et al. 1995). Although many currently identi-
fied metal hyperaccumulators do not meet these criteria, one favorable can-
didate is B. juncea, a high biomass plant that accumulates lead (Kumar et al.
1995). Screening of different plant species revealed that B. juncea had the
best ability to transport lead to the shoots, with >1.8% lead in the shoots. It
could also concentrate Cr6*, CcP, Ni2+, Zn2% and Cu2+ (Raskin et al. 1994).
However, a large genetic variability was seen in the ability of B. juncea culti-
vars to accumulate Pb, with 0.04% to 3.5% in shoots and 7% to 19% in roots
(Kumar etal. 1995).
   Plants suitable for rhizodegradation of organic contaminants should pro-
duce appropriate compounds or enzymes and should not take up the con-
taminant of interest. The identification of specific plant-formed compounds
that are known to stimulate cometabolism of a contaminant and the screen-
ing of plants for these compounds can increase the success rate of
rhizodegradation (Fletcher, Donnelly, and Hegcle 1995). The optimum
plants for phytodegradation will produce enzymes or have metabolic path-
ways capable of degrading the contaminants. For successful hydraulic con-
trol or phytovolatilization, the plant must take up and transpire a sufficiently
large amount of water. Metabolism within the plant should not result in
more toxic intermediate products that would be released into the atmosphere.
                                                                 i
                                 4.112

-------
                                                             Chapter 4
   Information on phytotoxicity of contaminants can be collected, such as
 was done with the PHYTOTOX database (Fletcher, Johnson, and McFarlane
 1988), to determine the compatibility of a site contaminant and potential
 plant candidates. A survey of vegetation at a site may be useful to identify
 plants that can tolerate the stresses of site conditions.  However, the presence
 of a plant may indicate only tolerance to the soil contaminants and not nec-
 essarily that it has any beneficial effect relative to site remediation, In a
 survey of vegetation at 15 hydrocarbon-contaminated sites, members of the
 Leguminosae family were most common amd the best competitors at these
 sites.  Their presence may have been due to lack of need for soil nitrogen
 that was consumed by microorganisms degrading the hydrocarbons,(Gudin
 and Syratt 1975).
   Root type, shape, and depth are important factors related to plant selec-
 tion for site remediation. Fibrous root systems have numerous fine roots
 spread throughout the soil while tap root systems are dominated by a deep
 central root (Aprill and Sims 1990).  The fibrous root system provides maxi-
 mum contact with the soil and can extend to a 3-m (10-ft) depth in some
 grasses. However, root depth varies greatly among different types of plants.
 For example, trees may be useful for extraction of groundwater less than 9.1
 m (30 ft) deep (Gatliff 1994).
   Valuable information and experience, such as identification of particular
 species appropriate for varied climatic and soil conditions, was gained in the
 use of vegetation in connection with land application of hazardous waste and
 for hazardous waste site covers (Chaney 1983; US EPA 1983; McAneny et
 al. 1985).  Some criteria for establishing a vegetative layer as part of a Sub-
 title C hazardous waste landfill  cap may not be applicable to the vegetative
 or evapotranspirative caps used in phytoremediation. For example, avoiding
 deep tap root systems is desirable in the design of impermeable waste site
 covers to prevent penetration of an engineered cap underlying the vegetative
 cover. It might be desirable,  however, to use such a deep root system in
. some phytoremediation projects that do not Involve impermeable covers.

 4.5.2.2 Cultural Practices
   The proper cultural practices are crucial to establish and maintain healthy
 vegetation. Seeds used at a site should be obtained from the region of the
 site so that the plants are adapted to the climatic conditions characteristic of
 the site. Seed suppliers, agricultural extension agents, and horticultural asso-
 ciations can supply this information. The seeding timing, depth, and  density
 are dependent on the plant species used, as are the planting and cultivation
                                 4.113

-------
Soil Treatment Systems
methods. Some seeds may require special treatment prior to germina-
tion, such as some prairie grass seeds that require storage over winter.
Proper moisture, fertilization, and protection from pests and diseases are
also important during establishment and maintenance of the crop. The
amount and timing of irrigation, fertilizer, and pesticides should be care-
fully considered. Due to the added stress of a contaminated soil envi-
ronment, these factors may have to be carefully monitored if plant culti-
vation at a contaminated site is to be successful.  Complete  soil charac-
terization is important not only for soil bioremediation, but  is crucial for
establishing and maintaining vegetation.

4.5.3  Process Modifications
   To increase the bioavailability of metals to plants for phytoextraction, the
soil can be amended with metal chelates, a moderately acidic soil water pH
can be maintained, and reducing organic acids can be added to  alter the re-
dox status of the soil (Salt et al. 1995).

4.5.4  Cost Data
   Information on phytoremediation costs is not yet readily available due to
limited field applications and commercial experience in most areas. Some
speculative estimates or predictions have been made that vary widely de-
pending on the specific phytoremediation process used. Annual cropping
system costs of $200 to $10,000/ha ($80 to $4,050/acre) would represent
costs of $0.02 to $1.00/m3 ($0.015 to $0.76/yd3) for phytoremediation as-
suming a 1-m (3-ft) depth of roots (Cunningham, Berti, and Huang 1995b).
Costs for groundwater control and treatment using trees were estimated to be
less than for pump-and-treat methods in a 6-m (20-ft) deep aquifer (Gatliff
1994).  Predictions of the total treatment cost for phytoremediation of
metal-contaminated soils were $16 50 to $55/metric ton ($15 to $50/ton)
compared to $110 to $220/metric tori ($100 to $200/ton) for other treatment
methods (Raskin et al. 1994). A cost estimate to remediate 0.4  ha of 50-cm
(1 acre of 20-in.) thick sandy loam using phytoextraction was $60,000 to
$100,000 versus a minimum $400,000 for just excavation and storage of this
soil (Salt et al. 1995). Cornish et al. (1995) have prepared comparative cost
estimates for phytoremediation and soil washing for a hypothetical case of a
50-cm  (20-in.) thick layer of Cd-, Zn-, and 137Cs-contaminated sediments
from a 0.5-ha (1.2-acre) chemical waste disposal pond (Table 4.11). Al-
though the cost estimates are speculative, they do indicate the potential sav-
ings of phytoremediation over other site remediation technologies.
                                4.114

-------
                                                                            Table 4.11
          Cost Estimate for Phytoremediation vs. Soil Washing of Sediments in a Hypothetical Chemical Waste Disposal Pond
                                                        Cost Estimate $
                  Activity
                                     Phytoremediation *
                      Soil Washing
                                      Summary of Activity Components
£>.
~+l

cn
Site preparation


Production/processing


Waste management/ disposal




Project closeout


Environmental, safety, & health
(ES&H), engineering design (ED),
and project management overhead
(PMO)

Overall estimated cost

Costs per tonneb of sediment
 63,000


395,000


 35,000




 18,000


153,000
                                               664,000

                                                 200
 61,000


1,336,000


 175,000




 62,000


 490,000
                        2,124,000

                          600
Surveying, staking, and light clearing of area, access construction, excavation,
installation of water supplies, and construction of fences and other structures.

Operation and maintenance of treatment system for the duration of the project,
including heavy equipment rental and services.

Consolidation of waste materials and haulage of same to eventual storage facility,
as well as necessary testing prior to disposal. Costs for long-term storage have not
been calculated due to uncertainties regarding the ultimate disposition of some
wastes.

Demobilization of heavy equipment, removal of constructed facilities, and site
restoration

Calculated as 30% of above costs
                 Subtotal plus ES&H, ED, and PMO costs

                 Based on treating 2,500 m3 of sediment @ 1,420 kg/m3
      a. Costs shown for Phytoremediation assume a production period of 20 years.
      b. Denotes units of metric tons (1,000 kg per tonne)

      Source: Cornish et al. 19S5
                                                                                                                                                                  o
                                                                                                                                                                  :r
                                                                                                                                                                  Q

-------
Soil Treatment Systems
   Costs associated with phytoremediation include work plan and report
preparation, treatability and pilot testing, soil preparation, seeds, planting,
irrigation, fertilization, plant and soil sampling and analysis, and plant dis-
posal costs.

4.5.5  Design Validation
   Full-scale application of all forms of phytoremediation has not yet been
accomplished. Since it is an emerging technology, most work in
phytoremediation has been at the research level in the laboratory or
small-scale field plots.  Research experience with remediation of metals is
discussed in the previous sections. It is possible that more research has been
conducted for phytoremediation of metals than for organic compounds, but
field trials for metals phytoremediation remain limited and large-scale
implementation is not yet achievable, especially for soil as opposed to water.
Significant research has been conducted in some areas of phytoremediation
for organic compounds, such as for TCE in groundwater. Some laboratory
studies have been conducted with contaminated soil from Superfund sites
and field experiments have been conducted at numerous sites contaminated
with organics.
   In a laboratory and greenhouse study,  soil contaminated with PCP and
PAHs was obtained from a Superfund wood-preserving waste site to exam-
ine the effect of the contaminants on the germination and growth of prairie
and other grasses and to assess the potential for rhizodegradation. This
study indicated that rhizodegradation of PCP was likely only at relatively
low PCP concentrations. However, it is believed that this is strongly influ-
enced by site-specific factors. Under other site conditions, rhizodegradation
could occur at lower or higher PCP concentrations. An experimental plot
was established at another wood-preserving waste Superfund site to evaluate
whether PCP degradation could be enhanced in a LTU by rhizodegradation.
Adverse temperature and precipitation conditions at this site greatly in-
creased the difficulty of establishing a healthy vegetative crop (Pivetz, un-
published data).
   A laboratory study conducted with PAH- and PCB-contaminated soil
from an industrial site indicated a statistically significant increase (o=0.05)
in benzo(a)pyrene [B(a)P] biodegradation under a tall fescue crop as com-
pared to an unvegetated control, although B(a)P mineralization was low —
less than 0.3% of the B(a)P was incorporated into the plant (Qiu 1995b).
   Hybrid poplars have been successfully used In full-scale applications to:
(1) decrease nitrate from up to 100 mg/L in surficial groundwater to less
                                 4.116

-------
                                                             Chapter 4
than 5 mg/L, (2) take up 10% to 20% of atrazine applied in a field, and (3)
reduce infiltration at a landfill by capping it with 10,000 trees per hectare
(4,050 trees per acre)(Schnoor et al. 1995).

4,5.6 Permitting Requirements
   Requirements for implementation of phytoremediation include obtaining
regulatory acceptance, which will likely require more effort than for proven
remedial technologies.  The potentially longer time frame for
phytoremediation to be effective may influence regulatory decisions, in a
manner similar to that for intrinsic remediation, which may also take longer
than more active remedial technologies. Reports on a field trial of T.
caerulescens indicated that it would take 13 l:o 14 years to remediate the
heavy metal-contaminated site (Salt et al.  1995) using phytoremediation
technology.  Despite potentially longer time frames for remediation,1
phytoremediation has been studied or proposed for part of the remedial solu-
tion at several Superfund sites, including for TCE-contaminated groundwater
arid for TNT-contaminated soil and surface water.

4.5.7 Implementation
   The potential negative impacts of phytoremediation should be fully evalu-
ated before implementing this technology. The interactions of the plants and
all contaminants at the site should be studied, For example, application of
fertilizer to optimize plant growth may increase the mobility of some metals
in the soil because many common nitrogen-containing fertilizers lower the
soil pH.  If the plant does not take up the metals, they may be leached to the
groundwater. Unintended increased mobility of contaminants and increased
exposure pathways can be limitations of phytoremediation.
   If toxic contaminants are bioconcentrated in the plants, the ultimate fate of
the plant must be considered to prevent chemical or toxin ingestion by animals
or humans. Research on uptake and degradation of metals and organic com-
pounds was initially fueled by concern that vegetation used in land treatment
systems might be a potential hazard in this manner (US EPA 1983).
   The selected plant may be inherently  toxic to other plants (allelopathy),
animals, or humans, and its presence at a site can have risks that are signifi-
cantly greater than the risks posed by the contaminants. Substances ex-
tracted from Datura innoxia (thorn apple)  can be beneficial for removal of
metals in certain situations (Drake et al.  1996); however, this does not justify
widespread planting of this attractive but very toxic plant at a site because
the risk of a child fatally ingesting a leaf is much greater than the risk of the
                                4.117

-------
Soil Treatment Systems
child being harmed by ingesting large amounts of the contaminated soil.
The deliberate introduction of noxious or invasive weeds, such as purple
loosestrife, can have adverse economic, ecological, or aesthetic impacts
outweighing any benefits to site remediation.  The impacts of establishing
any particular plant should be considered in the planning stage. Other con-
siderations include the impact of plants and biomass on other technologies
being used at the site and the potential for plants (such as trees) to be physi-
cal impediments to movement or operations at the site.
   The presence of plants may decrease contaminant biodegradation if the
plant organic matter is used as a carbon source instead of the contaminant.
The presence of debris of the salt marsh plant Spartina alterniflora de-
creased the amount of degradation of oil in laboratory sediment columns,
perhaps from competition for limited oxygen arid nutrients by both indig-
enous oil-degrading microorganisms and by microorganisms degrading the
plant organic matter (Nlblina, Araujo, and Bond 1995).

4.5.8  Performance Evaluation
   Evaluation of phytoremediation performance depends on the type of
phytoremediation process employed, but generally involves determination of
the percent removal of a contaminant or attainment of a remedial goal in the
contaminated matrix. The degree of phytoextraction is easy to assess be-
cause extraction and analysis techniques for metals in plants are straightfor-
ward (Brown et al. 1994).  The concentration in the soil, plant leaves, and
shoots can be measured, and the bioconcentration and accumulation factors
can be calculated. If the biomass production rate is measured using standard
agronomic techniques, the mass and rate of metal removal can be calculated.
   i •' j	i1'  ' ii' ,!,,•"'', "„;',!.. ,j   , j ,, , p,'!' ,i  "f\ :;,,'j,. ,.,;."! '"B hi!1',1, • T ,i n'; | ^j '",* " M'. riij1'   'i';r „ ' ,|  "    • ,
-------
                                                             Chapter 4
contaminated with low levels of TCE from a site naturally vegetated with a
grass, legume, composite herb, and Loblolly pine. However, use of radiola-
belled compounds in field situations is virtually impossible.
   The contaminant may not degrade uniformly throughout the soil since
rhizodegradation is postulated to occur in a discrete zone around the plant
root. Unless the plant roots are distributed uniformly throughout the soil, the
selection of soil sampling locations or the interpretation of analytical results
will have to take into account the location of the sample relative to the root.
Microbial counts or respiration in these soil samples can also provide infor-
mation on the degree of rhizodegradation.  The problem of sample location
would not arise in water treatment systems because influent and effluent
concentrations representative of the larger system can be easily measured.

4.5.9  Research Needs for Further Technology
Implementation
   Interest in phytoremediation has increased greatly since the late 1980s.
University, industry, and government research groups have been active in
examining various aspects of phytoremediation, producing numerous re-
search publications. Interest in phytoremediation has resulted in symposia
related to this technology, focusing on organics (Anderson and Coats 1994)
and metals (Interdisciplinary Plant Group 1995; DOE 1994). Some
phytoremediation technologies have been applied commercially, and infor-
mation on phytoremediation has reached the general public and business
community through articles in newspapers and business magazines. This
last aspect is important as it may lead to increased commercial and regula-
tory acceptance of this technology that would spur even more research ef-
forts and lead to further field-scale applications.
   The use of phytoremediation adds an additional layer of complexity to a
remedial action since it introduces another biological system that has its own
characteristics.  A multidisciplinary approach is needed to plan and imple-
ment phytoremediation, potentially involving ecologists, botanists, horticul-
turists, turfgrass and wetland specialists, and agricultural experts in addition
to hydrogeologists, engineers, microbiologists, and soil scientists.  Special-
ists in each field have valuable knowledge and expertise: for example, for-
estry experts can identify appropriate tree species  or hybrids., and provide
cultural techniques for the trees; botanists can identify useful plant exudates,
identify processes within the plant, and help with plant selection; and soil
scientists and agronomists can provide details on soil characterization and
plant cultural practices.  Much of the information  from these fields has not
generally been applied in the past towards remediation of contaminated soil

                                4.119
-------
Soil Treatment Systems
or groundwater.  This information needs to be distilled into a useful form and
then integrated into a remediation-focused effort. Full use of resources such
as farmers, agricultural extension services, arid even local garden clubs and
nurseries may provide additional information on potentially useful local
plants or cultural practices for application to site remediation problems.
  Research on the applicability of phytoremediiation to wastes with several
types of contaminants would greatly expand the potential use of this technol-
ogy. An important research goal for phytoextraction is to identify
fast-growing and high-biomass-producing plants that have the ability to take
up toxic metals (Raskin et al.  1994; Kumar etaf. 1995).  Genetic modifica-
tion may be necessary to develop such plants. Wackett and Allen (1995)
indicate that research on specific plant-microbe interactions and their effect
On biodegradation may be more productive than randomly screening combi-
nations of plants, soils, and contaminants.
  Extrapolation of lab results to field situations is problematic; therefore,
pilot testing in the field under controlled site conditions will be critical.
Evaluation of rhizodegradation performance may depend on the location of
soil samples because the rhizosphere distribution changes with time and may
not be uniform throughout a site.  Sampling techniques and evaluation of
rhizodegradation still need to be addressed.
  The emerging technology of phytoremediation consists of the use of
green plants for remediation of a wide variety of contaminants in soil,
groundwater, surface water, wastewater, and air. The specific processes and
design considerations can vary greatly depending on the contaminant type
and matrix being treated. The selection and use: of an appropriate plant is
critical. The depth of soil remediation is likely to be limited by the depth of
the plant roots. Practical implementation of phytoremediation will require
more information on techniques, costs, and integration of phytoremediation
with other remedial technologies.  Phytoremediation is the subject of inten-
sive research; however, routine field application and remediation success for
contaminated soils has not yet been achieved.
    -	!:  if	  , •, i •; .  '  :/]:	I  i	; ,., .... i' ,, , .. • v,	-.  !,;,:;!,- '  .  ,   -, ;;";.ji i,,; ,'..•,
                                 4.120
-------
                                                       Chapter 5
       GROUNDWATER TREATMENT
                        SYSTEMS
5.7  Introduction

  Site, soil, and waste constituent characteristics that are important in the
evaluation and design of groundwater treatment system technologies are
summarized in Table 5.1. Technologies applied in situ (i.e., enhanced biore-
mediation, intrinsic remediation, biosparging, and migration barriers) are
affected by site conditions (e.g., groundwater table fluctuations, heteroge-
neous layered soil lenses) that negatively impact the effective distribution
and transport of reactants throughout the contaminated site. Ex-situ,
aboveground reactor technology is less sensitive to these restrictions be-
cause its implementation is unaffected by site conditions once contaminated
groundwater is extracted from the aquifer. All technologies are impacted by
aquifer conditions that affect microbial activity (i.e., soil pH and nutrient
limitations), although the ex-situ system can be designed to modify these
properties in a controlled reactor environment. All technologies are affected
by waste constituent characteristics that affect a contaminant's toxicity,
biodegradability, and bioavailability. Toxicity and biodegradability limita-
tions can be controlled in an aboveground reactor through modification of
extracted groundwater conditions via dilution, addition of carbon sources to
Stimulate co-metabolic degradation, etc.  Both biosparging and aboveground
reactors rely on contaminant mobility for aquifer treatment so a
contaminant's solubility has a significant impact on the effectiveness of
these methods.
  The following sections describe each of the groundwater treatment tech-
niques listed in Table 5.1, and discuss technology applications and limita-
tions appropriate for groundwater remediation systems applied at the
field-scale.

                               5.1
-------
 Groundwater Treatment Systems
                              Table 5.1
    Impact of Various Site, Soil, and Waste Constituent  Characteristics
         on Groundwater Treatment Technology Performance*

Site




SoU




Waste Constituent





Climatic conditions
GW table fluctuations
Surface structures
11
Layered formation
Product existence/distribution
Fine grained
High-water content
Low-water content
Nutrients
^ ;
VolatUity
Biodegradability
Bioavailabiltiy
Water solubility
Toxicity
RP
N
I
N-
V-
V
V-
N
N
I- l
I. ' '" '
N
V
V
V
V-
IR
N
I
I
V
V
I
N
N
I-
I-
I
V
V
V
V-
AGR
N
I
N
V
v" •"
I-
N
N
N
N'
I
V
V
V
V-
BS
N
I
I
V-
V
V-
N
N
I
' I 	
V
V
v 	
I
V-
;MB '
N
I
I
V-
V
V-
' ' N '
N
I
' t 	
N
I
v
V
"v:1
 *RP = Raymond Process
 IR = Intrinsic Remediation
 AGR = Aboveground Reactors
 BS = Biosparging
< MB = Migration Barriers
 GW = Groundwater
 N = Not Important related to the performance of the technology
 I = Important related to the performance of the technology
 V = Very important related to the performance of the technology
 + ^ Characteristic positively impacts the performance or selection of the technology
 - = Characteristic negatively impacts the performance or selection of the technology
5.2  Raymond Process (Enhanced
Bioremediation of Aquifers)
5.2.1 Principles of Operation

   Although land farming and wastewater treatment have been commercial-
ized for many years, the first process to become widely known as a bioreme-
diation process was in situ treatment of aquifers. This process was first field

                                5.2
-------
                                                            Chapters
tested in 1972 by Richard L. Raymond of Sun Tech in conjunction with the
American Petroleum Institute (Raymond, Jamison, and Hudson 1976;
Raymond et al. 1978; Brown, Norris, and Raymond 1984). The process
became commercial in the mid-1980s and consists of:
        •  extraction of groundwater;
        •  aboveground treatment of the extracted water;
        •  amendment of the water with an electron acceptor and, typically,
           nutrients; and
        •  reinjection of the water so that it will sweep through the contami-
           nated soils and groundwater towards the groundwater recovery
           system as shown in Figure 5.1.
                              Figure 5.1
                    Well System for Liquid (Delivery
Mixing Tank:
Nutrients
Surface IpHftHHH
Injection
Well
Water Table V

•

Air Pump
— ,-nfl
1
r
i-
V
"



t

Produced w
T. recirculated
P""1!' surface usir
j— J~ =j options.
Producing
Well
V
ater can be
or treated at the
g other remedial


    Groundwater
    Flow	^"
Sparger Device
               Nutrient anoli«ygen-Rich
               Groundwater
Source: Ward et al. 1995
   The process relies on the injection of electron acceptors and nutrients to
stimulate microbial growth and enhance the rate of biodegradation of the
organic contaminants.  Depending upon the solubility of the contaminants,
                                 5.3
-------
Groundwater Treatment Systems
the hydrogeology, and the system design, varying proportions of contami-
nant mass reduction will occur as a result of removal of contaminants in the
recovered water.
            111                    '         i ''"•'"
   The first systems used oxygen that was supplied by sparging air within
the injection well(s) as the electron acceptor (Raymond 1974). Subse-
quently, other sources of electron acceptors including hydrogen peroxide
(Raymond et al. 1986), nitrate (Hutchins et al. 1991), and sulfate (Beeman et
al. 1993) were used as electron acceptors. The use of alternate electron ac-
ceptors for petroleum hydrocarbons (Reinhard 1994) and chlorinated sol-
vents (Bouwer 1994) are discussed in the Handbook of Bioremediation
(Norris et al. 1994).
   In its current state of development, the Raymond Process is applicable to
remediation of petroleum-based hydrocarbons (including commercial fuel
blends), creosote, oxygenated solvents such as alcohols and ketones, some
chlorinated compounds, and (to some extent) PAHs.  It is also being field
tested with some success for treatment of chlorinated solvents such as
trichloroethylene and trichloroethane. Commercial treatment of pesticides,
herbicides, PCBs, high molecular weight hydrocarbons, and most energetics
is not likely  to be available for some time.
   Generally, the process is most suited to relatively permeable soils and is
most easily applied to homogeneous aquifers. The aquifer hydraulic con-
ductivity and saturated interval thickness determine the rate at which
groundwater can be transported to deliver electron acceptors and nutrients.
The minimum acceptable hydraulic conductivity depends upon the mass
loading of the contaminants and the oxygen demand that must be met by the
added electron acceptor.  Sample calculations are provided in Section 5.2.9
of this monograph.
   As discussed in Section 5.3, it is now recognized that many sites, espe-
cially those where the contamination is limited to petroleum hydrocarbons
and/or other readily biodegradable compounds, do not represent an immedi-
ate risk to human health or the environment because natural attenuation  pro-
cesses will prevent migration to downgradierit receptors. As a result, many
regulatory agencies are accepting or even requiring a combination  of risk
evaluation and intrinsic remediation to be considered at UST sites. ThuSj
many sites that have been remediated using the Raymond Process would,
today, be addressed by intrinsic remediation or a combination of the two
technologies.
  A comprehensive introduction to the Raymond Process can be found  in
the preceding monograph, Innovative Site Remediation Technology:
                                 5.4
-------
                                                             Chapter 5
Bioremediation (Ward et al. 1995), in Chapter 2 of the Handbook ofBiore-
mediation (Norris et al. 1994), in A Guide for Railroad Industry Use of In
Situ Bioremediation (Brubaker et al. 1994), and in In Situ Bioremediation:
When Does It Work? (Rittman et al.  1993).

5.2.2  Process Design Principles
   As a result of the release of hydrocarbons to the subsurface, contamina-
tion will impact unsaturated soils. If sufficient quantity is released relative
to the retention capacity of the unsaturated zone, free-phase hydrocarbons on
the groundwater surface (light nonaqueous-phase liquids or LNAPLs), con-
taminated saturated zone soil, and contaminated groundwater can result.
Even excluding NAPL, normally only a small fraction of the contaminant
mass within the saturated zone will be in the dissolved phase. Therefore,
this process must be designed to address the mass of contaminants that are
adsorbed or entrained by the soil because the dissolved contaminant mass
generally represents less than 1% of the total mass of contaminant distrib-
uted throughout the impacted area.
   The remediation system must supply the necessary amount of electron
acceptor to degrade the total mass of contaminant within the contaminated
saturated zone and capillary fringe minus the mass of contaminant that is
removed with the extracted groundwater. For remedial designs that inplude
bioventing, the mass of contaminants removed or biodegraded through
bioventing as a result of water table fluctuations should be estimated. When
oxygen is used as the electron acceptor, between 1 and 3 kg of oxygen must
be supplied for each kilogram of hydrocarbon not removed in the recovered
water or through bioventing. In essence, the problem consists of engineering
a system that can cost-effectively meet this requirement within an acceptable
time frame and at a lower cost than other acceptable alternatives.
   First, it is necessary to know that all constituents of interest are biodegrad-
able, will be extracted, or will be addressed by an integrated design. Given that
these conditions are met, the aquifer must be evaluated to determine unaccept-
able pH conditions, presence of inhibitory compounds, availability of electron
acceptors and nutrients, contaminant distribution, and hydrogeology. The first
two criteria can be functionally evaluated by standard plate count techniques
mat will identify conditions potentially deleterious to microbial activity^  If
electron acceptors and nutrients were already present at sufficient levels, there
would be no need to design a system. Nutrients, especially phosphorus, may be
sufficient to support biodegradation of the contaminant mass. However, elec-
tron acceptors are usually insufficient and their rate of introduction to the aqui-
fer is normally rate-limiting for biodegradation during active remediation.

                                  5.5
-------
 Groundwater Treatment Systems
   The critical feature of the design is the distribution of the electron
 acceptors). For design purposes, background levels of electron acceptors
 can be left out of the calculations.  If background levels of electron acceptors
 are significant compared to the contaminant mass, let alone the uncertainty
 in the contaminant mass, then intrinsic remediation will, in all but a few
 cases, be the remedy of choice.  Designing a system to meet the oxygen
 demand requires an understanding of the three-dimensional distribution of
 the contaminant mass and the details of the site hydrogeology. Mass distri-
 bution is determined from soil borings.  Site hydrogeology is determined
 first from well logs, but must eventually include slug tests or preferably 24-
 to 72-hour pump tests and some level of modeling (Falatico and Norris
 1990; Bedient and Rifai 1993). Two dimensional analytical flow models are
 frequently sufficient, require modest data input, and can be used to simulate
 numerous injection-recovery scenarios, thereby permitting selection of the
 most favorable combination of injection and recovery wells.  Numerical
 models can account for heterogeneity and provide a more accurate represen-
 tation of groundwater flow but require more input and are much more costly
 to develop and use.
   Modeling is used to select the number and location of wells using patterns
 such as those shown in Figure 5.2, and to estimate groundwater injection and
 recovery rates. The groundwater injection rates are used along with the
 achievable concentrations of electron acceptors in the injection water to
 estimate the delivery rates of the electron acceptor and thus, the time that the
 system will have to operate to meet the remediation goals. The rate of physi-
 cal removal also needs to be factored into the calculations. These calcula-
 tions are discussed in more detail in Section 5'.2.9.
   Injection and recovery rates must be balanced so that injected nutrients
 and electron acceptors flow through the contaminated zone and can be cap-
 tured by the recovery wells if not used by the microorganisms. For efficient
 capture, the injection rate typically needs to be less than the recovery rate by
 about 20% to 30% to satisfy regulatory agency concerns regarding residual
 levels of nutrients and to minimize the probability of spreading the contami-
 nants. From a practical point of view, small excursions of contaminants are
 addressed through natural attenuation. To avoid recovering and treating
 excessive quantities of clean water, it may be necessary to recover less water
 than the wells could yield if pumped at full capacity.  As shown in Figure
 5.3, pumping at maximum recovery rates can extract water from outside the
 contaminated area, excessively diluting the feed to the aboveground treat-
ment system, resulting in the need for a larger system than necessary to treat
the contaminated aquifer.
                                  5.6
-------
                                                               Chapter 5
                                Figure 5.2
                   Raymond Process Well Configurations
                              -Groundwater Flow-
                                o
                                o
                                o
                                                                o
            a. Straight Line Pattern
    In-Line Series Pattern
          c. Isource Area Capture Pattern
d. Repeating Five-Spot Pattern
 • Injection Wells
 O Recovery Wells
   One common approach is to locate injection wells upgradient of the
source area and recovery wells on the downgraclient side as shown in
Figure 5.2a and 5.2b (Brabaker et al. 1994).  This takes advantage of the
natural gradient and also serves to pull back some portion of the plume.
Another approach is to place extraction wells within the center of the plume
and inject through a ring of wells located near the perimeter of the source
area as shown in Figure 5.2c. This design maximizes the recovery of .
dissolved-phase contaminants as well as residual LNAPL and creates a flow
from the clean area to the impacted area. For large sites or for sites with
relatively low hydraulic conductivity, a repeating five-spot pattern
(Figure 5.2d) creates flow within the aquifer thait reduces the travel time of
                                   5.7
-------
Groundwater Treatment Systems
the injected water as compared to the previously discussed patterns. The last
two patterns also lend themselves to subsequently reversing or otherwise
modifying the flow directions.  Reversing the flow direction reduces the
average distance mat electron acceptors and nutrients must travel and can
eliminate dead spots (flow stagnation) that may occur (especially for the
five-spot pattern).
                                '              '         '              '
                               Figure 5.3
Inefficient Groundwater Capture Resulting from an Excessive Pumping Rate
• Recovery Walls
o injection Wells
   ;:             '	  ,,,„"„ / ;  , ' ,  , , I,,,:,, ,   „ f :| ' '	   "',  • . i ,|. , •  i    „ 	H j,,,, i|« t
   The interval over which the injection and recovery wells are screened is
important.  At a minimum, wells need to be screened across the smear zone
and must allow for normal water table fluctuations.  Screening over a larger
(deeper) interval will allow higher recovery rates and may allow higher in-
jection rates. The impact of the  screened dimension will depend upon the
details of the soil particle-size distribution in"the vicinity of the wells. In-
creasing the groundwater recovery and injection rates will not be of benefit
if the increased flow occurs primarily through deep, high-permeability chan-
nels where only a small fraction of the contaminant mass exists.  Design of
                                  5.8
-------
                                                             Chapter 5
injection/recovery systems must incorporate a three-dimensional picture of
both the hydraulic properties of the aquifer and the distribution of the
contaminants).
   After an injection/recovery well pattern has been selected, it is necessary
to estimate whether the configuration can deliver sufficient election accep-
tors in a reasonable time to meet the demand created by the mass of con-
taminants.  The mass of contaminants can be estimated from the volume of
impacted soil and the average concentration, or more sophisticated estimat-
ing techniques can be used. The mass of oxygen required is adequately
estimated as three times the contaminant mass associated with the soils. The
time required to deliver the required oxygen is estimated by dividing the
required oxygen mass by the oxygen flux, calculated from groundwater
injection rate, the water density, and the oxygen (or other electron acceptor)
concentration in the injected water.
   Table 5.2 gives some examples of the time required to introduce various
electron acceptor sources and blends at a series of injection rates to degrade
454 kg (1,000 Ib) of contaminant mass.  The estimated mass of contaminant
to be biodegraded should be adjusted for the mass of the contaminant that
would be extracted by the recovery system. For gasoline and jet fuels, this
mass could be 20% to 40% of the total, depending upon the initial condi-
tions and the proximity of the recovery wells to the highest soil contamina-
tion levels. For heavier molecular weight fuels, the mass recovered in the
aqueous phase will be insignificant compared to the uncertainty in the esti-
mations of the adsorbed-phase mass and the amount removed by groundwa-
ter recovery can be ignored for these calculations. If the estimated time to
supply the electron acceptor is too long to meet project objectives, then the
well pattern is not satisfactory. In some cases, the hydraulic conductivity of
the aquifer will be insufficient to permit satisfactory rates of electron accep-
tor introduction without incorporating an excessive number of wells.
   The above-described simple approach ignores heterogeneity in groundwa-
ter flow and contaminant distribution. Depending upon the type of well
pattern used and the degree of heterogeneity, this approach may be satisfac-
tory. Frequently, it is prudent, if not always practiced, to es timate treatment
times for individual areas within the contaminated zone. This added level of
complexity may lead to the inclusion of several additional wells located to
more aggressively address areas of high contamination or low hydraulic
conductivity.
   An excellent approach is to include, as part of project design and budget-
ing, the installation of additional injection and/or recovery wells after the
initial system has been operated long enough to learn how the aquifer is

                                  5.9
-------
Groundwater Treatment Systems
actually going to respond. This observational approach provides the best
opportunity to respond to the presence of permeable channels that have
low-adsorbtivity capability and to low-permeability intervals that have
greater adsorptive capacity and thus may contain the majority of the con-
taminant mass. Where this dichotomy is extreme, it may be best to use an
active system to address the more transmissive soils and allow intrinsic
remediation to address contaminants that will slowly diffuse from the less
transmissive soils to the more transmissive soils.  In many cases, natural
attenuation will be sufficient to prevent significant migration of contaminant
mass toward dowiigradient receptors. A discussion of intrinsic remediation
for groundwater management is provided in Section 5.3.
                               Table 5.2
       Treatment time for Each 454 kg (1,000 Ib) of Hydrocarbons*
                     Electron Acceptor           Treatment Times (Days) at
                      Concentration             Indicated Injection Rates
Source                   (mg/L)         4.5 m3&      45m3&
In Well Aeration
In Well Pure Oxygen
NO 3+ Pure Oxygen

ftn\J^
ir " 	 1,1
"22

8
"40 ' 	
80

200
500 	 '"

55
11
5.5

3.5
1.4"

55
1.1
0.55

0.35
0.14

0.6
0.1
0.06

0.04
0.01
'1
1 ,: " " : , ":, , ' , " ; !, i!:'11;,
•Assumes 100% utilization and stolchiometric requirements.
   One other important consideration is site-specific geochemistry (Norris
1992; Ward et'al. 1995). The addition of oxygen sources to an aquifer can
cause precipitation of reduced kon that is frequently present in contaminated
groundwater or, worse, precipitation of iron from reinjected groundwater
leading to plugging of wells or reduced permeability in the immediate vicin-
ity of the injection well. Groundwater chemistry is also a critical issue with
respect to nutrient selection. Calcium^ iron, and magnesium will precipitate
in the presence of ortho phosphate (salts of phosphoric acid). Therefore, it is
preferable to use potassium tripolyphosphate (KTPP) because it can

                                 5.10
-------
                                                             Chapters
sequester these metals and prevent precipitation in the vicinity of the well
bore. Tripolyphosphate must be present in molar excess of the total moles of
calcium, iron, and magnesium. The appropriate calculation for the minimum
KTPP concentration is presented in Section 5.2.9. The use of potassium
salts rather than sodium salts minimizes the possibility of swelling of clays.
If the chemistry of the treated water is not appropriate for nutrient amend-
ment and reinjection, it will be necessary to dispose all treated water and use
water from another source, such as city water or water from a deeper aquifer
for reinjection.
   For planning purposes, the total amount of nutrient required can be esti-
mated from a C:N:P ratio of 200:10:1 that is based on the ratio of those
elements in cell material with the assumption that 50% of the contaminant is
converted to cell mass. This is an overestimate because the soils and j
groundwater nearly  always contain most of the required phosphorus and
some nitrogen in usable forms and these elements are recycled when the
cells die. On the other hand, transport of nutrients is retarded by the soil. In
practice, nutrients are first added at a rate that would provide nutrients over
the first 25% to 50% of the estimated treatment time. Nutrient addition is
adjusted or terminated based on detection at downgradient monitoring loca-
tions; after nutrients reach the extraction wells, nutrient addition should be
terminated so that nitrogen and phosphorus levels will not exceed regulatory
limits at the end of the remediation process.
   Extraction wells are designed as previously described. Installation tech-
niques are the same  as for any groundwater recovery system. It is important
to pay close attention to well completion and development techniques to
minimize loss in recovery rates from redeposition of fines and to maintain
the design flow rates for long periods. Groundwater recovery can be accom-
plished with many different pump types. Pump selection and design de-
pends upon the total flow, depth to water, and potential for silting. A good
description of groundwater recovery pumps can be found in Groundwater
and Wells (Driscoll 1986).
   Groundwater treatment systems are designed based on the composition of
the groundwater, requirements for offgas treatment or polishing of effluent, flow
rate, regulatory compliance requirements, and level of operator attention.
Table 5.3 lists types of groundwater treatment systems, the classes of com-
pounds that can be treated, and the general level of operator attention required.
   Monitoring well requirements and treatment of the extracted groundwater
are discussed in Section 5.2.6. Treatment of extracted groundwater must
take into account the chemistry of the  water provided for reinjection.  While
regulatory requirements focus on removal of contaminants, the inorganic

                                 5.11
-------
                                              	1"
                                                                         nil in
Groundwater Treatment Systems
                                   Table 5.3
                 Common Groundwater Treatment Systems
  Treatment
   System
    Contaminants Treated
                                         Comments
Air Strippers
Activated
Carbon
Bioreactors
UV Oxidation
Membrane
Separation
VOCs-BETX, gasoline, etc.
TCE, PCE, MTBE


Most organics; inefficient for
highly water-soluble compounds
Most organics; some chlorinated
organics are less biodegradable
Most VOCs and organic ketones;
alcohols are not readily oxidized
Most organics and inorganics
 •  Considerable maintenance
 • May need off-gas treatment
 • Iron removal may be necessary

 • Simple to operate
 • Costly O&M
 • Nondestructive technology
 • Filtration pre-treatmerit may be required

 • Suspended growth for high concentrations (e.g.,
  >100 mg/L as COD)
 « Fixed growth for low concentrations (e.g.,
  <106 mg/L COD)
 » Nitrogen and phosphorus addition may be
   required
 ». Sludge handling and disposal required

"•' Compound-specific treatment
 •  May require catalyst (e.g., Fenton's Reagent)
 '• Destruction to CO2 +H?O possible for selected
  organics
 • Pre-treatment for iron required
 •  Can create oxidation byproducts such as acetone
      ; i| Ji"""" ' "!{•• ' '"i ' '	'	"!"  v 	:;, 1 Si  	!','-
 • High capital cost
 • High O'&M cost
 •  Nonselective treatment
 • High level of pre-treatment required
chemistry (calcium, iron, magnesium, carbons, etc.) will determine whether
pretreatment is needed to prevent operational problems during remediation.
Pretreatment for metals can add significantly to groundwater treatment costs.


5.2.3  Process-Flow Diagrams
                  ' I '  • ' . .  '   !'	 •	 '	! .'«    '"   l|l '  I , .  • •     ',,      ,      ii	
           ,!,      , ' I    ,„  , '   	„  ' '   !   '!!„ ,       f  ,!    „     " "i ',  	 , , „   '  ,, i1	",'!	
   Critical features that are frequently incorporated into the design of a
Raymond Process system are shown in Figure 5.4.  These include compo-
nents for pretreatment for inorganics and metals, if necessary; treatment;  a

surge tank with controls unless none of the recovered groundwater is rein-
jected; addition of nutrients; addition of the electron acceptor; transfer lines;
and filtration before injection.  Process and instrumentation (P&I) drawings
should be prepared for each major component including transfer lines as

discussed in Section 5.2.8.
                                                                       "''I"< I SP1' "-"it:''!'
                                      5.12
-------
                                                               Figure 5.4

                                       Groundwater Treatment, Amendment, and Recirculation
                                   H,
en

CO
Nutrients
                                                         High-Level Control
                                                         Low-Level Control
                                                                                                             Discharge
     ©Injection Wells

     O Extraction Wells
                                                                                           §
                                                                                           Q
                                                                                           ig.

                                                                                           CD

                                                                                           Oi
-------
         Groundwater Treatment Systems
         5.2.4 Process Modifications
            The Raymond Process, as first practiced, provided aqueous solutions of
         nutrients in batch additions to the injection wells. Oxygen was supplied by
         the use of air diffusers (air spargers) located at the bottom of the injection
         wells. Nutrients were dissolved in water on-site and added in batches. The
         simplest process modification was  the inclusion of a storage tank and meter-
         ing pump for continuous addition of nutrients. Three other modifications
         have been practiced: (1) alternate electron acceptor sources, (2) injection and
         recovery systems other than wells, and (3) air sparging directly into the aqui-
         fer as discussed in Section 5.5.  The other oxygen and alternate electron
         acceptor systems include those discussed in the following paragraphs.
            Pure Oxygen Systems.  The use of pure oxygen in place of air offers the
         possibility of introducing the electron acceptor at a five-fold increase in rate
         and, presumably, a similar reduction in the treatment time. Oxygen is generated
         on-site or brought on-site hi liquid (cryogenic) jfbrm (Prosen, Korreck, and
         Armstrong 1992).  Other aspects of operation are virtually the same.
         Oxygen-generating units,  such as those used hi remote hospitals, can provide
         sufficient oxygen for modest systems at reasonable capital investments. If liq-
         uid oxygen is used, an evaporator and additional equipment are required and
         thus this approach is more applicable to larger systems. The injection system
         can be much the same as for air, using spargers or diffusers in the injection well.
         Oxygen utilization is more efficient if the injection water is saturated with oxy-
         gen aboveground  in one of several types of systems. These systems provide
         high turbulence to allow dissolution with a short contact time. After the oxygen
         is dissolved in the injection water, the linear velocity of the water in the delivery
         system must be maintained above certain levels to minimize dissolution of the
         oxygen. The oxygenated water is then introduced mrough an annular pipe to
         the bottom of the injection well.  Fully-pressurized systems where the well bore
         is maintained full of water reduces the hazard of high oxygen levels hi the pres-
l-.'-j!:1   •   ',     '>'"	 "i .'.,•''' •'   I"   •	 ,!'  ',  .i":""1 ,  , "-[' i. .I?,!", •':  ('!;|['"!	1'II.H-  '.  "'i.  '!'"   ..'    '. 'I"'	I	•
         ence of flammable vapors.
            Hydrogen Peroxide Systems. Hydrogen peroxide, H^,was ^asi used in
         the mid-1980s because it is miscible in water arid decomposes in the aquifer
         to yield oxygen and water (Brown and Norris 1988). Two kilograms of pure
         hydrogen peroxide produce almost 1 kg of oxygen. Hydrogen peroxide
         systems use metering pumps to transfer hydrogen peroxide (35 to 70%) from
         storage tanks for continuous addition to the injection water. Hydrogen per-
         oxide at higher concentrations can serve as a tjactericide and at too high a
         concentration, will yield oxygen too rapidly, leading to loss of oxygen to the
         unsaturated zone or gas blockage of the aquifer. Where applicable, hydro-
         gen peroxide can be introduced at concentrations from 100 mg/L to

                                           5.14

 ,"t( •   	'  :    '  ' '      ,     .!  I,,1 .  ' .      	   i '  " '!    , j"" i: ''•        "          i I "li'ii, '.
;.,.
-------
                                                            Chapter 5
1,000 mg/L (typically, 200 mg/L to 500 mg/L). At 500 mg/L, oxygen is
theoretically provided at a rate that is 25 times greater than that achieved by
the sparging of air and more than 10 times faster than that achieved with
pure oxygen. However, hydrogen peroxide stability is frequently a problem
and must be evaluated before completing the system design (Flathman et al.
1991; Lawes 1991). If hydrogen peroxide stability and the groundwater
flow rate under operating conditions are insufficient to permit hydrogen
peroxide and/or elevated levels of oxygen to penetrate several meters into the
aquifer, alternative designs will have to be implemented. Shorter
remediation times should result if hydrogen peroxide utilization is efficient.
   Nitrate Systems. Nitrate systems use metering pumps to transfer nitrate
from tanks containing concentrates for continuous addition to the injection
water (Hutchins et al. 1991). Nitrate can serve as  a nutrient source and as an
electron acceptor for aromatic compounds (except, reportedly, benzene) and
some chlorinated solvents, but not for aliphatic hydrocarbons. Nitrate is
highly soluble in water and is not retarded by soils. Thus, nitrate is techni-
cally an ideal electron acceptor for some organics. However, its use is regu-
lated, and many regulatory agencies limit the nitrate concentrations in the
injection water to 10 mg/L as N (46 mg/L as NO3")
   Sulfate Systems. Sulfate systems use metering pumps to transfer sulfate
from tanks containing sodium or potassium sulfate concentrates for continu-
ous addition to the injection water (Beeman et al.  1993). Sulfate has been
used at a few sites for the reductive dechlorination of tetrachloroethylene
(PCE)  and trichloroethylene (TCE) under anaerobic conditions. Reductive
dechlorination requires a degradable organic substrate/electron donor, such
as sodium benzoate. Sulfate is highly water soluble and is not retarded by
soils, so it is easily distributed through the aquifer. Apparently, there are no
regulatory restrictions that would limit the amount of sulfate added; how-
ever, sulfide is an anaerobic byproduct that can be of concern, and sulfate
concentrations greater than 250 ppm produces a laxative effect, and should
be avoided.
   Multiple Electron Acceptor Systems. Some designs have successfully
incorporated more than one electron acceptor. Both oxygen and nitrate were
successfully used at the French Limited  site in Crosby, Texas (Thomsom et
al. 1995).  Nitrate penetrated more rapidly than oxygen  and appeared to
serve as an electron acceptor for the degradation of toluene and several chlo-
rinated ethenes. Oxygen can be introduced downgradient of where sulfate is
introduced to promote reductive dechlorination. Oxygen is then available to
enhance biodegradation of vinyl chloride, which may be formed by reduc-
tive dechlorination.
                                 5.15
-------
                                Groundwater Treatment Systems

HI  V     „	!	•
   Other Injection/Recovery Systems. Figure 5.5 shows alternatives to in-
jection/recovery well systems (Cookson 1995).  Systems that add amended
water at the surface, beneath the surface but above groundwater (percolation
systems), or beneath the groundwater surface (trench systems) have been
used in various combinations, including combinations with recovery wells.
   Percolation Systems. Introduction of amended water from the surface
using percolation systems are limited to sites; where the unsaturated zone
consists of soils with adequate percolation rates. Sands and gravel are most
appropriate. As me fines content of the soil Increases, percolation rates de-
crease, adsorption of nutrients increases, andt potential losses of electron
acceptors, especially hydrogen peroxide, increase. A detailed discussion of
these processes can be found in Bioremediation Engineering: Design and
Applications (Cookson 1995).
   The system shown in Figure 5.5a presents a number of permutations of
the addition of amended water to unpaved surfaces using sprinklers, irriga-
  i.": i1    	    , i" .i11: • i ',!!'  i „! '   ' i.'i,:i	; ' ""iii *	.iMv,' „.	
-------
                                                      Chapters
                           Figure 5.5
    Alternative Injection/Recovery Well Systems for the Raymond Process
 Bed Rock |_
                    a. Spray Irrigation for Gravity-Feed System
      \\\
  Original
 Water Table
              Nutrients
                                            ///
                                       Induced Water
                                         Level
                          fs-*\\' /*.- •V^/;^>»>rs;Contamiination,/f^^li.J-"^ '  • • •>
.'. . . jX^. A\V\\VVN'^ xV^--»^.-A\V^^- V •  • • •  .
 Bed Rock
                    b. Infiltration Trenches for Gravity-Feed System
Source: Cookson 1995
                             5.17
-------
Groundwater Treatment Systems
                           • ,   • . I   i"!1:"  •.   |  f .   i   .  :•.   '   •   •  '  -
area as soon as amended water percolates to the water table in subsurface
percolation systems.
   Percolation systems generally incorporate groundwater recovery systems
that capture the introduced amendments and contaminated groundwater to
satisfy regulatory requirements or to protect downgradient receptors. First, it
is necessary to calculate, determine, or estimate percolation rates and
mounding effects (Cookson 1995) arid, in some cases, account for retarda-
tion effects. For highly-porous, unsaturated soil with a shallow water table,
percolation rates will probably exceed requirements. Where the aquifer is
highly transmissive, mounding may be minimal. For less permeable soil and
deeper water tables, percolation rates and mounding effects may limit this
application method. The groundwater recovery system must then be de-
signed, using modeling techniques discussed in Section 5.2.1, to capture
contaminated and amended water.  As with well systems, the rate of ground-
water recovery must exceed the rate of introduction, and aboveground treat-
ment will be required.
   Systems using trenches extending into the aquifer operate much like well
systems except that radius of influence limits no longer apply  as long as the
trench(es) traverse the entire treatment zone. Trenches also provide more
surface area and are less prone to plugging than wells.

5.2.5  Pretreatment Processes
   Recovery and reinjection wells must be developed or redeveloped prior to
implementation to ensure maximum performance of the wells. Well devel-
opment procedures differ depending upon  the type of well and the character-
istics of the formation.  Procedures are described in Groundwater and Wells
(Driscoll 1986).
   Pretreatment of water to remove iron and calcium may be necessary to
prevent fouling of the water treatment system. Removal of heavy metals
may also be necessary for regulatory purposes. Typically, inorganics and
heavy metals are removed through precipitation, although reverse osmosis or
ion exchange may be used as Discussed in  Section 5.4.6.  Depending upon
the treatment method used for recovered groundwater, adequate metal re-
moval may readily be achieved as discussed in Section 5.2.6.
   Nutrient blends can be prepared on-site, but it is generally more economi-
cal to have the blends prepared off-site by  a chemical distributor. The dis-
tributor should be required to filter the nutrient concentrate before shipment.
The injection water should also be filtered downstream of where the nutri-
ents and the electron acceptor are introduced and ahead of the injection

                       '    '  ' "  *	  ",      I ' i •:   '   •• • ,           • i  >	
                                 5.18
-------
                                                            Chapter 5
 wells. If precipitates are introduced into the well, the well, gravel pack, and/
 or formation may become plugged, reducing; well performance and increas-
 ing treatment times and cost.

 5.2.6 Posttreatment Processes

   Where the concentration of degradable compounds in the recovered water
 exceeds several mg/L, it is more cost-effective to treat the water
 aboveground because reinjection of untreated water will increase the de-
 mand upon the injection system to introduce electron acceptors, resulting in
 longer treatment time.  Reinjection without treatment is not usually an op-
 tion because contaminated groundwater typically must be treated before
 reinjection to meet regulatory requirements. One exception is where water
 can be discharged to a sewer or existing wastewater treatment plant and
 another source of water is used to introduce nutrients and electron acceptors.
   Monitoring well locations and screened intervals need to be determined as
 part of the design. To modify operating parameters based on evaluation of
 performance as discussed in Section 5.2.18.2, the wells must be located to
 provide performance data in a timely fashion.  Typically, this means locating
 monitoring wells or piezometers based on the travel time of the injected
 water. The monitoring points located closest to the injection wells should
 intercept the injected water at a point corresponding to one to two weeks
 groundwater travel time from the injection wells. Additional monitoring
 pouits need to be spaced at points corresponding to, for example, 25,50, and
 75% of the groundwater travel time between injection and recovery points.
   Treatment of the recovered water is the same as for pump-and-treat sys-
 tems. Air stripping towers or low-profile strippers, activated carbon, and
 bioreactors, as described in Section 5.4, are the most common treatment
 systems.  Air-stripper towers may require an offgas treatment system.  Air
 strippers are typically sized using computer software incorporating input
 parameters that include flow rate, temperature, volatile organic compound
 (VOC) concentration, and the Henry's Law constants for all constituents.
 Activated carbon system requirements can be estimated by suppliers based
 on known flow rates and groundwater quality data. The major problems hi
 sizing either air strippers or activated carbon systems are obtaining good
 estimates of groundwater flow and the concentration of specific compounds
 in the recovered water.  Usually systems are overdesigned to ensure compli-
 ance.  Groundwater with mixtures of VOCs, semivolatiles, chlorinated sol-
 vents, and acetones or ketones may require two methods of treatment.
Table 5.3 summarizes the commonly used groundwater treatment processes.
                                5.19
-------
Groundwater Treatment Systems
   In addition to treatment of organics, it may be necessary to remove iron and/
or heavy metals. Iron treatment is usually accomplished by oxidation and pre-
cipitation and includes filtering, drying, and sludge disposal.  If heavy metals
are present or if organics are removed with the iron, the sludge may need to be
disposed as hazardous waste. Treatment of iron and/or heavy metals can easily
exceed the treatment costs for the organic constituents.
   For bioremediation at gasoline service stations and similar UST sites, the
water treatment system must be able to operate for several days or weeks at a
time without maintenance.  At larger sites, such as many Superfund sites,
daily or even around-the-clock operators niay be required for other purposes,
and thus, more labor-intensive treatment systems may be tolerable.
                           . •  •   '	!	''	|- -    • 	 - •••  '  •	•    - • |  -'	
   The treated groundwater may be discharged to sewers, surface water, and/
or rejnjected. Unless another source of water is being used for reinjection,
the treated water stream.is normally split between some form of discharge
and nutrient/electron acceptor amendment and reinjection. Before reinjec-
tion, it is necessary to filter the water so that particulate matter is not being
introduced to the well; even small amounts of solids can eventually reduce
injection rates.
     • ',  i       '     ,    ••  •  :•::, ,' •   "  ')	'.:,,!    I  ,    " •'  •  •"     ,  '	: r i (*;•  ','•.
5.2.7  Process Instrumentation and Control
   The main components of a Raymond Process system are listed in Table
5.4.  In addition to the components common to a pump-and-reat system,
Raymond Process systems include nutrient and electron acceptor delivery
systems, transfer pumps and lines, injection wells, controls, and meters. The
system components are designed to remove and treat water from the aquifer,
amend the water,  and reintroduce the water to the aquifer in a manner that
does not lead to reduction in permeability. All of these processes need to be
integrated so that water flow is balanced.
   As shown in Figure 5.4, the effluent from the groundwater treatment sys-
tem is split between discharge and recirculation. The recirculation water is
typically sent to a surge tank which should hold several hours of flow.  The
surge tank is necessary because the rate at which injection wells will accept
water fluctuates arid typically decreases over time until the wells are treated
or redeveloped. Transfer pumps are used to deliver water to the injection
header. Injection wells are either gravity fed or pressure fed. Controllers are
used to operate valves and/or the transfer pump»s.  Water level in the surge
tank is maintained between specified limits with high/low switches. The
system can be designed to automatically direct more of the treated ground-
water to the sewer or other discharge point, rather than interrupt groundwater
                                 5,20
                                         	j
-------
                                                              Chapters
                                Table 5.4
              Major Raymond Process System Components


             Recovery Wells              Groundwater Treatment System
             Injection Wells              Surge Tank
             ManifoldsATransfer Lines        Nutrient System
             Force Mains                Electron Acceptor System
             Transfer Pumps              Controllers
             Valves/Meters               Programmable Logic Controller
recovery, thus maintaining groundwater capture patterns. The system de-
sign, in most cases, should call for operation at less than the maximum
achievable injection rates so that the system can be adjusted to compensate
for moderate losses in injection capacity over time.
   The nutrient and electron acceptor feeds are located between the surge
tank and the header(s) to the injection wells. The nutrient tanks should be
sufficient to contain several weeks' supply of nutrients. Level indicators on
the nutrient tanks document the amount of nutrient added. Totalizer readings
and delivery records are used, as are periodic analyses of both the nutrient
solution in the tank and samples collected at points sufficiently downstream
of the nutrient feed, to document nutrient delivery to the aquifer.
   The electron acceptor delivery system depends upon the total system
design. Nitrate and sulfate addition is essentially the same as that for nutri-
ents. Hydrogen peroxide delivery systems are similar except the properties
of hydrogen peroxide place special demands on the materials of construction
and require provisions for pressure release.  Small remediation systems may
use drums or other containers, while large systems use storage tanks up to
7,570L (2,000 gal). Transfer lines  must not isolate hydrogen peroxide with-
out pressure release, otherwise the  slow evolution of oxygen gas as the hy-
drogen peroxide decomposes will rupture the line.  It is imperative that a
hydrogen peroxide supplier or a qualified distributor provide the delivery
system and, for larger systems, install the system.
   Gas delivery systems consist of  an oil-free air compressor (to avoid intro-
ducing contaminants) and filter, along with pressure release valves, pressure
gauges, and control valves. Molecular sieves can be used to remove nitrogen
from air so that a higher oxygen content can be introduced into the aquifer.
A more complete description of such systems is provided in Section 5.5.5.
                                  5.21
-------
 Groundwater Treatment Systems
   If city water is used as a water source, it may be necessary to add a small
 amount of sulfur dioxide to destroy residual chlorine. Alternatively, a sec-
 ond groundwater system may be used to recover water from a deeper aquifer
 for supply purposes.
   While temporary interruptions in service may not cause further environ-
 mental damage, they can be quite expensive. Thus, it is important to specify
 good quality equipment with a reputation for reliability. Appropriate materi-
 als of construction including piping that is compatible with the water source
 must also be specified. Downtime delays remediation and requires  addi-
 tional maintenance that will quickly exceed me savings realized from pur-
 chasing less reliable equipment. While it is not necessary or always prefer-
 able to buy the most expensive equipment, equipment is not the place to
 economize.
   Enhanced performance and reduced labor costs can be achieved by the
 use of automated control systems. Programmable logic controllers  (PLCs)
 hay! become quite  sophisticated.  PLCs provide standard relay-type opera-
 tion for control of surge tank water levels and injection and recovery flow
: -"rates. Control can |>e based on response to meters 'or on a programmed time
 sequence of events. A computer pfograiri for the specific system is  devel-
 oped in conjunction with the system vendor. Modifications to the program
 are readily made in the field or by remote command. The  return on the
 investment in PLCs can be substantial based solely on reduced labor costs.
 However, the biggest potential for return is the ability to maintain the system
 closer to optimum and reduce the time for required remediation.
                                          	it;
 5.23  Process and Instrumentation Diagrams
 	[' Vi	  i1   ,  , «i • in !' „ , "  • " ' !,n , ,,|!iil r'•',••!	IJ'H",' I1!,*"' |M", ,,,,,! ,,,|.,,|,I1: „.:.'!,,  ,! , '«"!' ,..,•• ,i »  •,, ",IN . , , %',•",. ij Jihillk " „
    Process and instrumentation (P&I) diagrams are necessary to ensure de-
 sign quality and to facilitate efficient and correct installation. The level of
 detail required depends upon whether the design firm is also doing the instal-
 lation.  If the work will be put out for bid, the level of detail needs to be
 greater because of the potential for misinterpretations and to limit the oppor-
 tunity for change orders. For Superfund and RCRA projects, the P&I draw-
 ings will need to be highly detailed and will undergo review at several stages
 of the process.
    Several types of P&l diagrams are necessary. These include the following:
         •   total system showing major components;
         •   injection, monitoring, and recover}' wells showing screened inter-
            vals, grouting, and surface completion;
                                  5.22
-------
                                                             Chapter 5
        •  water treatment system and offgas treatment system, if included;
        •  nutrient and electron acceptor delivery systems;
        •  piping from recovery well(s) to treatment system, to discharge
           and to surge tank, to injection well, and from delivery systems to
           injection water header;
        •  electrical systems; and
        •  meters, gauges, and controllers.
   Figures 5.6, 5.7, and 5.8 show typical P&I diagrams for the total system
layout, for the nutrient/electron acceptor delivery system, and for the piping
system. Typical diagrams for wells, electrical, and groundwater treatment
systems (Section 5.4.6) are adequate.

5.2.9 Sample Calculations
   Many of the calculations involved in this process are used elsewhere or
are relatively fundamental.  These include calculations of contaminant mass,
nutrient and electron acceptor requirements, and groundwater flow. The
dissolved mass is estimated by multiplying the aquifer pore volume by the
average dissolved contaminant concentration.
   Total contaminant mass associated with the soils can be estimated from:
                                                                  (5.1)


   or,
                M(lb) = V(ft3) * D(lb / ft3) •  avs                     (5.2)

where:
        M   =   mass of contaminants;                        .
         V   =   soil volume;
         D   =   soil density (typically 1,350 kg/m3(105 lb/ft3]); and
       Cav   =   average soil contaminant concentration.
   The stoichiometric electron acceptor and nutrient requirements are pre-
sented in Table 5.5.  These are only starting values and need to be adjusted
for the factors discussed in Section 5.2.2, including the mass of contaminant
removed through groundwater extraction.
                                 5.23
-------
Groundwater Treatment Systems
    :i:	•;;..  , .<: .^ ' ••	,s.^,.	•,  •.-.. Figure 5.6
  Groundwater Recovery, DewdteringVahd Oil-Water Separator Systems
From 2
Recovery Wells •< 	
Groundwater Only
,,, ' , i
From?
Recovery Wells > i
Total Fluids
l
' HOA
/fiS\ 	 (
1
T:30l H
Oil-Water
Separator



' '
	 .
T-302 Liquid , 1
Product Storage ¥-1
To
Containerizatlon
and Disposal

HOA ! HOAxjs.
i
i
k !
DW Points ^ '<-S TT71
! P-311 I™1
•jl, 7 Dewatering Pump s^—\ 	
DW Points * ;i <^
P-310
Dewatering Pump
TnR
. . . , ,
OA
(jl§) *•
-®f
t>tl — 	 	
_JL§HH...
^wf
> — i i
"t*/1^ •*•
P-308
^ — i
^isy c
*
PLC Control O
Panel ^

|ii" i , N '" : .i! '
H>

JM]
— fe-P
! Groun
• Treat
;
i

    DW Points
                                       P-309
                                   h Dewatering Pump
           Dewatering
            Point
   The total volume of injected water that will be required is calculated from
|he cpncenlxation of the electrpn acceptor(s) ami/or nutrients as introduced

into the aquifer.
                                                                   (5.3)
                                  5.24
-------
                                                                         Chapter 5
where:
         Mea   =   required mass of electron acceptor [kg (lb)];
          V.   =   volume of water to be injected [M3 (ft3)]; and
         C     =   concentration of electron acceptor in the injection
           63
                    water (mg/L).
                                     Figure 5.7'
                               Air Injection System
                 PLC Control
                   Panel
            HOA
     AIExhaust  F-101
     Separator Air Inlet
             Filter
                             •   Q3  S-101    ^
                          ••{Ml	~ AI Exhaust F_io2
                           •—'   A A Separator Ai 0xhai
                                          Filter
                               B-102
                              Air After
                               Cooler
                                                                         ^.To 16 Additional
                                                                           AI Point Sets
                                                       ;»  > f- >  ^ :>
                                                       I  I i a  i I
                                                  -C-3-C
                                                 4- 1/2 in. ID by 5/8 in. OD
                                                  )  HOPE Hose
                                                 + 6BO ft Maximum
                                                    0.040 in. Slot Screen
                                               Air Injection
                                               Point A-IOI
                                                (Typical)
                                        5.25
-------
Groundwater Treatment Systems
    o
   t>
    0
 »!>

 t?
 5 o

                           5.26

-------
                                                             Chapter 5
                               Table 5.5
          Effective Range of Electron Acceptor Concentration
                   and Cost for the Raymond Process
Electron
Acceptor
and Source
Air(02)
Oxygen (O2)
Sodium Nitrate (NO3")
H202(02)
Effective
Concentration
(mg/L)
8
'40
45
235
Kg Toluene
Kg electron acceptor
source
15
3
52
6.5
Cost ($) for Electron
Acceptor per kg
Toluene
0.00*
0.40
5.00
12.00
 'Only costs are capital and O&M.
   The rates of groundwater recovery and injection are determined by con-
ducting pump tests and incorporating the results of the test in a groundwater
flow model that can simulate capture and flow lines from the injection to the
recovery well(s). Anticipated recovery rates can be estimated from soil clas-
sifications and tables found in books, such as Freeze and Cherry (1979).
This process is highly dependent upon the judgment of the geologists who
classify the soils.  More often than not, this process overestimates sustain-
able yields.  Further, unless the recovery well is within the source area and
surrounded by injection wells, it may not be  advisable to operate the well at
maximum yield because significant volumes of clean water will be captured
requiring a significantly larger groundwater treatment system than would
otherwise be necessary.  In general, injection wells will sustain lower flow
rates than recovery wells.
   Based on the anticipated injection rates and the concentrations of nutrients
and/or electron acceptors, the mass of electron acceptor that can be introduced
in one month or one year is calculated. It is then a simple matter to estimate the
remediation time from this value and the demand calculated earlier.

                T = ^L                                            (5.4)
                    Ri

where:
         . V.   =   volume of water to be injected  {M3 (ft3)];
         R^   =   injection rate [m3/d (ft3/d)]|.
                                  5.27
-------
Groundwater Treatment Systems
   The total volume of water that can be added over the estimated
remediation time can then be used to determine the required average concen-
tration of nutrients that will be needed to meet demand. The concentration
of nutrients at startup should be two to four times this value, with the antici-
pation that the nutrient concentration will be adjusted based on observations
made during operations. If precipitation of calcium, iron, and/or magnesium
is indicated, the minimum concentration of KTPP is calculated from:

                K5P3010(g) ^ Mg(g)"  Fe(g) + Ca(g)                '"
                   448    ~   24     56      40

   Jar tests can also be used to test the calculated nutrient formulation. Jar
tests are conducted by adding a volume of water to each of several jars con-
taining various amounts of nutrient concentrate. Visual inspection is used to
determine the lowest concentration of nutrients that will not result in precipi-
tate formation. If the calculated minimum concentration or that determined
from jar tests exceeds the concentration based on demand, nutrients should
be added intermittently at the higher concentration^ Because water quality
will change over time, the calculations and/or jar tests should be repeated
after one pore volume of water has passed through the contaminated aquifer.
   The time for injected water to reach the recovery wells or sweep through
the contaminated zone should not be excessive. Long residence times equate
to long remediation times because numerous pore  volumes of injected water
will be required for remediation.  Additionally, long residence times result in
greater loss of nutrients and electron acceptors tlirough nonproductive reac-
tions and adsorption. The loss of electron acceptors to other reactions is
particularly critical when hydrogen peroxide is used. Long residence times
are m'3st easily addressed by closer spacing of wells as discussed in Sec-
tion 5.2.2.  Residence times can be approximated from the injection rate and
the volume of soil through which  the water will pass.  First the velocity of
groundwater flow is calculated as:
                                                                 (5.6)
   or
                VQft'/d)-   Q(ft3/d)
                 A    '
                         W(ft)»h(ft)«n
                                 5.28
-------
                                                            Chapter 5
where:
         Q   =   injection rate [mVd(ftVd)];
        W   =   width of sweep zone [m(ft)];
         h   =   depth of sweep zone [m (ft)]; and
         n   =   porosity.
   The time of sweep through a given zone is then calculated from the
groundwater flow velocity and distance from the injection well to point of
interest (i.e., a monitoring or recovery well).

5.2.10 Safety Requirements
   In addition to the common health and safety considerations  associated
with environmental activities and covered by the OSHA the following must
be addressed:
        •  electrical systems must meet local codes including  ground fault
          protection;
        •  mechanical equipment must be supported on concrete pads;
        •  pressure release valves must protect each pressurized system and
          be located such that personnel cannot be struck by the discharge;
        •  fugitive emissions must be controlled through treatment or the
          use of stacks;
        •  chemicals must be stored properly, including segregation of
          noncompatible chemicals (e.g., hydrogen peroxide  and combus-
          tible materials) with appropriate warning labels and first aid in-
          formation clearly accessible;
        •  hydrogen peroxide systems must be designed by suppliers or
          distributors, constructed of appropriate materials, and contain
          adequate pressure relief. Lines should not be located over per-
          sonnel access areas;
        •  liquid oxygen tanks  should be placed on concrete rather than
          asphalt pads because of fire hazard;
        •  eyewash  and first aid supplies should be kept in an  accessible and
          easily identified location;
        •  project-specific health and safety reviews should be held along
          with personnel training (Hazard Communication); and
                                 5.29
-------
Groundwater Treatment Systems
        •  site security must be sufficient to eliminate casual, unwanted
           visitors who may hurt themselves or damage equipment and
           increase potential harm to site workers and/or official visitors.
                         , i      -    i • i"      j " '     '           • '• '  ii -'"
5.2.11  Specifications Development
   The selection of all equipment, meters, valves, piping, controllers, and
chemicals should be compared to the appropriate specifications. Vendors
and suppliers should be required to provide the appropriate documentation
regarding materials, quality, and performance.

5.2.12 Cost Data
   The costs associated with a Raymond Process system are highly
site-dependent and a function of the hydrogeology, treatment zone dimen-
sions, contaminant properties, loading contaminant mass and distribution,
clean-up targets, regulatory design limitations, and specific design used.
Because of the large variability in costs among relatively similar sites, cau-
tion and engineering judgment must be used to extrapolate these representa-
tive costs to a specific site.
   The cost of any system is a function of engineering, capital, and operation
and maintenance (O&M) costs.  O&M costs are associated with operations,
maintenance, labor, chemicals, and electricity and can easily exceed other
costs. Additionally, project management, and reporting costs can be signifi-
cant. Designs that increase the rate of introduction arid distribution of elec-
tron acceptors increase capital and monthly operating costs, but can shorten
the duration of the remediation project.
                  	               . ,         i
   The electron acceptor must be selected based on the contaminant and
known metabolic pathways as discussed in Section 2.1. After this criterion
is met, the selection is based on cost-effectiveness. However, the cost of
supplying the electron acceptor may be less important than the impact of the
selection upon total remediation time because this determines the number of
months or years over that which O&M costs wilt'be incurred. Table 5.5
shows the effective concentrations that might be achieved with oxygen
sources, nitrate, and sulfate; the relative amounts of toluene that could be
biodegraded; and the cost of the electron acceptor per kilogram of contami-
nant treated. Table 5.6 indicates typical durations of Raymond Process treat-
ment applications using various electron acceptors along with an estimated
total project cost for each application. The total cost consists of capital
qosts, O&M costs accrued over the lifetime of treatment, as well as the inci-
dental costs, which may include limited use of the property or third-party

                                5.3o':'"
-------
                                                             Chapter 5
impacts. Table 5.7 provides a typical cost breakdown for a 3-year hydrogen
peroxide treatment project.
                               Table 5.6
          Typical Cost* of Aquifer Bioremediation Using Various
              Electron Acceptors in the Raymond Method
Source Electron Acceptor
Air 8 mg/L
O2 30 mg/L
HjOj 200 mg/L
NO 3" 50 mg/L
NO 3" 200 mg/L
O2+NO3' 30/50 mg/L
O2 + NO3' 30/200 mg/L
NQ3" + H2O2 200/200 mg/L
Months
100
40
20
40
15
25
10
7
K$/Month
10
11
12
10.1
10.4
11.1
11.4
12.4
Total (K$)
1,000
440
240
404
156
278
114
87
•Including capital, O&M, and incidental costs.
5.2.13 Design Validation
   Design validation should include review of all calculations and equipment
specifications including comparison with other projects. The design should
also consider problems that have occurred at: other sites. Sustainable recov-
ery and injection rates should be tested at representative wells. Permits
should be reviewed so treatment process discharges meet regulatory and
permit limits including both air or water quality and flow rates if discharges
are made to sewers.
   Treatability studies do not always need to be conducted. Easily con-
ducted tests including pH, bacteria counts, analysis of potential toxins, elec-
tron acceptors and nutrients, and degradation byproducts, as well as histori-
cal water quality data can provide a good indication of whether bioremedia-
tion is an appropriate remedial technology for a given site.  Microcosm stud-
ies can validate that microbial populations capable of degrading the
                                  5.31
-------
 Groundwater treatment Systems
                               Table 5.7
               Typical Raymond Process Cost Breakdown
ill
li'iiiii1 ii
Cost Element
Unit Costs ($)
No. of Units
ENGINEERING
Work Plan 3,000 1


Aquifer Tests
Laboratory Tests
':,;" Modeling
Design
Subtotal, Engineering
TREATMENT COSTS
Well Installation





Transfer Lines
Pumps & Controls
Nutrients System
H2O2 System
Water Treatment System
O&M (annual)
Management (annual)
Reporting
Subtotal, Treatment Costs
10,000
!i '" ''' 15,000
•i Hill'"' '..:..!*'; .if"'.'1
5,000
' 	 «, • • . .; 	 i •• . *
25,000
	 i1
2,000
20/ft
1 	 	 i •
7,000
20,000
20,000
25jooV'
110,000
12,000
10,006
'"" ;'l
1
1
1
20
600
1
1
	 1
	 3
3
5
Costs ($)
1 ii
3,000
10,000
15,000
5,
-------
                                                           Chapters


  In general, requirements for design validation will be more strict for
Superfund and RCRA sites than for state-mandated sites, such as UST sites
at service stations.

5.2.14  Permitting Requirements
  Permitting requirements vary among states and sometimes within states
where local or regional agencies exist.  Typically on Superfund and RCRA
sites, permits are not required, but the conditions of the permit regulations
must be met. The types of permits that apply include:
        •  well installation (some jurisdictions);
        •  discharge to groundwater;
        •  discharge to surface water bodies;
        •  discharge to sewers;
        •  air discharge;
        •  building;                                           '
        •  electrical;
        •  unique local regulations;
        •  wetlands; and
        •  coastal waters.
   Most of these permits are obtained as for any other remedial technology.
The discharge to groundwater permit may be stringent in some locations
because nutrients and electron acceptors are being introduced. For instance,
nitrate may be limited to concentrations below its maximum contaminant
level (MCL) of 10 mg/L as N. Some states allow nutrient injection, but
require  that ammonium and phosphate levels not exceed background or
regulatory levels at the completion of the project.
   As part of the permitting process, public notification and hearings
may be required.

5.2.15 Design Checklist
   Table 5.8 provides a relatively comprehensive list of items that need to be
addressed in the design and implementation of the Raymond Process for
aquifer remediation.
                                 5.33
-------
 Groundwater Treatment Systems
                                           Table  5.8
                          Raymond Process Design Checklist
 Site Characterization
 Site Features
 Type and distribution of contaminants
 Soil characterization
 Contaminated soil and ground water volume
 Aquifer characteristics
  1  ;,:'(; !' ..... • ...... vvif'T'''-1 ?,.' " ,-•     ( "i •'•
 Groundwater chemistry
 Property lines
 Nearest power source
   ;J> •,••' ''!'•'> '"	|l if "«,.; ,'  I .,-  ;l ' I  • •
 Building/confined space locations
      :      '-ij-/ i,..--'    : • ,|  .'
 Underground utilities
 Concrete/asphalt surfaces
 Normal use and traffic patterns
Aquifer Testing Results
Well Design
 Adequate hydraulic conductivity
 Appropriate microbiology
                          I  '
 Appropriate electron  acceptors
 Nutrients compatible  with groundwater
 Estimated treatment  time acceptable
                         "t
 Screened intervals
 "',,	•;„  ', 11  ' *#•••.  ;	'• '" .',  	Bi, • ]'V'>!!'',":, i|,;
 Well locations
 Completion methods .
                                                                                         ua" .ill*.'"
Water Treatment
All regulated compounds treated
Precipitation potential addressed
   ',,/ jit	 " ' ', ,1,    ",	:p 'ill , ',:!	 li "'ifij 	 ''
Sufficient capacity
Residuals addressed
Discharge water quality acceptance
Groundwater Transfer System
Lines deep enough to avoid freezing/heat traced
Valves, flow meters, control valves
Surge tank controls
In-line filters
                                             5.34
-------
                                                               Chapter 5
                             Table 5.8 (cont.)
                   Raymond Process Design Checklist
 Amendment Addition          Adequate storage
                         Compatible materials of construction

                         Check valves/pressure relief

                         In-line filters


 Health and Safety            HASP complete/kept on site
                         Dig safe contacted

                         Ground fault protection
                         Electrical interlocks
                         Fire Department notified (H2O2)

                         Security
                         Emergency shutdown procedures


 Miscellaneous               Permits secured
                         O&M manual reviewed
                         Monitoring plan/schedule
                         Treated soil placement

                         Site restoration
5.2.16 Implementation
   Implementation requires coordination of several activities. Wells must be
installed using the appropriate drill rigs and installation techniques, and
completed at the proper elevations for monitoring wells (e.g., surface or
aboveground  completion). Injection and recovery wells must be completed
either at the surface or beneath ground in vaults depending upon where the
force mains and transfer lines are placed. Force mains and underground
transfer lines should be installed from well manifolds to points of connec-
tions to tanks and pumps.
   Equipment support pads, treatment buildings, storage buildings, electrical
service, security fencing and lighting, access roadways and paths, if needed,
and grading should be completed before equipment is delivered. Equipment
and tanks should be placed on support pads or in a building as specified.
   Treatment equipment, pumps, meters, valves, and controllers should be
installed and tested to the extent practical. Repairs or replacements should

                                   5.35
-------
                             Groundwater Treatment Systems
                             be made immediately.  The system should be inspected to ensure that
                             specifications have been met. Finally, all appropriate local agencies
      !	,   ,  ',',..  ,;::,,!,	,.".".".     should be notified.         ,  '  ,  "''"',',!	'..,	  ''"'	,', '          	"" ""	"""„

                             5.2.17 Start-up Procedures
      .^ ,        :  .;i, ;         '!.•••       ,  ';:::   ': ••''.•''•  lfl.	•'	:••.-::;.. *?:•.'' :/,l ' ::  , ':.',•  ;:	':: •',. •;  :; ,;.: ;;L,!;''yi'-
                                Start-up procedures vary somewhat with the design, electron acceptor
                             used, and groundwater chemistry.  Startup begins with checking the! system
                             to ensure that no installation-related problems exist.
                                Baseline measurements of water quality and bioremediation parameters are
                             made prior to initiating groundwater recovery. A detailed list of wells to be
                             sampled, parametersito be analyzed in each well, and sample preservation re-
                             quirements should be prepared, reviewed, and useS as a guMe in the field. A
                             similar list, including measurements of temperatures, flow rates, and pressures
                             across the system, should be prepared for the duration of the start-up period.
' .               'ii	•'  '•'             .    .••:••'.••.','  ••     "i ' .s. <:,- :  !»!,'•" ':':	• ,-. :	  <,txi,t. "is,',"!!1 •• ••;	-v:	:!••:!••  ,  ,• • :.: .'.•;'*:,IB v>
                                Groundwater recovery is initiated with 100% of the treated water being
                             polished with activated carbon and discharged to a sewer line or surface
                             water body until water quality can be demonstrated to meet regulatory re-
                             quirements. After this condition has been met'! a portion of the groundwater
                             can be diverted to the surge tank.  When the level in the surge tank exceeds
                             me'low level" the transfer pump is turned on and water is introduced to the
                             injection wells through the manifold or header.
      ih          	 \	 ,'"' ,      ,    !    , „ "     ' '    i       	" .! '    i  ''	''' ', inn'"1' M IM i 	'	I ,i N!"  ,. i  '    ii '. '   ' '	'  i!	!	 '"'
                                Groundwater recovery and injection are continued without nutrient or
                             electron acceptor addition for a specified period depending upon the ground-
                             water geochemistry.  Nutrient addition is typically initiated prior to electron
                             acceptor addition.  If hydrogen peroxide is used, nutrient addition is main-
                             tained for 1 to 2  weeks prior to initiating hydrogen peroxide addition. Hy-
                             drogen peroxide is first added at a relatively low concentration, 20 to 50 mg/
                             L, and successively increased about once per week until the design concen-
                             tration of 100 to 500 mg/L is attained.
                               During the start-up period, the schedule for measuring and recording flow
                             rates, taking meter readings, and collecting samples for analysis is followed.
                             Samples of the influent and effluent to the groundwater treatment system are
                             collected for analysis. Nutrient and electron acceptor concentrations in the
                             feed solutions and  in the injection water are determined.
•'   •  :    •   :"i!"  	   •'"•'      • •  •	: ."   	   ' " • •  	•••".  	''•':'•".	;V  .--j- ».',-;	:	"• ,;••'•	'•>•	•  ' '     -'	 '••	'•	'	- ;
                               Water levels are measured for evaluation and verification of the computer
                             model used to design the system. This will allow periodic adjustment of the
                             recovery and injection rates from individual wells to achieve the desired
                             groundwater flow patterns.

                                          .       ,,.                                              I
               p.  •• •   , •  i!      '  „     .            :                        I
         • ' 	'   :::   '           '"   •'  • '•   ;" ' '               5.36
                ' J!    ,.          ' >,     <    '       '                      I  I I I                 I ||  I  II
-------
                                                            Chapter 5
   During this period, flow rates and other parameters are modified to meet
specifications and/or to optimize performance. Since the system will not
behave entirely as expected, modifications to the O&M manual should be
anticipated based on the outcome of this start-up period.  Monitoring and
documentation procedures are also evaluated and changed as necessary.

5.2.18 Performance Evaluation
   Performance evaluation includes determining changes in grouridwater
quality, nutrient and electron acceptor concentrations at monitoring loca-
tions, groundwater recovery rates, injection rates, groundwater treatment
efficiency, and changes in piezometer surfaces.  In all but the smallest sys-
tems in relatively homogeneous formations, it is highly unlikely that the
system, as originally designed, will provide the best practical remediation.
The O&M plan should incorporate procedures to evaluate performance and
modify operations. This requires that the monitoring plan be designed to
identify optimization refinements as well as to satisfy regulatory require-
ments and measure progress of treatment. Potential changes in operational
procedures include modifications to: (1) reduce operator time; (2) change
requirements; (3) minimize monthly operating costs; or (4) better distribute
nutrients and electron acceptors to reduce the treatment time. As discussed
in Section 5.2.1, the project design and budget should provide for the instal-
lation of additional wells, if needed.

5.2.18.1  Operation Practices
   Operations include injection and recovery well maintenance; balancing
groundwater recovery, discharge, and reinjection; maintenance of groundwa-
ter treatment; management of nutrient and electron acceptor addition; evalu-
ation and response to monitoring data; routine maintenance of equipment,
controls, and monitoring equipment; and housekeeping.
   Maintaining flow in the injection wells is frequently the most time-con-
suming activity listed above. The rate of remediation is closely linked to the
rate of introduction of the electron acceptor. If the rate of water injection
decreases, the number of months over which O'&M, reporting, and manage-
ment costs are incurred increases. Design and well completion procedures
for injection wells are important, as is scheduling well redevelopment at a
frequency based on experience with different types of aquifers, mass of or-
ganics to be degraded, and design flow rate. Performance of and anticipated
problems with injection wells can be evaluated from changes in the cycling
frequency of the high/low controllers for gravity-fed wells and by pressure
changes  in pressure-fed wells.  Wells can be treated by surging to remove

                                 5.37
-------
                                      Grpundwater Treatment Systems
                        ,.,'lf
   fines. Biological growth and precipitation of calcium or iron can be ad-
   dressed by adding dilute hydrochloric acid to the weli and subsequently
   recovering the spent acid after several hours. This may kill bacteria within a
   few feet of the well, but this is not a concern as remediation of this area
   should have occurred within the first few days of operation.  A better ap-
   proach is to add a batch of 1 to 3% hydrogen peroxide to kill bacteria. This
   results in a more easily-removed (particulate) biomass than does the use of
   dilute acid which often results in a slimy biomass that is difficult to remove
   from the well.
     To optimize the groundwater flow patterns, the model used, to design the
   treatment system is used during operation to adjust flow rates based on the
   current piezometric surface. The model output is used to determine if nutri-
   ents and electron acceptors are being proportionally introduced based on the
   contaminant distribution.  If not, model simulations can suggest modifica-
   tions to flow rates at individual injection and recovery wells. The impact of
   the changes in flow rates can be evaluated within a few days by measuring
   the changes in the piezometric surface and again running the model. Use of"
   the model provides operational guidance more rapidly than waiting months
   for monitoring data to indicate how to change well operation; however, the
   model must be used in conjunction with the field monitoring data.
     The distribution of treated water between injection and discharge is con-
   trolled by the level of water in the surge tank. Manual adjustments to valves
     I 'i, , i'''i ''  ,' ,  •  I", ' i i"1,! 11'" |,' ,1 , 'ii., ',,i"',i,  ' r 'i'i'i,:,'i, i ''i |,j|ii"!»! !l:l!'.|,|.!IP|'l'.'M'!'i:',''	i.lili'lP1'"  , ",,'"i ,11 •• I "t , ' • , * , ,' '•,' ,nih,!> , „  , • ,  1	r ,"i ,,: I11!,,	'InJ  	P'iiMli" '
   are made to prevent excessive cycling within the surge tank while ensuring
   adequate feed rates.
     Operational practices of the groundwater treatment system depend upon
   the treatment process used. Groundwater treatment operational practices
   and monitoring are critical because of the potential to discharge water that is
   put of regulatory compliance to surface water, groundwater, or sewers, all of
   which canresult jn.fines and poor relations with the regulatory agencies or
  the public. Operational practices for biological treatment systems are dis-
  cussed in Section,,$,4.|<>,Activated carfepn systems are relatively simple.
  The principal activity consists of monitoring the water quality in the infli'uent,'
  effluent, and between the carbon units.  The use of three units in series per-
 •-;', •"_ 	 	 		I	'	' • '	  I	 i 	
  mits greater loading than using two in series, but  requires an additional
  analysis to be performed. The first unit in series is removed when spent, and
ij. . • "•„ 'I'iji i '  ."'A'1; • ^ /'IP	,,,	,iiii"iiiii! in i •!' . ,ii "  i "i11 AH 'ii'1,1",1!!*  :	KIN	i,,,' iiiiniinni!,1,,,'«!>,!" •, "in1 • , i1  i11!,!1!"1  •  /..pi	„',,'.',. ,'p !«,„  ,	nip •,,«j i*	A ,1, i inniiiiiiPiiKtupi,, v
  a new.iinit is placed at the effluent end of the series.  Air strippers, either
  towersi or low-profile strippers, routinely require maintenance to remove
  precipitates of iron and/or calcium and biomass.
                        ; ;:  .•;.  ,      •   •  •.     _   	5.38'

"I "lll'il	 Hip I  |  i )  HI    i  i.     I   ill i    i          I    I!|	,'!!      I i        i      i{
-------
                                                                   Chapter 5
5.2.18.2 Operation Monitoring

   As with most remedial systems, monitoring includes baseline sampling
and analysis. Monitoring is most intense during the first few days of opera-
tion and decreases in frequency over the first few months.  To the extent
practical, monitoring should be conducted with instrumentation and auto-
matic recording devices. For remote sites, it is particularly advantageous to
link these devices to an off-site location to reduce travel and labor costs.

   Generally, for a Raymond Process system, the parameters listed in
Table 5.9 are monitored. Initially, the most critical parameters are the influ-
ent and effluent water  quality parameters from the groundwater treatment
system.  If the groundwater treatment system is not meeting the specified
criteria, the entire system will need to be shut down. The groundwater treat-
ment system influent quality is also used to cumulatively measure the mass
of contaminant being removed as one indication of remedial progress.
                                   Table  5.9
       Required Monitoring Parameters  Used in the Raymond Process
         Parameters
                                   Comments
                                                             Locations
 Organics



 Nutrients




 Electron Acceptors)



 Inorganic Ions
 pH, Temperature, eH, and
 Conductivity
 Groundwater Levels

 Recovery/Injection, Discharge
 Flow Rates, and Pressure
Specific (e.g., benzene), and
surrogate (e.g., TPH or TOX)
Nitrogen (e.g., NO3',, NH4+,
TKN, and Phosphorus (e.g., total
phosphate)
DO, NO 3-, or SO 42
Fe, Ca, and Mg for Precipitation
Concerns and Others if Fresh
Water Supply is Injected
Monitoring/Recovery Wells
Treatment System, and
InfluentfEffluerit

Monitoring/Recovery Wells
Storage Tanks, and Injection
Header
Monitoring/Recovery Wells
Storage Tanks, and Injection
Header

Monitoring/Recovery Weils,
Treatment System     :
Influent/Effluent, and Injection
Header

Monitoring/Recovery Wells
Treatment System
Influent/Effluent Injection,
Header, and Storage Tanks

AH Wells

Injection/Recovery Wells' and
All Transfer Lines     ;
                                    5,39
-------
	i;	;	I ln
	 i	
                                      Groundwater Treatment Systems
                                         Monitoring for groundwater quality changes is necessary to meet regula-
                                      tory requirements and to evaluate remedial progress. Data interpretation
                                      requires a detailed understanding of the process — degradation will occur
                                      first near the injection wells, to some extent the more degradable compounds
                                      will be treated first, the more"solublei compound
                                      grouridwater extraction, and solubilizafibn effects from biosurfactants will
                                      frequently result in increased dissolved-phase coriceritrations. Thus, these
                                      data need to be interpreted in the context of the mechanisms of remediation
               	      r    ;  ,,     ,  '  ,|i, , , | .   ,   ' '    ill! '! " . | ' ^      ,.„,„,,, 	 	   ,„ ,, „,,,	 ,,	, ,, , 	  	,	   , , „, , , i i,	
                                      and the sequence of their occurrence along gi-oundwater flow paths. It is
                                     , important tnat the regulatory•agencies understand mat increased concentra-
                                      tions of various constiwents^grburidwater|iriay occur before improve-
                                      ments to groundwater quality are achieved.
                                         Biodegradation parameters, especially nutrient and electron acceptor
                                      concentrations, are initially intensely monitoredin the vicinity of the injec-
                                      tion wells to evaluate flow patterns relative to contaminant distribution. Dis-
                                      solved oxygen (DO), pH, conductivity, eH, and temperature can easily be
                                      measured on-site using readily available meters.  (For DO values less than
            ••   :>\",\ •'•''.  l!-;'i: . til  iff  •   ',' .1 mg^, some practitioners prefer kits'	based on trie Winkler method although
                                      there is some question of the benefit of accurate DO measurements for DO
                                      levels less than 1 mg/L). Phosphate, ammonium ion, nitrate, nitrite, carbon-
                                      ate, sulfate,'magnesium, mariganeseVcnlbridei and iron can be measured
                                      using water analysis kits while on site. During startup, such rapid access to
                                      data can be beneficial.  For routine operations!, it is not always cost-effective
                                      to use the kits, and the results are often not as defensible as data  obtained
                                      from an analytical laboratory.
                                         It is equally important to measure the depth to water in monitoring wells
                                      frequently during me first several months of operation.  As previously men-
                                      tioned, the groundwater model used to design the well layout and flow rates
                                      can be used to evaluate groundwater flow patterns and to adjust flows from
                                      individual wells.  The model should also be evaluated and calibrated as nec-
                                      essary during the course of the project based on measured field data.
                                        The monitoring program should also include measurement of water flow
                                      rates from individual wells, headers, groundwater treatment influent, discharge
                                      line, and 'individual injection welds. The most practical memod of obtaining
                                      these flows is to use totalizers and recordme data onla predetermiried schedule.
                                        Nutrient and electron acceptor concentrations are measured in the concen-
                                               	   ••     „, ,A „,   	   „!	   	,	   , ,,   ,  ,| 	
                                      trate tanks following deliveries and in the header at a location downstream of
                                      the mixing point. These values are compared to tank level records for con-
                                      sistency and to the O&M schedule to verify stable performance of the sys-
                                      tem over time.
                                                                         _
                                                                   (;V Ki'F-: iJiiiili^                 .' ,!''-l
-------
                                                          Chapter 5
5.2.18.3 Quality Assurance/Quality Control
  Quality Assurance/Quality Control (QA/QC) procedures applicable to the
Raymond Process include practices common to other bioremediation tech-
nologies (i.e., the use of blanks, blind duplicates, and spiked samples for
laboratory and field measurements).  Quality practices specific to the
Raymond Process are:
       •  determination of composition of the nutrient and electron accep-
          tor concentrates;
       •  comparison of current groundwater recovery flow rates and cu-
          mulative flow from individual wells to the current flow rates and
          cumulative flow from the injection header and discharge line;
       •  comparison of the distribution of nutrients and electron acceptor
          concentrations as indicated by monitoring data to those predicted
          from the groundwater flow pattern simulated by the computer
          model (tracers can also be used);
       •  comparison of biological parameter data with changes in ground-
          water quality and mass removed from groundwater extraction;

       •  review of health and safety practices;                   ;

       •  review of operating practices and training of new field per-
          sonnel; and
       •  routine evaluation of monitoring and metering equipment,
          valves, etc.
5.3  Intrinsic Remediation


5.3.1  Principles of Operation
   Intrinsic remediation is the process of site assessment, data reduction, and
interpretation that quantifies the natural assimilative capacity of a given
aquifer system to treat groundwater contaminants through physical, chemi-
cal, and/or biological means without human intervention.  Through the in-
trinsic remediation plume management approach, the nature and extent of
soil and groundwater contamination, and the extent and rate of natural con-
taminant degradation at a site are determined.  This approach is appropriate
if the plume has not impacted a downgradient receptor and the rate of
                                5,41
-------
  Groundwater Treatment Systems
  contaminant release from the source area is equal to or less than the contami-
  nant degradation rate observed at a site.  Dissolved plumes containing those
  compounds shown in Table 2.2 to be biologically transformed under aerobic
  and anaerobic conditions could potentially be managed using an intrinsic
  remediation approach.
   ;  "',	i .I,,1 "', '"", i"l- .Mil!,1, ' "I ,	i,||i.'" "i ', n.!|, !!'	!'l"|| ,	ili'i , i",!1! J,i\ini|4|i,f:»i''il'i''',  ,• ,n 4	11 -I"'' 11 " '  •'  ''""''' '      ''' " '   I ' •!'• '*« "i1'  ' '," "
    A number of field sampling protocols are available from a variety of
  sources describing approaches for collecting and analyzing data necessary to
  verify that iritririsic remediation processes are taking place (Wiedemeier
  1994; Wilson et al. 1994).  The connection of these data with decisions re-
  garding source removal activities or with estimates of source lifetime have
  generally not been presented in the literature. An approach for implement-
  ing intrinsic remediation concepts from data collection through source re-
  moval and source lifetime considerations has been developed for the US
  EPA and the ILST Air 'Force (Dupont	et	a£	'i"99l7""l^'tJWRt' i'997).
   ,,'PI	'Mjijjli  r,'1!,, , !' " , „ ,„',!„„ ';!;'' |,, ! , '• \, ', , tf,,,,'r  li'.'i^ , JV ,.*•, , ,f ,„„ „ TJL.T ;•::, •„	 , n,,,,, „„	 ,, ,;,,.'! •„,• • ,,,r	F,	„  „	r |f , 	„•.,„„•;; • ,,n •« ri ;•••,••• • ,„
  Tjiese concepts and procedures''are presented in the following sections.
  5.3.2  Process Design Principles
    Intrinsic remediation assessment involves a seven-step process outlined in
  Figure 5.9. For a given site, this process involves: (1) determining whether
  steady-state plume conditions exist, (2) estimating contaminant degradation
  rates, (3) estimating the source mass, (4) estimating the source lifetime, (5)
  predicting long-term plume behavior with arid without source removal, (6)
  deciding whether to use intrinsic remediation and/or source removal, and (7)
  developing a long-term monitoring strategy if intrinsic remediation is se-
  lected for plume management.

  5.3.2.1  Determination of Steady-State 'Plume Conditions
    Verifying that steady-state conditions exist for a contaminant plume at a
  given site is critical in establishing that intrinsic remediation processes are
  taking place and are likely to provide "continued"' plume containment under
  current site conditions.  Steady-state plume conditions occur when the rate
  of contaminant release from the source area is equivalent to the rate of con-
   ;•; ; ,',", ',;	•'•,:;; j	:;, •:  1, ,        i              I   J.		 .,,	,,,-„(.,.,	 ,
  taminant assimilation by biotic and abiotic processes taking place within the
  aquifer.  Steady-state conditions can be identified by observing contaminant
  concentrations at specific groundwater monitoring locations over time.
  However, a better approach is to evaluatecontaminant concentration an3
,','„, 	•• ' ' .«   '  '  ,,!,'! !	,•„!!!'•	• '•	 .:,!"! '	* IS SI'	^ ",„	 * "„»',!',!'»'!,«" i!!!l||!!!l	,!,, 	,,j	,!!! !!|l| 	!!„,! '"!!,«,' '!'!*' '„"!,'"	",'!!!',' ',   ! "!, , ,'",'' '  !
-------
                                                            Chapters


contaminant concentration and aquifer volume for each monitoring point is
summed to yield a total dissolved mass for the plume.
   Aquifer volume is determined from the product of the aquifer porosity,
the average aquifer thickness (generally the length of the largest sampling
interval used within the monitoring network at a given sampling time), and a
plume surface area associated with each sampling point. One procedure that
can be used to obtain an estimate of area associated with each sampling
point is the Thiessen Polygon Method.  This method was developed in the
field of hydrology for estimating areas associated with point rainfall mea-
surements within rain gage networks. The Thiessen method assumes  that the
concentration measured at a given sampling point is equal out to a distance
halfway to the sampling points located next to it in all directions. The rela-
tive weights (areas) represented by each sampling point are determined by
the construction of a Thiessen polygon network, the boundaries of which are
formed by the perpendicular bisectors of lines connecting adjacent points
(Chow 1988). The construction of an example polygon network is shown in
Figure 5.11. The outer boundary of the Thiessen polygon network is  esti-
mated based on the outermost well locations.  lit is important for boundary
definition to be consistent if mass calculations are to be comparable among
sampling events. It is also important to note that this method can be used for
estimating mass within a monitoring network consisting of as few as three
monitoring wells. However, an increase in sampling point density through-
out the plume will improve the accuracy of the plume mass calculations, as
interpolation among data points will be improved due to the shorter interpo-
lation distance.  Ideally, from 10 to 20 monitoring points throughout the site,
both inside and outside the contaminant plume, can be used to provide irea-
sonable accuracy in plume mass estimates and plume delineation for  a rea-
sonable cost.
   In addition to estimating the total mass  of a compound within the dis-
solved plume at a given time, the representative: center point of the combined
plume mass can also be calculated. This representative mass center is i
termed the centroid of the mass (CoM) and is calculated by taking the first
moment of inertia of the mass at each sampling location within the contami-
nant plume about specified X and Y axes.  Example calculations for both
dissolved plume mass and CoM are provided in Section 5.3.9.
   These CoM calculations are useful for tracking and interpreting the move-
ment of contaminants, reactants, and products within the contaminant plume
over time. They can also aid in assessing the status of the plume and inter-
preting its migration pattern over time as indicated in Table 5.10.
                                 5.45
-------
II       II
                                                  Groundwater Treatment Systems
                                                                                                                           ""if . v.i ••!	I1 ?'|W1PI'Mi"".' . ' '.'ff'-t
                                                                                         Figure 5.11
                                                                  Example Thiessen Polygon Network Construction
                                                                    i   /, ,	i .r,,,.
                                                                   Monitoring Points
                                                                                                               Plume Boundary
                                                   a. The outer boundary of the sampling network is identified based on logical, physical boundaries of the
                                                      problem. Each sampling location is then connected to all adjacent points to form a series of polygons
                                                      with the sampling points as their corners.
                                                                   Monitoring Points
                                                                                                               Plume Boundary
                                                   b. The lines between these sampling points are bisected, and perpendicular lines are drawn at the
                                                      bisection points. These perpendicular lines are then extended so that they intersect one another.
                                                                                                         Plume Boundary
                                                                        Monitoring Points
                                                   c. The intersecting lines are connected to form polygons associated with each original sampling location
                                                     to yield unbiased and consistently generated areas. These areas can then be used to generate associated
                                                     groundwater and soil volumes that allow the determination of the mass of contaminant within the
                                                     assigned plume boundary and the changes in that mass over time.
                                                    • Bisectors
                                                    > Bisector Extension
                                                    I Associated Area Boundary
                                                                                            5.46
-------
                                                           Chapter 5
                             Table 5.10
   Changes in Contaminant Mass and Mass Center Coordinates and
     Corresponding Interpretation of Plume Mobility and Persistence
Contaminant Mass
Increasing

Constant
(Steady State)
Constant
(Steady State)
Decreasing

Decreasing
Centroid of Mass
Moving
Downgradient
Moving
Downgradient
Stable

Moving
Downgradient
Moving
Upgradient
Interpretation
Continuous Source; Unstable Plume; Contaminant
Migration
Finite Source; Plume Migration; Minimal Natural
Attenuation
Continuous Source; Stable Plume; Contaminant
Attenuation
Finite Source; Plume Migration; Contaminant Attenuation

Finite Source; Plume Attenuation; Rapid Contaminant
Attenuation; Optimal Intrinsic Bioremediation
  If plume centerline analysis and CoM calculations suggest that the plume
is growing over time, steady-state conditions have not been reached, and
either ongoing monitoring should take place to ensure future attenuation of
the plume, or active source removal and/or site remediation should occur if a
sensitive receptor is or will be impacted in the near term.  If the contaminant
plume is shown to have reached steady-state conditions, further quantifica-
tion of the nature and extent of plume attenuation taking place under site
conditions is warranted.

5.3.2.2 Estimation of Contaminant Degradation Rate
  Contaminant degradation rates can be estimated based on dissolved plume
contaminant mass data if a declining mass of contaminant is observed pver
time, or on contaminant groundwater concentration data if the source pro-
duces steady-state dissolved mass in the plume over time. If steady-state
mass is indicated, degradation rates for the contaminants can be estimated
directly from centerline concentration data or through the calibration of a
contaminant fate and transport model to field groundwater data. Figure 5.12
presents the logic associated with the estimation of field-determined degra-
dation rates and  suggests that if aquifer flow data are available, the use of a
fate and transport model accounting for advection, dispersion, sorption, and
degradation is preferred over the use of plume centerline concentration data
alone. In addition, the use of less degradable "plume resident tracer" com-
pounds in the calibration process is desirable for the calibration of the trans-
port component  of the fate and transport model if data for these tracer  :

                                 5.47
-------
                  Grpundwater Treatment Systems
    i o, >  i1 TO:
compounds are available. The use of the less-degradable tracers simplifies
modeling because the transport components of a model can be calibrated
without having to consider degradation reactions.
                                                Figure 5.12
                        Decision Logic for Evaluating Contaminant Degradation Rates
   , SO1:.]"  llili1!!	'
t 'V '	.iJHU'Mii . 
-------
                                                           Chapters
  Dissolved Plume Mass Changes Over Time.  Dissolved plume mass
changes over time can be used as an indicator of the type of plume existing
at a site as summarized in Table 5.10. When the total mass of contaminant
in the dissolved plume is decreasing over time, a finite source is suggested.
Both its position and concentration profile would not be expected to be
steady-state, and it would behave as a pulsed source. To estimate the degra-
dation rate of contaminants within the plume resulting from this pulse
source, the changes in total contaminant dissolved plume mass should be
analyzed over time.
  A classical approach to the evaluation of contaminant degradation rates in
biological systems is to analyze the changes in contaminant concentration or
mass over time to establish the relationship between concentration or mass
versus reaction time using zero- or first-order reaction rate laws. Zero-order
reactions are described by a contaminant reaction rate independent of con-
taminant mass (i.e., a constant mass degradation rate over time) or:
                dM/dt = -k0                                    (5.8)

where:
        ko   =   the zero degradation rate constant (mass/time),. The
                 integrated form of this equation is shown in Equation 5.9:
                M = M0-k0t                                     (5.9)

where:
        M   =   contaminant mass at time t (mass); and
        M   =   the initial contaminant mass at time t = 0 (mass).
          O                                                    i
   If the reaction is governed by a zero-order degradation rate lavs/, a plot of
contaminant mass versus time produces a lineair relationship, the slope of
which equals ko and whose intercept value should equal Mo.
   First-order reactions are described by a contaminant reaction rate that is
dependent on contaminant concentration or mass (i.e., a mass degradation
rate changing over time), or:

                dM/dt = -k,M                                  (5.10)

where:
        k,   =   the first degradation rate constant (I/time). The integrated
                 form of this equation is shown in Equation 5.11:

                M = M0e-k''                                     (5.H)

                                 5.49
-------
Ill 11	'
 11	k
                         ,;; i'fB  • jii:ji!
,!!! ' III!	
                                     Grpundwater Treatment Systems
   r .	          I           I          : ;| :	Ill „!„,,;, •.':'' ,' . jf '',„! ,, .. .
   I/1 ' ,;,   11, ,       i         11   lii  n    i|in pi  'i  », lip:1,",K >',, ,;,|,,|!l!iii,,l,1 «' "hi,,18	„ '        i  in
   A plot of contaminant mass versus time produces a non-linear relation-
ship that can be linearized by plotting the natural log of contaminant mass
versus time. The slope of this linearized relationship is equal to k,.
   !•: ';>j  ',	,,I,H,,,I;, '-I'lji,,1,. i1,' Hi-	:i"i	,;„',,,^r	i1 „;, ,i „ iiiaini	a1,:,1	L>	'.i	•••£	'»"	„ v; vj ji, T,;,	m >	. ••	BI	  ,  • ,L,e	
   Plume CenterUne Concentration Data. Plume centerlme concentration
data can be used to quantify cb                    rates if the dissolved
plume mass does not changesigniiicaritly" overtime (i.e., if a continuous
steady-state source is indicated) (Table 5.10). Using the data reduction ap-
proach described above for dissolved plume mass, contaminant concentra-
tion data can be analyzed using zero-order reactions with Equation 5.12:
                                                     dC/dt = -k
                                                                 (5.12)
                                        ,  ,,
                                     where:
                                              ko  =  the zero degradation rate constant (mass/volume/time).
                                        The integrated form of this equation is shown in Equation 5.13:
                                     .where:
                                                                            .                        ,
                                              C  = ; contaminant concentration ^at time t (mass/volume); and
                                              C   =  the initial contaminant concentration at time t = 0 (mass/
                                               o
                                                      volume).
                                        A plot of contanoinarit concentration versus time produces a linear rela-
                                     tionship, the slope of which equals ko and whose intercept value should
                                     equal Co.
                                        First-order reactions using contaminant concentration data are written as:
                                                                                                     "(5.14)
                                     where:
                                                  =  the first degradation rate constant (I/time).
                                        The integrated form of this equation is shown in Equation 5.15:
                                                       = C0e-klt
                                                                (5.15)
                                        A plot of the natural log of contaminant concentration versus time is lin-
                                     ear' when firs^prde^de^adation is" taking "placet	wMTthe slbpe'of thislinear-
                                     ized relationship equal to kr This data analysis approach has been incorpo-
                                     rated into an intrinsic remediation protocol developed by Chevron Research
                                     and Technology Company (Buscheck, ftefliyTanci1 Nelson 1993).
                                                                      5.50
                       ,' , , „;,!:  ,"7 Jill •  '"'
                       irji	tils:,, rsa/'	:	
-------
                                                           Chapter 5
   Calibration of Analytical Fate and Transport Groundwater Models.
When a continuous source is observed at a site, calibration of analytical fate
and transport groundwater models provides the best estimate of contaminant
degradation rates as these models integrate transport, retardation, and degra-
dation using site-specific contaminant and aquifer properties. An analytical,
one-dimensional flow, three-dimensional dispersion model developed by
Domenico (1987) is one such model that can be used in intrinsic remediation
methodology. Use of this model accounting for flow and contaminant sorp-
tion characteristics, in addition to degradation yields, a "dilution-corrected"
degradation rate. When calibrated with nondegradable "tracer" compounds
(i.e., dimethylpentane or trimethylbenzene isomers) in the source are^, the
model provides improved degradation rate estimates for the more reactive
compounds (e.g., benzene, toluene, ethylbenzene, and xylenes BTEX) of
health significance. The use of this model for intrinsic remediation assess-
ment is presented in a case study in Chapter 8 of this monograph.

5.3.2.3 Estimation of Source Mass/Lifetime
   With an estimate of the rate of contaminant degradation at a site, manage-
ment decisions regarding the appropriateness of source removal actions and
the effect of such actions on the projected lifetime of contamination at the
site can be made.  The logic associated with source mass and lifetime deter-
minations is shown in Figure 5.13. Specific calculations for the estimation
of contaminant source mass and lifetimes as a function of plume type are
detailed below in Section 5.3.9.
   If a continuous source is found at a site, contaminant mass within the
source area will continue to contribute mass to the groundwater, maintaining
the contaminant plume footprint that has developed over time. To estimate
the potential lifetime of this plume, an estimate must be made of the mass of
contaminant existing as residual saturation both above and below the
groundwater table, along with that residing in  any free product occurring at
the site. Ideally, these estimates should be based on soil core arid free prod-
uct samples collected throughout the site. This total mass estimate requires
that the soil and free product volumes associated with each  soil core or
monitoring location be defined using a procedure such as the Thiessen Poly-
gon Method previously described.
   If soil core data are unavailable, the source area mass can be estimated
from groundwater concentration and free product measurements in observa-
tion wells and groundwater monitoring points. Concentrations of contami-
nants in groundwater above the equilibrium values expected based on
Raoult's Law (water concentration = solubility • mole fraction in

                                 5.51                         ;
-------
i .. hi   i	 ':



il >. i1!1"     "li'ir
                                Groundwoter Treatment Systems
                                  -"":	•   •"                     Figure 5.13
                                Decision Logic for Evaluating""Contaminant Source Mass and Source Lifetime
                                                           3/4. Estimate
                                                          Source Total Mass
                                                            and Lifetime
                                                            Plume Type?
                                             Pulse
                                                                             Continuous
                                          Lifetime Based
                                          on Degradation
                                         Rate of Dissolved
                                             Mass
          Lifetime Based
          on Total Mass
             at Site
                                          Last Dissolved
                                           Mass Value
Mass Above
Groundwater
  Table
Mass Below
Groundwater
  Table
                                                                             ,,!,,,,
                                residual-phase product) suggest residual saturation in soils below the water
                                table. With the extent of mobile product and/or residual saturation estimated
                                based on groundwater measurements, and residual fuel saturation values as a
                                function of soil type (Parker, Lenhard, and Kuppusamy 1987) available in
                                the literature, source area mass can be estimated. An example of such calcu-
                                lations is provided hi a case study in Chapter 8.
                               : • it" • •; ::•• "'   ' ; •.     ', i1;;: i  ' r „; j'*,i:v; v £.15 • • \-f. • wi':"'! -fii:; f"11 \t»: y»wt^ ''' •',  ••••/. = ,•;	;; i,;-; /,  •'. • 'i*;,,, * • f ji% !H ^ , i ;
                                  Once the total mass of contamination is estimated above and below the
                                groundwater table, estimates for the total lifetime of the plume can be  made
                                based on the total mass disappearance rate described in Section 5.3.2.2.
                                These calculations are presented in detail in Section  5.3.9.
                                                                   5.52
-------
                                                               Chapter 5
5.3.2.4  Prediction of Long-Term Plume Behavior

   The long-term behavior of a contaminant plume is impacted both by the
characteristics of the source, which affect the duration of the release of con-
taminant into the aquifer, and by the characteristics of the aquifer itself,
affecting the transport and degradation of contaminant once it is released
from the source area. Figure 5.14 presents the decision logic related to
long-term source behavior, identifying differences in analysis of the plume
based on whether it is a pulse or continuous source.
                               Figure 5.14
       Decision Logic for Evaluating Long-Term Contaminant Plume
                        5. Predict
                        Long-Term
                        Behavior
                         Plume
                         Type?
        Pulsed
                                         Continuous
       Dissolved
        Mass
      Degradation
Without
 Source
Removal
 With
 Source
Removal
                                Total Mass
                               Degradation
                                        Vadose Zone
                                          Source
                                          Removal
                           Vadose and
                           Saturated
                          Zone Source
                           Removal
                                   5.53
-------
             Groundwater Treatment Systems
               Iftheplume is a.pulsed source, no residual source area exists, and the
             long-term behavior of the plume is related to the projected lifetime of the
             plume. If the site contains a significant source area, producing a "continuous
             source" plume, long-term plume behavior can be evaluated based on various
             source removal scenarios. If no source removal is to be carried out, a
             worst-casescenario develops in terms of the length of time the plume will
             persist, as the plume lifetime calculations for the sum of vadose zone, satu-
             rated zone, and dissolved plume masses estimated above apply. If contami-
             nant source removal is being considered, the effect on plume lifetime of
             mass removal from various locations at the site> ami at various levels of re-
             moval efficiency can be evaluated.
                                          ,  • /   , ,      •  |.L i   i      •        .  (•• '.•',
               Once source removal strategies are investigated, the complete long-term
             behavior of the contaminant plume can be predicted using a groundwater
:'*M:  ''!!!  .•. ,, i^fateand'trarisport'mbdeL As indicated by Gofderetal. (1996), source re-
             moval activities can be modeled by superposing a "negative" continuous
             source plume on top of the existing steady-state plume concentration profile.
             This "negative" plume is generated using a source concentration equal to the
             negative of the initial source concentration at a point in time corresponding
             to the time of source removal.  With this superposition, movement of the
             steady-state plume away from a source area that has been eliminated follow-
             ing source removal activities can be modeled.' Ah example of such a model-
             ing approach is provided in a case study in Chapter 8. It is important to note
             that the steady-state contaminant plume profile represents the highest
             downgradient concentration profile that would be expected at a given site.
             "The concentrationsi at a given^ point in space will decrease over tune follow-
             uig source removal activities if the plume is truly at steady-state and if all
        ;  .; ^ other site conditions remain the same.	('	i	'^	   ^	^	_
             5.3.2.5 Decision Making feegardlh'g'lnfrlnslc Remediation
            '»•• ' '  i, ; '  , .    I. • "'!  i, . p :i!,JN,i'!,,,:il	»',!' 	I ,r	 '	 	 ,'" '	,,,(,,„•,:, i,,,i,'	•' 	,|IIM ,,„	',„,','	a fijl Jl|ll|,''»	II	,,',,,,i,,:"	 	 ,,l71l,l,i,,:	,ll,j|.,,l ',; ,• •:„, i]	•,	 iV.Ji ., (,!'	|n!i,|,M'	i
               The analysesi described above provide a basis ^        decisions re-
             garding the applicability of an intrinsic remediation plume management
             approach for a site. Figure 5.15 provides a summary of the logic necessary
             to complete the decision-making process based on the impact the plume has
             on downgradient receptors and the "potentialfor intrinsic attenuation reac-
             tions to contain and control the plume.

               The final questions that must be answered regarding application of an
             intrinsic remediation management approach at a site are: (1) whether a sen-
             sitive receptor is being impacted now or in the future when the plume is
             projected to reach steady-state conditions and (2) whether the projected
             lifetime of the plume is acceptable to owners/operators, regulatory agencies,
              11       '  '  "                                        '
-------
                                                                  Chapter 5
                              Figure 5.15
        Decision Logic for Evaluating Applicability of Intrinsic
           Remediation Plume Management Approach
                                          6. Intrinsic
                                         Remediation for
                                             Site?
                                       Impacted Receptors
                                          Now or When
                                         Plume Reaches
                                          Steady-State?
                          No
                                                                 Yes
                      Plume Lifetime
                     Within Acceptable
                        Limits?
                                       Apply Active
                                       Remediation
                Yes
                                    No
           Evidence of TEA
          Pool Sufficient for
            Contaminant
            Assimilation?
          Apply Active
          Remediation
     Yes
                          No
Consider Intrinsic
  Remediation
   Approach
Apply Active
Remediation
                                  5.55
-------
«" ' i!,
1 , „
                                   Groundwaier TreatmentSystems
                                   and other interested parties.  In general, if an existing or projected receptor
                                   impact exists, active source removal and plume control/remediation will be
                                   required unless institutional controls (i.e., deed restrictions, etc.) can be put
                                   into place to restrict the long-term use of contaminated soil and/or ground-
                                   water. The issue of plume lifetime tends to be more complicated. If signifi-
                                   cant contaminant mass remains in the source area of a site, the resulting
                                   plume may persist for decades. If remediation goals are established with
                                   shorter tirneframes (i.e.? for property transfer reasons, etc.), this assimilation
                                   time will likely be unacceptable and active remediation may be required.
                                      the focus of the previous discussion has been on quantifying the transport
                                   and degradation of contaminants under actual site conditions. After a pro-
                                   jected s"0"urce lifetime is deemed acceptable and. intrinsic remediation is con-
                                   sidered viable at a site, final supporting evidence for verification that degra-
                                   dation reactions are biologically mediated must be provided through an
                                   analysis  of the changes in background terminal electron acceptor (TEA)
                                   mass compared to that within the plume itself.  If contaminant biodegrada-
                                   tion'fs taking place, indigenous organisms will consume TEAs (O2, NO3",
                                   Mn4*, Fe3+, SO42% CO2) at a rate and to an extent that should correspond to
                                   contaminant loss observed at the site. The stoichiometry associated with
                                   microbiai metabolism known to occur~under various TEA conditions (Table
                                   5.11) allows a determination of the potential contaminant assimilative capac-
                                   ity of background groundwater moving into the source area and available
                                   within the plume itself. If this theoretical assimilative capacity is equal to or
                                   greater than the level of contamination observed at the site, expressed both
                                   on maximum concentration and total mass of contaminant bases, biological
                                   intrinsic  remediation processes can be expected to play a major role in con-
                                   taminant attenuation. If assimilative capacity is limited, some source re-
                                   moval and/or active site remediation action is likely warranted. An example
                                   of the evaluation of "potential "site' assimiiati.ve capacity is provided in Section
                                   5.3.9 and in a case study in Chapter 8 of this monograph.

                                   5.3.2.6  Long-Term Monitoring
                                      If an intrinsic remediation management approach is selected for a given
                                   site, the last step in the assessment process is the development of a long-term
                                   rnonitqring strategy.  Figure 5 16 shows that the requirements of the monitor-
                                   ing strategy are twofold:  compliance monitoring and intrinsic remediation
                                   process monitoring.
                                      Compliance monitoring must be conducted to provide data to the regula-
                                   tory agency to confirm that plume containment and risk management con-
                                   tinue to take place at the site.  Compliance monitoring normally involves an
                        I1!,,!!1,;!!) [„  ..I'li!,!! „    '    ' w^   ' '       '"'„;'          I I     111  II II       I      I         ' „[  '' ',. ,'''„ : ] iiifll" •' •;	'',
                        •	'-'-  '•'••-" •'    :     ••• •  •    '•     '  .             5.56
-------
                                                                          Chapter 5
                                     Table 5.11
         Potential Hydrocarbon Assimilative Capacity Relationships
             for Electron Acceptors of Importance at UST Sites*
TEA Indicator
Oxygen
Nitrate
Fe^toFe2"
Sulfate
Organic to CH 4
Compound
Degraded
Aromatic
Alkane
Aromatic
Aromatic
Alkane
Aromatic
Alkane
Aromatic
Alkane
Molar Relationship
(gmol/gmol HC
Degraded)
-7.5
-9.5
-6
+ 30
+ 38
-3.75
-4.75
+ 3.75
+ 4.75
Mass Relationship
(g/g HC Degraded)
-3.1
-3.5
-}.07
+ 2.15
+ 24.7
-4.6
-5.3
+ 0.77
+ 0.88
* These molar relationships were determined using the stoichtometric relationships presented in Weidmeier et al.
(1994) No specific kinetic rate or metabolic pathway should be infeired from these data  Negative values indicate
TEA use in reaction is quantified. Positive values indicate product generation is quantified. For solid-phase
reactants (i.e., iron and manganese), quantification of product generation will normally underestimate the total
assimilative capacity with these TEAS.
                                     Figure 5.16
               Requisite Components of a Lc>ng-Term Monitoring
                   Approach at an Intrinsic Remediation Site
                           7. Long-Term
                           Monitoring for
                               Site
       Compliance
       Monitoring
  Intrinsic
Remediation
 Monitoring
                              r
                                                                          T_
                         Update Site
                         Conceptual
                           Model
  Update
  Model
Calibration
Update Source
  Lifetime
 Predictions
                                         5.57
-------
 Groundwofer Treatment Systems
                                                                   I
 upgradient, background monitoring well; one to two monitoring wells within
 the contaminant plume; and one to two downgradient compliance wells used
 to detect contaminant migration toward potential receptors.  Groundwater
 elevation, contaminant concentration, and minimal groundwater quality data
 (pH, temperature" total dissolved soli3s) are generally required to be reported
 for these monitoring wells.
    Although the information generated for compliance monitoring is re-
 quired, it is insufficient for intrinsic remediation process monitoring; data
 from additional monitoring locations and for additional analytes should be
 collected for process monitoring. Figure 5.1? shows a monitoring well net-
 work tijat |s appropriate for initial intrinsic remediation evaluation during the
 site assessment phase as well as for long-term compliance and intrinsic
 remediation process monitoring. In addition to the data collection require-
 , ;  • :• ,,„„;!iii1!;;; ''!:,"'/ ri,1; jj,,,,,?», |  , 	,::|l ,| |  |L	|. ,,„. », - r  ; , r„ .,„,,„.,  „.., m,	..„. y.^	— „,,	i,,™, . .,.,	vis ••„•,•,••,!„!	,„ ..m..., , », •,,••,„ .;,; mi.,,	•*	:,  K,,\,,,m, , ,.B	.,„;.
 meiifs for compliance monitoring, intrinsic remediation process monitoring
;;. should,include^analysis of TEAs that aire	consumed	(b2," N03% SO42') and
 products that are formed (Mn2+, Fe2*, CH4) during contaminant biodegrada-
 tion, and assessment of water quality characteristics (alkalinity and oxida-
 tion/reduction potential) that indicate of biological processes within the con-
 taminated aquifer. Based on data collected from process monitoring, the
 conc&ptuai model of the site (mefueling gn"d use assumptions) and model
 calibration results can be periodically updated to provide ongoing refine-
 ments to source lifetime predictions and to risk assessment considerations
 	;,'" Hi, I!1 ,,|,lirj	,„;„..i1,	,',i'	 •„ Li1',	 i,!1'	''Hlf III11'I'll',,, ,  ":* 	,	n	 ,	,	,	 „  , 	, 	  	
 for,the site, 	  ',	 ,  .    .,	„„,,	'	,.	,	IM	  \'t	,'	
    Finally, the frequency of groundwater monitoring must be established as
 part of the long-term monitoring plan. Compliance monitoring schedules
 generally require quarterly to annual sampling. However, under most cir-
 cumstances; annual sampling will be the shortest time interval necessary for
 intrinsic remedjatipn process monitoring; because the low groundwater ve-
 locities observed at most sites do not warrant more frequent sampling inter-
 vals. At a site with a groundwater velocity of 6.51 ni/d (0.04 ft/d),
 unretarded groundwater moves less than 4.5 m (15 ft) in a year.  With a re-
 tarded velocity 1/3 to 1/6 that of groundwater (appropriate for benzene and
 xylene, respectively) contaminant movement of less than 1 to 1.5 m (3 to 5
 ft) would be expected over a 1-year time period. With a monitoring grid
 spaced at 10-m (30-ft) intervals, a 1-year change in plume position cannot be
 detected. Again, the sampling  interval should be assessed on a site-specific
 basis, but generally, an annual to biannual sampling schedule should be suf-
 ficient toensure that adequate data are collected while minimizing the sam-
 pling and analysis burden at intrinsic remediation sites.
                                 5.58
-------
                                                           Chapter 5
                              Figure 5.17
       Groundwater Monitoring Network for Both Compliance and
                Intrinsic Remediation Procei>s Monitoring
                                                            Point-of-
                                                           Compliance
                                                              fells
      O
   Background
 X Shallow Groundwater Monitoring Points
 O Monitoring Wells
 5.3.3 Process-Flow Diagrams
   Because intrinsic remediation is an in situ plume management approach
 that relies upon indigenous physical, chemical, and biological processes for
 the attenuation and destruction of contaminant mass, no engineered process
. flow scheme is involved.  The major system requirements relate to contami-
 nant plume delineation and long-term groundwater monitoring.  A typical
 monitoring network to provide the necessary initial site assessment informa-
 tion and long-term monitoring of the mobility and transformation of con-
 taminants within the groundwater plume is provided in Figure 5.17.

 5.3.4 Process Modification

 5.3.4.1  Hydrocarbon Plumes
   The intrinsic remediation process design principles discussed in Section
 5.3.2 apply primarily to dissolved contaminants that degrade under a variety
 of TEA conditions to short-lived intermediates. This situation is typical of
 fuel-contaminated groundwater where hydrocarbon constituents of the re-
 leased fuels are degraded both aerobically and anaerobically to CO2 and
 water with few intermediate products being detected in the plume. The

                                  5.59
-------
            Groundwoter Treatment Systems
	ll't	•;,	I:
intermediates that have been observed are; low-molecular-weight volatile
fatty acids generated in the oxidation of monoaromatic hydrocarbon fuel
constituents (Cozzarelli, Eganhouse, and Baedecker 1990; Cozzarelli et al.
1994)1 The presence of these intermediate compounds downgradient from a
source area indicates biological fuel contaminant degradation, and quantifi-
cation of such compounds may be warranted if additional evidence of intrin-
sic bioremediation is required. Cozzarelli et all (1990,1994) provide a de-
tailed description of the analytical methodology for separation and quantifi-
cation of these volatile fatty aci3s in groundwater.
                   111 l"l"''' "ill    il  I        III        ' :" 'i'1     nil I
                                           I          t    II:     I
   Additional verification of the biological nature of contaminant degrada-
tion processes taking place in hydrocarbon-contaminated groundwater sys-
tenis can be provided using plume-resident tracers (PRTs) as suggested in
Section 5.3.2.2. To be an effective tracer, mese compounds must be less
degradable than the compounds of interest (i.e., the BTEX components) and
must reside in the source area at concentrations high enough to produce a
measurable downgradient plume. If these bioresistent compounds exist in
the source, their relative concentration in the resultant groundwater plume
should increase with respect to the more degradable BTEX components as
the plume moves downgradient of the source area. The relative concentra-
tion (Wilson et al. 1994) or relative mass flux across the plume (Dupohtet
al. 1996) of the PRTs versus the BTEX components can be used to indicate
the biodegradation of 'BTEX' constituents in the plume. In addition, these
PRTs have been used to calibrate the flow portion of a groundwater fate and
transport model prior to calibration and degradation rate determinations for
the BTEX comts*(i[iraret dK>6,1997'fUWRL	1997)7' '"' 	' '*'  "'
              A number i	
            2,3-dimethylpentane;
            l,2,4-trlmethyl6enzene; and"i"^3,*>trimemylbenzene. Again, the resistance of
            these compounds to biodegradation within aquifer systems provides addi-
            tional evidence that biologically-mediated contaminant destruction is taking
            place at a site, and ifie quantification of these compounds may be necessary
            to provide additional verification to regulatory agencies that contaminant
            mass degradation,	not simply	dilution, is t'aklng1'place.,

            5.3.4.2  Chlorinated Solvent Plumes
              Industrial solvents, such as TCE, pcg5 carbon tetrachloride (CT), and
           chloroform (CF) are some of the most common pollutants found at contami-
           nated groundwater sites. Those compounds that are highly chlorinated (i.e.,
           PCE and TCE), are also highly oxidized and cannot serve as a source of
           energy (electron donor) to indigenous microorganisms under aerobic

                                            5.60
-------
                                                           Chapters
conditions. However, in the absence of oxygen and under highly-reducing
conditions, these chlorinated solvents actually act directly as alternative
electron acceptors through the process of reductive dehalogenation.
   Reductive dehalogenation results in the degradation of TCE to environ-
mentally acceptable products, such as ethyleue (Freedman and Gossett 1989;
de Bruin et al. 1992; DiStefano, Gossett, and Zinder 1992), ethane (de Bruin
et al. 1992) and CO2 in the laboratory through the general pathway shown in
Figure 5.18.  Mohn and Tiedje (1992) provide a review of microbial reduc-
tive dehalogenation research. Many studies, however, have reported the
accumulation of intermediate products in the dehalogenation process,
namely cis-dichloroethylene (cis-DCE) and vinyl chloride (Imbrigiotta,
Ehlke, and Martin 1991), which pose a threat to human health and the envi-
ronment. DiStefano, Gossett, and Zinder (1991) summarize of the literature
in this area, and McCarty (1994) has presented a recent summary of the
current state of knowledge regarding anaerobic transformations of chlori-
nated solvents in contaminated groundwater isystems.
   The pathway shown in Figure 5.18 describes the biologically-mediated,
anaerobic transformation of PCE and TCE in contaminated  aquifer systems.
Within the framework of intrinsic remediation previously described, ;the
detection of daughter products of these biotransformation reactions serves as
supporting evidence to suggest the natural containment of chlorinated sol-
vent groundwater plumes. In demonstrating the attenuation of these chlori-
nated solvent plumes, the following two primary aquifer characteristics must
be shown:
        • the aquifer is highly reducing (e.g.,  evidence of sulfate reduc-
          tion and methanogenesis exists at the site) such that anaerobic
          dechlorination would be expected to take place under field
          conditions; and
        • the PCE/TCE plume mass is stable (as determined according to
          procedures defined in Section 53.2.1) with daughter product
          mass data which indicate accumulation of intermediates from
          these dehalogenation reactions, or reduction in daughter product
          mass if further degradation of these intermediates can be verified.
   To provide the information necessary to make these determinations, in-
trinsic remediation assessment for chlorinated solvent plumes should include
sampling and analysis for water quality parameters indicative of aquifer
oxidation/reduction status (dissolved oxygen, nitrate,  sulfate, dissolved iron
and manganese,  dissolved methane, and dissolved hydrogen sulfide), parent
and intermediate compound groundwater concentration data (PCE, TCE,


                                 5.61
-------
 Groundwater Treatment Systems
              Pathway for the Anaerobic Dehalogenation of
                 PCE and Various Intermediate Products
   CC12 = CCI2 Tetrachloroethylene
                                                              Carbon
                                                            Tetrachloride
 Source
>: McCarthy
           1994
                                                            I       II
                               K i in III
cis- and trans-DCE, vinyl chloride, and ethylene), and electron donor species
driving the anaerobic dechlorination process (dissolved total organic carbon
or chemical oxygen demand).
"  "::'":	•	:""•	::":	 '"::; ::':  '•"•	::" "'' -:'	"•""' • -''>r	::T: ;:'T:•.::;:",' ":iH"'•' 7':"v:" '•"' - '":'   .:'  ":	T:'"I"?	:'
   This last parameter is important due to the inefficiency of electron trans-
fer, which  has resulted in a requirement of as mucli as 130 times the esti-
mated"reducing equivalents to sustain'de'Ealogen'atibh^  Gross parameter
measurements for electron donor concentrations appear appropriate as nu-
merous organic substrates including;'methane (Corapcioglu andHossain
1991; Enzien et al. 1994); ethanol, acetate, and lactate (de Bruin et al. 1992;
Gibson and Sewell 1992; Paylqstathis and ^uang 1993); methanol and
glucose (Freedman and Gossett 1989); and propionate, crotonate, and bu-
tyrate (Gibson and Sewell 1992) have been used by researchers to stimulate
reductive dehalogenation in the laboratory. Fiorenza et al. (1994) reported
                                 5.62
-------
                                                          Chapter 5
reductive dehalogenation of PCE and TCE at a manufacturing plant in
Ontario, Canada, using organic contaminants, such as naphtha components
and volatile fatty acids, as electron donors. An example of verification of the
intrinsic remediation of a chlorinated solvent plume is provided in a ease
study in Chapter 8.

5.3.5 Pretreatment Processes
   Since this is an in situ, plume monitoring approach, no engineered pre-
treatment processes are necessarily required. However, based on the assess-
ment methodology presented above, some source removal and active source
remediation may be recommended to accelerate site cleanup. Refer to Sec-
tion 4.2 for information regarding the in situ treatment of contaminated
source area soil; Sections 4.3,4.4, and 4.5 for ex-situ source area soil treat-
ment; and Sections 5.5 and 7.2 for integrated approaches for source area
product recovery and soil and groundwater treatment.

5.3.6  Posttreatment Processes
   Since this is an in situ, plume monitoring approach, no engineered post-
treatment processes are required.

5.3.7  Process Instrumentation and Control
   Since this is an in situ, plume monitoring approach, no process instrumen-
tation or controls would be required. If sourc e removal or treatment is being
carried out at an intrinsic remediation site to accelerate site remediation,
process instrumentation and controls for necessary soil handling and treat-
ment systems would be required. Refer to this appropriate source area treat-
ment technologies for their specific process instrumentation and control
requirements.

5.3.8 Process and Instrumentation Diagrams

   Since this is an in situ, plume monitoring approach, no process and instru-
mentation (P&I) diagrams would be generated. If some source removal and
active source remediation is recommended based on the assessment method-
ology presented above, P&I diagrams appropriate for the selected technol-
ogy would be required. Refer to the appropriate source area treatment tech-
nologies for their specific P&I diagram requirements.
                                 5.63
-------
!> iljX RSI illllljK |I1B '!! " 'T '<" ISIS pill!"', r ...... ',• HI I"! ....... l/
                     W: '! ..... • '""!• ...... ;; .' i/litRi ..... ' n flj ....... fMml",  '      ,            ,  !       .'^',''	.'."   ,''..""  L'
                            The total mass of a compound within the dissolved plume at a given point
                          in time is determined from the product of:  (1) contaminant concentration
                          associated with each sampling location (C.); (2) aquifer porosity (0); (3)
                          average aquifer thickness (H); and (4) plume surface area associated with
                          each sampling point generated from the Thiessen Polygon Method described
                          in Section 5.3.2,1,2 as follows;
                  1	  '   '  • "    '                                  i
                                *  '                                      i                    ' i
                                          MassT = J niasSj = J) C,(0XH)(Ap              (5.16)

               iii  '?!,,,ii  ;l •.. ''where:  	.	,
                               mass.   =   dissolved contaminant mass associated with sampling
               H  •';,;,;:  '  ;;„}ii:'  , ;'\"• * ''\, • " ••'.\;","...'.''^.location j (mass); arid" i'   '"''' _"'''"']"'	/  "'	,	  ', '" _'''"'"i
                                  A.   =   Thiessen Area j associated with sampling location i
                     ;;_;,;" :' ,' "}t:, !• "^ ,  	 :f •, ^ .^(ICHgth2).
                            Thie representative center point of the combined plume mass (CoM) is
                         calculated by taking the first moment of inertia of the mass at each sampling
                         location within the contaminant plume  about specified X and Y axes. Math-
                         ematically, this can be expressed as follows for the center of mass X and Y
                   III  !"'   '' il,	;	«•. ; -it  	i. "Mi,,!••»•,!.,11,11  	,	,	 '.'.	• „„ "	,	.,,,	I	 	• 	r	VMII:;TII;	 ,	'	•', »"	 ',|l ••	It. i. •. »v	 -;,r 	VI	,	,
                         coordinates,  respectively:
                        '!, !|.    . ". .''.lli,  '' "    "I1 !,.  • .||' ! ,1ft,, 	' ' '!'        ! i ,•"'•!  " J!  '..ililll!11'1 	.'	I" '.".ll1!,! '!' .  ,.| , !  '. 	'"  '  . ' „ ."I  ,'' ' , «  "'ill.' ' Ml1,,,

                                          "V    iol
                                          x = ——	—

                                                                                            (5.17)


                                               n
                                   ,,  ..   ,
                                    where:
                                v  «.'•.>   '.:.:'.
                                                     .     ,.  . . •.. ,..  •. ,.             :     . .  .    . . ••
                                                      x and y coordinates of each sampling location within the
                                                      Thiessen area network.
                                                                      s.64
it, . an ....... .: ..... r ii." 1 1 „ ii
                   .....
i, .isjiiviiiiiiii ..... iiiii..!!.!! ..... i1 iiiiiiiiii^ ..... gwia ilillil1 iiaiiis " ..... niniiii ....... ta
                                                                  .....    !, v;     i
                                                              ii 'jit, Jiiii!^           ...... 11 ....... i: it iii
                                                                                              ..... Ai: ; 3-111 : > .:;
-------
                                                            Chapter 5
   Table 5.12 summarizes typical calculations for total mass and center of
mass results generated from a monitoring network for which contaminant
concentration and Thiessen area values were obtained. Results from these
calculations indicate a total dissolved benzene plume mass of 3,696 g, with
CoM X and Y coordinates of +66.5 ft and -17.2 ft, respectively.

5.3.9.2 Estimation of Source Area Mass
   When a continuous source plume is identified at a site, source area mass
should be estimated based on soil core measurements above and below the
groundwater table. Figure 5.19 indicates the configuration of soil cores and
associated geometry used in Equations 5.19 and 5.20 for average borehole con-
centration, Cave, and total contaminant mass, Mp estimates in a source area:
                       IC.AJ
                  avg.j     ",

                         tr''J                                   (5-19)

where:
        C..   =   soil contaminant concentration in core j at depth i in the
                 core (mass contaminant/mass soil);
        h..   =   core j interval thickness at depth i (length); and
          n   =   total number of soil cores collected at the site.

   Total source area mass is calculated as:
where:
         A.  =  Thiessen area associated with core j (length2).
   The denominator in Equation 5.19 is the thickness of vadose zone con-
tamination for mass above the groundwater table, while it is the thickness of
contaminated soil below the groundwater table for saturated zone mass de-
terminations. Generally, total mass calculations provided via Equation 5.20
are carried out separately for mass above and below the groundwater table so
that the vertical distribution of contaminant mass can be evaluated.
   If soil core data are unavailable, then contaminant mass within a source
area can be estimated based on a determination of the extent and. composi-
tion of free product or residual saturation at a site. The extent of free

                                  5.65
-------
0,










Sampling


Typical Total
Data from

XCoor.
Location Number (ft)
CPT
CPT
CPT
CPT
CPT
CPT
CPT
CPT
CPT
CPT
CPT
CPT
CPT
CPT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
-8.4
54.6
-475
^5
-363
108.7
-415
59.4
-513
63.4
-16.0
64.7
108.7
19.0





Table 5.12




Mass and Center of Mass Calculation Using Field-Determined
Shallow Groundwater Monitoring Points (CRTs) and Monitoring

YCoor.
(ft)
-80.6
-122.4
-52S
-42.8
-74.6
55.4
•22
13.7
245
58.1
165
355
115
-85

Associated
Thiessen
Area (ft2)
1,490

1,767
1,272
2,152

1,745
1,737
2,669
1,600
1,108
1,008
2,492
1,207

Elevation Elevation Water Col.
TOC (ft) BOC (ft) Depth (ft)
97.43 6.71 - 424
plugged
9754 651 355
97.9 732 350
97.11 6\65 42}
9851 plugged
96.88 6.75 1.17
99.64 &86 3.19
97.10 733 2.77
9853 758 3.70
9738 6.65 3.80
9858 834 4.02
99.67 93 325
98.45 7.68 332

Volume
(ft3)
4,967

5,889
4,240
7,173

5,818
5,791
8,895
5,333
3,693
3,360
8,304
4,023

Benzene
ftig/L)
0.0

0.0
965
W

0.0
2,442
0.0
0.0
0.0
70.0
4,344
71.0






Groundwater
Wells (MWs)

Benzene
Mass
(g)
0.0
0.0
0.0
116
02
0.0
0.0
400
0.0
00
0.0
6.7
1,021
8.1

Mass-X
(g-ft)
0
0
0
-518
-7.4
0
0
23,794
0
0
0
431
111,016
154

Mass-Y
(g-ft)
0
0
0
-4,954
-153
0
0
5,501
0
0
0
239
11,704
-72
0
3
C.
Q.
Q
sf
i1
(D
-------
CPT 15 143 154 763 98.10 7.12
CPT 16 262 -354 98.63 plugged
CPT 17 9.1 33.8 2,479 97.68 6SO
CPT 18 -43 -615 693 97.65 7.00
CPT 19 520 -40.6 3,540 9951 854
CPT 20 285 -58.4 1,046 98.42 7.85
CPT 21 37.8 -825 5,384 98.78 820
MW 1 103.0 35.7 2,043 99.73 849
MW 3 0.0 0.0 791 97.48 6.70
MW 4 -172 -1375 4,086 9620 5.89
Max Depth (ft) =

423 2,543 406 292 419
0.0 0
350 8,261 2.0 05 42
523 2,311 2.0 0.1 -0.6
356 11,799 4,170 1,393 72,469
280 3,485 4,893 483 13,772
3.75 17,944 1.0 05 192
851 6,810 1,226 236 24,345
10.30 2,637 3.0 02 0
11.11 13,618 1.0 0.4 -6.6
11.11 X= 3,696 245,890
CoM Coordinates (ft, ft) 665
449
0
15.8
-8.1
-56570
-28202
4Z1
8,437
0
-53.0
-63,571
-172
Source: UWRL 1997
i













O
•g.
of
Ol
-------
                                        „•.":*"
 Groundwater Treatment Systems
                               Figure 5.19
                Configuration of Soil Cores and Associated
           Geometry Used for Calculation of Average Borehole
       Contaminant Concentrations gs Input to Total Mass Estimates
                                                       Top of
                                                     Contaminated
                                                        Zone
                              . Soil Core
                                         Yadose
                                          Zone
                                                      Water Table
                                                       Saturated
                                                        Zone
product can be estimated from observation well and monitoring point loca-
tions that contain free product during sampling. Because the thickness of
product observed in a monitoring well can be substantially greater than that
actually existing within the formation, care must be taken in interpreting
monitoring well product thickness results. Attempts should be made, how-
ever, to estimate free product volume both above and below the water table
so that the source contaminant(s) lifetime can be predicted. When the com-
position of contaminants within the free product, the lateral and vertical
extent of free product distribution, and the formation total porosity are
known, the mass of contaminant existing within the free product at the  site
(MfJ can be estimated as follows:
where:
      Area
 Thickness
         0
        c.   =
Mfp = Area(Thickness)6pfpCi
             .' '  " ;   ii1 '    ' i I

 estimated area! extent of free product (length2);
 estimated vertical extent of free product (length);
 formation total porosity (decimal);
 free product density (mass/volume); and
 contaminant concentration within the free product
 (mass/mass).
                                                                (5.21)
                                 5.68
-------
                                                            Chapter 5
   If high dissolved contaminant concentrations but no free product is observed
in monitoring wells, then product exists throughout the site as residual satura-
tion. Estimates of the maximum amount of residual-phase product existing at a
site can be made based on the characteristics of the soil at a site using the quan-
titative relationships presented by Parker, Lenhard, and Kuppusamy (1987) and
Mobil Oil Corporation (1995).  These relationships describe the typical residual
hydrocarbon saturation within the smear zone at and below the groundwater
table as a function of soil texture. For sands, a residual saturation of 25%; of the
total pore volume is suggested, while this value drops to 15% for sandy silts and
fine sands, and to only 5% for silty clays.
   With residual-phase product, the composition of the product can be in-
ferred from groundwater concentration data assuming that equilibrium exists
between the residual phase and the groundwater using the following relation-
ships based on Raoult's Law, an assumed molecular weight of the residual
product, and the known aqueous solubility of the individual compounds of
interest:

Equilibrium Concentration = Mole Fraction (Aqueous Solubility)      (5.22)

Mole Fraction = Equilibrium Concentration / Aqueous Solubility       (5.23)

                Moles in Product = Mole Fraction (Mfp) / MW^,     (5.24)

                Mass in Product = Moles in Product • MWcompound     ', (5.25)

where:
 MW        =   Molecular weight of compound (Ib/lb mol), and
     compound                   °                , „,    ..
     MW   =   molecular weight of the product (Ib/lb mol) =
                  120 Ib/lb mol for typical gasoline.
   This procedure for estimating residual-phase product and contaminant
mass in the source area was used at a site where former gasoline tanks had
released product to the soil and groundwater (Dupont et al.  1997). No free
product, only light sheens, had been observed in soil core and ongoing
groundwater monitoring samples. However, total petroleum hydrocarbon
(TPH) and BTEX groundwater concentrations approaching levels in equilib-
rium with free product were observed throughout much of the site. These
elevated groundwater data were used to delineate the apparent areal extent of
residual-phase material — approximately 776,,64 m2 (8,360 ft2). Based on
this area, the volume of soil existing below the site that contained this
                                  5.69
-------
 Groundwater Treatme nt Systems
 residual saturation and the vertical extent of potentially contaminated soil —
 4.11 m (13.5 ft) [3.05 m (10 ft) of measured groundwater contamination and
 1.06 m (3.5 ft) of capillary fringe and smear zone] — were estimated It was
 assumed that 10% of the pore volume of the fine sand and clay soil at the site
 contained this residual product material based on the findings of Parker,
 Lenhard, and Kuppusamy (1987). Based on an estimated total contaminated
 soil volume of 3,197 m3 (112,900 ft3) and a total porosity of 0.38, a total pore
 volume of 1,215 m3 (42,900 ft3) was calculated for the source area. The actual
 residual product volume was estimated to be 12T.5 m3 (4,290 ft3), resulting in an
 estimated 109,300 kg (241,000 Ib) of TPH below the site. From this estimate of
 product mass and the measured concentration of specific components in equi-
 librium with the product in the groundwater, the mass of each contaminant of
 interest within the source area residual product was estimated using Equations
 5.22 through 5.25 as summarized in Table 5.13.
• '• " * ' Table 5. 13 ' '! ' " ' ' : 	 "
Summary of Estimated Total Residual Contaminant Mass Based on
Residual Product Volume Estimates and Dissolved Plume Mass Measured
at a Field Site
Compound
Benzene
Toluene
Ethylbenzene
p-Xylene
Naphthalene
TPH
MW
(Ib/lb
mol)
78.1
92.1
1065
1065
128.2
120.0
Aqueous
Solubility
(mg/L)
1,780
759
135
221
30.6

Measured
Groundwater
Concentration
(mg/L)
4.9
3.2
1.9 	
6.3
0.79
. .. Xf, ','."!'
Source: Dupontet at. 1997
Mole
Fraction
in
Product
i i
0.003
0.004
0.014
0.029
0.026

'i : 	
Mole in
Product
(Ib)
5.5
8.5
" 285
575
52.1
/

Mass in
Product
(Ib)
431
782
3,025
6,111
6,681
241,000

Mass in
Plume
(Ib)
"..!• ] "'ill'"1 ', ,'::,,i!" • „
3.4
.•':; :i"! 	 t" i'"":
1.9
1 , . 1 '
9.6
1.2
'. 172'" ,''
, , 	 	 	
5.3.9.3 Estimation of Source Mass Lifetime
                                                  !              „, ll '  '"
   Based on the logic presented in Figure 5.14, the specific procedure to be
followed for source mass lifetime calculajions is Dependent upon the type of
source at a site. If a pulsed source is identified, little residual contaminant
                                 5.70
-------
                                                           Chapters


mass remains in the original source area, and the lifetime of the plume can
be estimated based on the dissolved plume mass from Equation 5.16, along
with the estimated contaminant degradation rate determined from Equations
5.8 through 5.11. A source lifetime (Tpu|se) can then be estimated as follows
for zero-order, and first-order degradation rates,, respectively:

               Tpulse,zero=(MassT)/k0                            (5.26)
                                          i

               Tpulse,first=-ln(M/MassT)/k1                     (5.27)

where:
        M   =  final mass that is to be reached at the end of the calculated
                plume lifetime.
   For a continuous source, the plume lifetime is the sum of the lifetime of
the dissolved plume mass plus  the mass remaining in the source area. The
total mass disappearance rate is equated to the contaminant degradation rate
determined from Equations 5.8 through 5.11 to yield an estimate of total
plume lifetime (Tcontinuous), for zero- and first-order degradation rate relation-
ships, respectively:

               Tcontinuoas>zero = (Massv + Masssz + MassT) / k0         (5.28)

               Tcontinuous>firat =  -ln[M / (Massv + Masssz + MassT)] / k, (5.29)

where:
    Massv   =  contaminant mass located above the groundwater table
                (mass);
    Mass    =  contaminant mass located below the groundwater table hi
         SZ
                the source area (mass); and
    Mass   =  contaminant mass located in the dissolved groundwater
                plume (mass).

5.3.9.4 Estimation of Source Mass Lifetime with Source Removal
   The effect of source removal efforts, either partial or complete, on the
overall lifetime of contaminant mass at the site can be evaluated through a
modification of Equations 5.28 and 5.29 presented above. For example, if a
significant mass of contaminant exists above the groundwater table (Lei,
Massv is large relative to MassT), the effect of vadose zone source removal
on the lifetime of the plume can be evaluated using Equations 5.30 or 5.31:
                                5.71
-------
 Groundwater Treatment Systems
                    imiouSlZero = [(1 - p)Massv + Masssz + MassT] / k0
           TcontinBOUS,fust = -ln{M / [(1 - p)Massv + Mass,, + MassT]} / kt
(5.30)
(5.31)
 where:
          p  =   % removal of vadose zone contamination (decimal).
   The impact of various removal scenarios on the plume lifetime can be
 evaluated using this general approach. If 100% vadose and saturated zone
 source removal is assumed, the continuous source plume lifetime equation
 reduces to that of a pulsed source as shown in Equations 5.26 and 5.27. An
 evaluation of the impact of source removal on predicted plume lifetime and
 the duration of management activities required at a site is presented in a case
 study in Chapter 8.

 5.3.10 Safety Requirements
   No safety requirements beyond those associated with proper conduct
 during environmental sampling and monitoring covered by OSHA are neces-
 sary for implementation of intrinsic remediation at a site.

 5.3.11 Specifications Development
   Specifications required to properly conduct an intrinsic remediation study
 at a field site apply primarily to:  (1) soil and groundwater monitoring point
 installation and construction and (2) soil and groundwater sample collection,
 handling, and analysis.  Recommendations related to monitoring point instal-
 lation can be found in US EPA RCRA groundwater and vadose zone moni-
 toring guidance (US EPA l986a, 1986b), whiie sample collection and analy-
 sis methodologies for a variety of analytes are described in detail in a variety
 of laboratory protocol documents available from US EPA (1984a, 1986c,
 1986d, 1989b). Many of these field and laboratory methods specifically
 relevant to intrinsic remediation studies have been compiled by a number of
 authors representing both the public (Wiedemeier et al. 1994) and private
 sectors (Buscheck and O'Reilly 1995; Mobil Oil Corporation 1995; Yang et
 al. 1995). The U.S. Air Force protocol prepared by Wiedemeier et al. (1994)
 is the most comprehensive in terms of its treatment of monitoring and ana-
 lytical approaches and should be consulted when preparing specifications for
 a field intrinsic remediation study.
   The specification document for conducting intrinsic remediation field
 assessments should emphasize sampling and analysis activities and, at a
minimum, contain the following:
                                        •I  i     '        ,      i „ ''''

                                5.72
-------
                                                           Chapter 5
        •  introduction to the site and test objectives;
        •  conceptual model of site and site contamination;
        •  sampling and analysis plan to be implemented at the site Including:
             •  hydrogeology assessment,
             •  water and soil gas sampling point installation requirements,
             •  soil core sampling requirements,
             •  groundwater and soil gas sampling requirements,
             •  field analytical requirements
             •  fixed-base laboratory analytical support, and
             •  sample handling, storage, transport, labeling, and tracking
               requirements;
        •  quality assurance plan (see Section 5.3.17.3);
        •  health and safety procedures to be followed during field and
          laboratory activities; and
        •  data summary and presentation requirements.

5.3.12 Cost Data
   The costs associated with implementation of an intrinsic remediation
plume management approach at a contaminated site involve initial site as-
sessment (monitoring point installation and sampling) and intrinsic
remediation process evaluation (data analysis, interpretation, and modeling)
activities, followed by ongoing site monitoring and process model improve-
ments during the life of the source area/plume.  Significant additional ex-
pense can arise if active source removal efforts are undertaken to reduce the
duration of monitoring required at a site.  Table 5.14 provides a summary of
costs associated with intrinsic remediation plume management.  Other sec-
tions of this monograph discuss costs associated with free product recovery
technologies and in situ and ex-situ remediation of the source area should
such efforts be necessary.

5.3.13 Design Validation
   Intrinsic remediation and natural containment of a contaminant plume
are validated through multiple lines of evidence as indicated in the in-
trinsic remediation assessment protocol presented in Section 5.3.2. At a
given site, validation of the intrinsic remediation plume management
approach is supported by:
                                 5.73
-------
	'i'1'llflBiiiji. 1li!.i|i:ii|ll||||H':|||	 'ir,;i|,;'|i,|i|!| ,1'T "
                    Groundwater Treatment Systems
                                                     Table 5.14
                             Typical Intrinsic Remediation Plume Management Costs*
                                                            Unit Cost
                                                                        No. of Units
Cost
                    INITIAL SITE ASSESSMENT/CAPITAL
                    REQUIREMENTS
                    Prepare Work Plan
                      "                  ,„      M    	lin
                    Install MWs & MPs
                    Conduct Initial Site Assessment/Data Collection
                                  • •;	-' '4 --i  ,.' ,!''.;"i'i."' v«r";
                    Data Analysis, Modeling, Intrinsic Remediation
                    Verification
                    TOTAL
                    ANNUAL SITE MONITORING/DATA ANALYSIS
                    COSTS
$4,000
$io,ood
Wl; f f . " 'f ' ['" i1 '••
$15,606
Lump Sum
600LF
Lump Sum
Lump Sum
   $4,000
  $18,000
  $10,000
 " $15,000
  $47,000
Annual Monitoring and Sample Analysis
Annual Data Reduction and Model Verification
TOTAL
$500/wetl
$5,000
1
• . rf • il
20 wells
Lump Sum

$10,000
$5,000
$15,000
                    Typical costs for site where depth to groundwater table = 9 m (30 ft), 20 monitoring wells and/or groundwater
                    monitoring points arelnstalled, and annual groundwater monitoring Is required. These values do not include source
                    removal/treatment costs. Refer to cost data for specific active source treatment technologies for these source removal
                    cost data.
                            •   confirmation of biologically-mediated reactions in the contami-
                               nant plume from chemical indicators of groundwater quality;
                                                         :         ,|
                            •   detection of biologically-generated chemical intermediates;
                            •   relative loss of degradable versus non-degradable contaminants
                               within the plume; and
                            •   contaminant fate and transport modeling.
                     '    "        ••  ••' :•   ,• ••••',  :,i;,:! ........ •"."'' I;*  •li:itorT'r':<  /'•   •]«>•*.•.:•;,]•&• : '  '
                      As previously indicated, ongoing monitoring and updating of the site
                   conceptual model is an integral part of the intrinsic remediation management
                   approach and should be used to validate and/or update contaminant fate and
                   transport predictions throughout the life of the plume.
                   5.3.14  Permitting Requirements
                                      ...  ,.•;•.'.-.,  ;• ,• :; ;;• , ...... • •; ..... :j ...  ...... .. .....           ,._..]. ,  ,,,  ,
                      Permitting requirements for intrinsic remediation systems are generally
                   minimal unless active source removal is part of the overall management
                   approach at a site. For the site assessment and ongoing monitoring phases of
                                                       5.74
-------
                                                           Chapter 5


the intrinsic remediation plume management approach, well construction
permits will be required in some states. Permits may also be required for
transport and disposal of contaminated drill cuttings from monitoring well
installation and contaminated groundwater generated during monitoring well
and shallow monitoring point purging prior to sampling. Depending upon
local requirements, the time and cost of managing these sampling residues
can be significant.  In addition, as part of the permitting process for most
remediation techniques, public notification and! public hearings may be re-
quired to justify selection  and implementation of intrinsic remediation for
contaminant plume management.

5.3.15 Design Checklist
   The previous sections provide detailed logic: regarding the application of
intrinsic remediation under site-specific conditions. The following is the
sequence of activities in the assessment and implementation of intrinsic
remediation for contaminant plume management.
        •  Determine steady-state plume conditions — evaluate plume
           centerline concentrations or estimate plume mass (CoM);
             •  If the plume is stable or shrinking, intrinsic remediation is
                viable.
             •  If the  plume is growing, continue to monitor unless
                source removal  or remediation is warranted due to re-
                ceptor impact.
        •  Estimate contaminant degradation rates — evaluate dissolved
           plume mass changes over time;
             •  If the plume mass is declining over time, a contracting
                plume is indicated.  Estimate degradation rate from mass
                loss over time.
             •  If the plume mass is steady-state over time, a continuous
                plume is indicated.
                  •     If groundwater flow data are available, calibrate
                        fate and transport model to field data.
                  •     If flow data are not available, use plume centerline
                        concentration data.
         •  Estimate the mass  of contaminant in the source area;
         •  Estimate the source lifetime based on degradation rates;
                                 5.75
-------
 Groundwater Treatment Systems
              •  If the plume is a pulsed source, base lifetime on dissolved
                mass degradation rate.
              •  If the plume is a continuous source, base lifetime on source
                mass above and below groundwater table in source area and
                in plume.
        •  Predict long-term behavior of plume;
                          	     	' 	..' "  ". ;'] 	      '           ,.''!,'
              •  If the plume is a pulsed source, long-term behavior is con-
                trolled by plume degradation rate estimated from dissolved
                plume mass loss over time
              •  If the plume is a continuous source, then:
                  •  Evaluate fate and transport with no source removal; and
                  •  Evaluate fate and transport  with various levels of
                           , ,, -   *  I,,, 'if .  • ," I  I ,,,„",', ,i,,|,'!' ,     .1,     |  ,'"',,'
                     source removal.
        •  Determine applicability of intrinsic remediation; and
              •  If a receptor will be impacted during the life of the plume,
                apply active remediation.
                                   11111     '  j                    '
              •  If the plume  lifetime using intrinsic remediation is not ac-
                ceptable, apply source removal.
              •  If the pool of TEAs is insufficient to assimilate mass in the
                source area over the predicted life of the source  area, apply
 '   "     ''  '	   	    '  	'' " '' *'	      1        •        '    	
                active remediation.
        •  Develop long-term site monitoring plan including:
             •  Compliance monitoring for regulatory purposes,
                               1	'•"'•    '"  1 "
             •  Intrinsic remediation process verification monitoring,
             •  Update conceptual model of site,  source lifetime predic-
                tions, contaminant degradation rates, and
             •  Update assessment of intrinsic remediation at the field site.

5.3.16 Implementation
   Implementation of the intrinsic remediation plume management approach
primarily involves the installation of groundwater and soil gas sampling
points and the collection and analysis of soil and groundwater samples from
a contaminated site. Wells and monitoring points should be installed by a
qualified driller with experience in environmental drilling, soil sampling
procedures, and OSHA 40-hour safety certification. All appropriate local
                                 5.76
-------
                                                          Chapter 5
and regional regulatory agencies should be notified. Refer to the appropriate
sections of this monograph that discuss active treatment methods should
such methods be necessary.

5.3.17  Start-up Procedures
  There are no specific start-up procedures for an intrinsic remediation
plume management approach as it is focused on site assessment, plume
delineation, plume modeling, and long-term monitoring. Procedures
appropriate for startup during the site assessment process include utility
right-of-way clearance, monitoring well and shallow groundwater sam-
pling point position identification, drill-rig mobilization, drilling, and
soil and groundwater sample collection. There are no active system
start-up procedures required unless source removal or treatment is
implemented as part of the intrinsic remediation management approach.
Refer to the appropriate sections of this monograph  that discuss active
treatment methods should such methods be necessary.

5.3.18  Performance Evaluation
  Performance evaluation of the intrinsic remediation plume management
approach was identified in the assessment protocol presented in Section
5.3.2. Performance of intrinsic remediation is based on the ability of natural
processes taking place within the contaminated aquifer to attenuate and de-
stroy the contaminants of concern before they migrate to downgradient re-
ceptors. Methods for verification of contaminant attenuation, which are
detailed in previous sections, include the evaluation of plume steady-state
conditions, determination of contaminant degradation rates, and quantifica-
tion of the long-term assimilative capacity potential of the contaminated
aquifer.  An integral component of this methodology is the long-term moni-
toring of the contaminant plume to up-date the conceptual model for the site
and verify contaminant degradation rates and assimilative capacity.

5.3.18.1  Operation Practices                                :
   Operations primarily involve routine soil gas and groundwater sampling
throughout the contaminated site and data analysis, interpretation, and mod-
eling for verification of contaminant attenuation.  Standard protocols for soil,
soil gas, and groundwater sampling are summarized by US EPA (1986a,
1986b), while methods for laboratory analyses required for documentation
of intrinsic remediation are available from a number of sources (APHA
1989; Nelson and Sommers 1982; Olsen and Sommers 1982; US EPA
                                5.77
-------
 Groundwater Treatment Systems
 1986c, 1986d, 1989c). Refer to the detailed protocol prepared for the U.S.
 Air Force by Wiedemeier et al. (1994) for a comprehensive discussion of
 sampling and analysis methods appropriate for evaluation of intrinsic
 remediation processes at a site.

 5.3.18.2 Operations Monitoring
   Monitoring of intrinsic remediation involves the quantification of the
 degradation of the contaminants of concern, magnitude and distribution of
 chemical species serving as electron acceptors during microbial degradation
 of these contaminants, and general aquifer properties indicative of ground-
 water conditions relevant to microbial degradation processes. Table 5.15
 summarizes these analytes along with methods used and the purpose for
 their determination. Refer to the detailed protocol by Wiedemeier et al.
 (1994) for a discussion of analytical methods appropriate for these analytes.
 5.4  Aboveground Reactors
5.4.1  Principles of Operation
   Biological treatment of domestic wastewater using naturally-occurring
microorganisms for removal of contaminants has been practiced for years.
Biological treatment using aerobic biodegradation has historically been the
primary treatment process for domestic wastewater.  Over the years, the
process has been improved and enhanced to Improve efficiency and perfor-
mance. Several variations of the basic biological treatment process are in
widespread use for treatment of domestic wastewater throughout the world,
and are described in detail in numerous textbooks and design manuals writ-
ten on the subject. The reader is referred to the first monograph in this series
(Ward et al. 1995) for specific design and application details for biological
reactors.  A summary of key principles and issues related to the biological
treatment of contaminated groundwater and soiids in ex-situ reactors is pro-
vided below.
   The basic biological treatment process involves the use of
naturally-occurring aerobic, facultative or anaerobic microorganisms to me-
tabolize organic material. The majority of organisms found in biological
treatment systems are of the facultative type, that is, they can function in
either an aerobic or anaerobic environment. Aerobic degradation uses oxy-
gen as an electron acceptor while anaerobic processes use inorganic carbon
                                                             )
                                5.78
-------
                        Chapter 5

Table 5.15
Soil, Soil Gas and Groundwater Sample Analyses that can
be used to Quantify Intrinsic Remediation Processes
Sample Type
Groundwater









SoU








Soil Gas






Measurement
02
fta+.Mn2+
pH
Redox
Aromatic HCs*
Total HCs
Boiling point splits
Chlorinated HCs
Dissolved Gases (CH4,
ethylene, vinyl chloride)
COD
Organic carbon
pH
Kjeldahl-N
Extractable P
Texture
Aromatic HCs
Total HCs
Boiling point splits
Chlorinated HCs
°2
OD2
CH4
Aromatic HCs
Total HCs
Boiling point splits
Chlorinated HCs
Method
Field DO Meter
Colorimetry
Field Glass Electrode
Field ORP Electrode
Lab GC**
Lab GC
LabGC
LabGC
LabGC
Lab Acid Chromffite
Lab Acid Chromate
Lab Glass Electrode
Digestion
Extraction
Physical
Lab GC**
Lab GC
LabGC
LabGC
Field O2 Meter
Field CO2 Meter
LabGC
LabGC
Field/Lab GC
LabGC
LabGC
Purpose
Electron acceptor
Biological product from
electron acceptor
Env. conditions
Redox potential
Substrate
Substrate
Substrate
Substrate/co-substrate
biological products &
intermediates
Biological products &
intermediates
Substrate/electron donor
Sorption
Env. conditions
Nutrient ;
Nutrient
Env. conditions, flow
Substrate
Substrate
Substrate
Substrate/co-substrate
biological products &
intermediates
Electron acceptor
Mineralization product
electron acceptor
Biological product
Substrate
Substrate
Substrate
Substrate/co-substrate
biological products &
intermediates
*HCs = hydrocarbons
**GC = gas chromatograph
5.79
-------
 Groundwater Treatment Systems
 as an energy source. The anaerobic process is typically applied to wastewa-
 ters of high organic strength (>1,000 mg/L BOD or COD) or for degradation
 of specific chlorinated organics (e.g., trichloroethene). However, for most
 ground-water treatment applications, aerobic biological treatment is appli-
 cable due to the low organic loadings typically encountered, even in what is
 considered highly contaminated groundwater systems.
   Aerobic biological treatment processes can operate as either suspended
 growth or fixed-film systems.  Suspended growth systems, such as the acti-
 vated sludge process,  consist of free-floating microorganisms suspended in
 the liquid or solid slurry suspension being treated. The waste/microorganism
 mix (mixed liquor) is  oxygenated in a tank (aeration tank) using mechanical
 or diffused aeration by the addition of either ambient air or pure oxygen.
 The microorganisms or the treated solid suspension are then separated from
 the treated liquid using sedimentation, filtration, or flotation processes. A
 portion of the separated solids are recycled back to the aeration tank in the
 liquid wastewater system to ensure that adequate microbial populations exist
 in the bioreactor to treat more waste. The portion of the microorganisms not
 recycled to the aeration tank are removed (wasted) from the system to  ac-
 count for organism growth and to maintain the desired microorganism con-
 centrations within the biological reactor. Suspended growth systems are
 typically applicable to wastewaters with BOD concentrations of 100 to 1,000
 mg/L, and are the only systems applicable for the treatment of soil slurries.
 At low BOD concentrations in liquid-phase bioreactors, insufficient growth
 rates due to a low food supply make it difficult to maintain a sufficient mi-
 croorganism population to overcome losses in the separation process. As a
 result, washout of the  microorganisms can occur, leading to process failure.
  Fixed film treatment systems, such as submerged biological filters,
 biotowers, rotating biological contactors (RBCs), biological activated carbon
 (BAG) and fluidized-bed reactors (FBRs) are generally more applicable for
 treatment of dilute groundwaters than are suspended growth systems. In a
 fixed-film process, the microorganisms are attached to a media and the sub-
 strate is passed over the attached biomass. Oxygen is  supplied in the RBC
 and biotower processes by diffusion from the atmosphere through the
 air-liquid interface on  the reactor media in contact with the atmosphere.
 Oxygen then diffuses to the microorganisms in the attached biofilm while
 metabolic endproducts (i.e., CO2, etc.) diffuse out of the biolayer, through
 the liquid-air interface and into the atmosphere. Oxygen is transferred to the
 wastewater to be treated in the FBR process by air or more typically, pure
 oxygen injection at the bottom of the reactor. This injected gas stream
 serves both as a source of electron acceptor, and as a means of mixing and
fluidizing the bed within the reactor. Submerged biological filters can  also

                                 5.80
-------
                                                           Chapter 5
be aerated using diffused aeration.  BAG units are typically aerated by satu-
rating the influent with oxygen prior to its entry into the reactor.
   Excess biomass is removed from fixed-film systems by the process of
sloughing. As the biomass grows on the media, organisms attached to the
media become starved as substrate diffusing from the surface of the biofilm
becomes depleted before reaching them. Without electron donors these
organisms die, the biofilm looses its attachment to the media surface, and the
entire biofilm sloughs off of the media due to turbulence.  This sloughed
biomass is subsequently removed from the reactor effluent by sedimentation,
filtration, or flotation.  BAG units are periodically backwashed to remove
excess biological growth as evidenced by excessive pressure drop across the
carbon bed. In the case of the FBR process, a mechanical system is used to
shear the biomass from the media when the density of the biomass/media
particles decrease sufficiently due to biological growth to elevate the bed
height above a predetermined point.
   All biological treatment technologies are not applicable to the treatment
of extracted groundwater. Generally, extracted groundwater is characterized
as having relatively low biodegradable organic concentrations (e.g., BOD or
COD) as compared to domestic or industrial wastewaters. Groundwater
characteristics from several Superfund sites in the northeast are provided in
Table 5.16 . These data demonstrate the variability in constituents and con-
centrations from different sites, thus indicating the importance of conducting
treatability studies to evaluate the performance of biological treatment with a
specific groundwater matrix. The importance of acclimation of the biomass
to a specific groundwater to evaluate performance is evident from the range
of influent characteristics that can be encountered. Of note are the compara-
tively low BOD and COD concentrations. Consequently, technologies de-
veloped for treatment of high strength wastewaters (e.g., pure oxygen acti-
vated sludge and most  anaerobic processes) are not generally appropriate for
groundwater treatment applications.

5.4.2 Process Design Principles

5.4.2.1 Technology  Application
   Figure 5.20 shows the range of organic strength at which potentially ap-
plicable aboveground biological treatment processes can typically operate.
In general, suspended growth processes, such as activated sludge and Pow-
dered Activated Carbon Treatment (PACT®), can effectively and efficiently
treat contaminated groundwater streams only at a higher range of organic
strength than fixed growth processes, such as submerged biological filters,

                                 5.81    I
-------
Groundwater Treatment Systems
; •• . ,( : • •••..;•;;.:•:;;' ;f;i ;.
Table 8.16
Example Grouhdwater Influent
Site A SiteB
Parameter Influent Influent
• ' • • 	 i i
West Chemistry (mg/L)
BOD (total) 13
BOD (soluble)
COD (total) 350
BOD (soluble)
Chloride 4
Total Cyanide
Amenable (Free) Cyanide)
TKN . , , .;; 	 ] ; . ..
Oil & Grease
TPH
pH(s.u.)
',, " ,!,,!',,,
Phenolics
TDS 445
TCC .
TSS
, , ,', ' • : i ' 1 ' ,
Hardness as CaCO3
Sulfate ' ' '"' "' ' M' ' '
Sulfide
Volatile Organics (ng/L)
Methyl methacrylate 57,184
	 " " ' | ' '
Acetone 302 557
Benzene 344 9
2-Butanone • 524
Carbon disulfide 4J 2
., " ' ,; .,: ,;,"'." " 1 . ."
Chlorobenzene
Chloroform 1
Chloroethane
Chloromelhane
; ' • 	 ' •• ' • 	 ' ' ' 1
1,1-Dichloroethane 85 8
i,2-Dichloroethane 42 1,652
1,1-Dichloroethene 2
BMDL = Below Method Detection Limits
Blanks indicate the analyte was not detected at that site.
••- -•• " : 	 -I"'" -'
Data
Site C Site D
Influent Influent
16 " '"51 	
16
	 iST ••'•'' l""" 253'"" 	 '" 	
179 	 !"
BMDL
BMDL
49
BMDL
BMDL
"6.85"' ' ' : "": ":"
0.10
954 4,688
' '" ' M' '" ''','"' H
84 ' ' "l55" 	
"" :'123"""'":: ": '''' 	 " 	 ;:"' 	 ' : '
1 ' i '• '•• • 	 •• •. 	 • ( 	 	 • „ 	 ,- -
BMDL

450 40
190 107
35
	 !! "I1,,"!.,,1 ,'!: ,s, J 	 ',:,"" 	 !', 1 "i,1 	 ,-!
4
250 1,736

1 :: '" 	
7

57

'" ,'••'<: ' ' , 	 „, ] ,,' 	 [ I,'.'
                             5.82
-------
                        Chapter 5

Table 5.16 (cont.)
Example Groundwater Influent Data
Site A Site B Site C
Parameter Influent Influent Influent
1,2-Dichloroethene (total)
cis-l,2-Dichloroethene 193 141
trans- 1,2-Dichloroethene 43
Total 1,2-Dichloroethene 141.
1,2-Dichloropropane 118
trans-l,3-Dichloropropene
1,4-Dioxane 126
Bthylbenzene 550 104 98
4-Methyl-2-Pentanone 111 210
Methylene chloride 48 4
Styrene 43
Tetrachloroethylene
Toluene 105 8411 480
1 , 1 , 1 -Trichloroethane
Trichloroethene 48
Vinyl chloride » 2
Xylene (total) 453 380
Acid Extractable Compounds (ng/L)
Benzoic Acid
2-Chlorophenol
2,4-Dichlorophenol
2,4-Dimethylphenol 48 11
2-Methylphenol 19
4-Methylphenol 18 16
Pentachlorophenol
Phenol 7* 10 4-*
2,4,6-Trichlorophenol
Base Neutral Organics (ng/L)
Acenaphthene 0.4
Aniline 20°
Benzo(b)fluoranthene
SiteD
Influent
7




3


9



6

3

62

10
14
12
10
5
19
25
516
19

10

10
BMDL = Below Method Detection Limits
Blanks indicate the analyte was not detected at that site.
5.83
-------
Groundwater Treatment Systems
". - • .. , " ." , . ' ;;•• .• . • : j • ' • • ' ' • ' .,' i 	 ; " ;
Table 5. 16 (contO
Example Groundwater Influent Data
; • 	 '.; 	 -... > r ,, ., ..
Site A Site B Site C
Parameter Influent Influent Influent
Benzo(k)fluoranthene
Benzyl alcohol 2
Bis (2-Chloroethyl)ether 3.9
Bis (2-Chloroethoxy)methane 2.2
Bis (2-Ethylhexyl) phthalate 22 U
4-Chloroaniline 42
2-Chloronaphthalene
... ... . . . . ,| | . . . . .
Chrysene
Dibenzofuran
Diethyl phthalate 28 1.1
Dimethyl phthalate 2
Di-n-butyl phthalate 19 0.6
1,2-Dichlorobenzene 7.6
1,3-Dichlorobenzene 12
1.4-DichIorobenzene 43 26
3,3-Dichlorobenzidine
Fluoranthene
Fluorene
Isophorone 1.0
2-MethyI naphthalene 20 13
Naphthalene 29 2)
Nitrobenzene
N-Nitrosodiphenylamine 2.0
Phenanthrene
Pyrene
1 ,2,4-Trichlorobenzene
Metals (ng/L)
Aluminum (total) 253
Aluminum (soluble) 66
Antimony
Arsenic (total) 23 3 37
Arsenic (soluble) 25 11
Cadmium (total) 6 BMDL
i 	
	 i. 1 . r,""1 . "!
SiteD
Influent
X)
13" ; ' 	
TO

K)
2059
K)
5' " "
10



1,543
79
192

K)
5 '

35'" ' "
543

10
.< , , n,' ,i| 	 „
10
703
;::



377

4
, j , ... , i „
BMDL = Below Method Detection Limits
Blanks Indicate the analyte was not detected at that sita.
.,],'•'' : '! '; . '
                             5.84
-------
                                                              Chapters
                            Table 5.16 (cont.)
                   Example Groundwater Influent Data
Parameter
Cadmium (soluble)
Chromium (total)
Chromium (soluble)
Copper (total)
Copper (soluble)
Iron (total)
Iron (soluble)
Lead (total)
Lead (soluble)
Manganese (total)
Manganese (soluble)
Mercury (total)
Mercury, (soluble)
Nickel (total)
Nickel (soluble)
Silver (total)
Silver (soluble)
Zinc (total)
Zinc (soluble)
Site A Site El Site C
Influent Influent Influent
2 BMDL
18 BMDL
BMDL
17 13
8.7
104,426 46,858 318
101,530 239
9 282 33
5 BMDL
76 167
74
1 BMDL
BMDL
16
19
BMDL
BMDL
13,499 44
55
SiteD
Influent
!
108

396

52,200
1
176



55

160

2


-
BMDL = Below Method Detection Limits
Blanks indicate the analyte was not detected at that site.
biological activated carbon, rotating biological contactors (RBCs) and fluid-
ized-bed reactors (FBRs). Slurry reactors (high solids content suspended
growth reactors) are applicable for highly concentrated solids streams and
highly contaminated soils that are slurried prior to treatment.
   Slurry reactors are liquid/solid contact biological reactors that are analo-
gous to conventional biological suspended growth reactors, e.g., activated
sludge systems. Slurry reactors are used to treat contaminated solid matri-
ces, such as soils or sediments. They are employed to mitigate the environ-
mental factors typically encountered with treating relatively recalcitrant
constituents in soil or sediments.  By suspending the soil or sediments in an

                                  5.85
-------
                                                                                                                                     a
                                                                                                                                     I
                                        Figure 5.20
                            Overview of Biological Processes
                   Aerobic
         Fixed-Film or
       Attached Growth
Rotating Biological Contractor (RBC)
  Submerged Aerobic Filter (SAP)
   Fluidized-Bed Reactor (FBR)
 Biological Activated Carbon (BAC)
                                        BIOLOGICAL PROCESSES
       Suspended
        Growth

  Activated Sludge (AS)
   Powdered Activated
Carbon Treatment (PACT®)
                                                                               Anaerobic (AN)
Attached
 Growth
                                                                                               Suspended
                                                                                                Growth
                                                                                                                            (D
BAC
                                  SAP
                                  RBC
                   FBR
                    AS
                   PACT
                             AN
 10
                                   50
                    100

         Concentration of Organics, mg/1
                            1000
-------
                                                            Chapter 5
aqueous system the availability of carbon sources, inorganic nutrients, and an
electron acceptor (typically oxygen) are greatly improved by maximizing mass
transfer rates and contact between the contaminants and the microorganisms.
   Treatment of soils with a slurry reactor is nearly always considerably
more expensive than treatment using a soil pile or by land farming. The use
of this technology is generally limited to treatment of more recalcitrant com-
pounds or highly contaminated soils or sediments. Typical wastes treated in
this manner are oil refinery wastes, principally sludges from storage and
treatment lagoons, and wood preserving wastes, such as impoundment slud-
ges and the surrounding soils contaminated with creosote and/or pentachlo-
rophenol (PCP).
   Although aerobic reactors are most common, some systems are designed
to operate under anaerobic conditions or to cycle between anaerobic and
aerobic conditions. Slurry reactors can be operated in single or sequencing
batch modes, or in either continuous or semi-continuous modes.  Slurry
reactors have been constructed in lined lagoons, unlined lagoons (for treat-
ment of lagoon solids),  or in constructed reactors.
   In all systems, the two main design criteria are mixing and aeration.
While mixing can, in some cases, be provided by aeration, it is typically
achieved through mechanical means, especially for treatment of soils, which
require more energy to maintain in suspension than do sediments.
   Most of the constituents found in contaminated groundwaters can be re-
moved using aerobic biological treatment. Any of several removal mecha-
nisms including stripping, sorption, and biodegradation may occur alone or
concurrently and contribute to overall contaminant removal. The fate of
various organic compounds in aerobic biological treatment processes is
shown in Table 5.17. The removal mechanisms are described below.

5.4.2.2 Stripping
   Stripping can be a significant removal mechanism in aerobic treatment
processes because the aeration process used to provide oxygen for aerobic
degradation also strips volatile compounds. Depending on the particular
constituent, both stripping and biodegradation may occur simultaneously.
The proportion of removal that occurs due to either of these mechanisms
depends on the constituent in question and,  to a lesser extent, the type of
treatment process being utilized.  For example, vinyl chloride is extremely
volatile and will generally strip before significant biodegradation occurs.
However, a less volatile compound, such as acetone, will generally biode-
grade readily with only minor removal due to stripping.
                                 5.87
-------






























Fate










of Organic


Overall Percent Removal


Pollutant

.0, 2,4-D
,§§ ,f Acenaphthylene
Acetaldebyde
Acetone
- Acrolein
Acrylamid
Acrylic Acid
Acrylonitrile
Aldrin
Aniline
Anthracene
Antimony
: Arsenic
Barium


Ace. %

90
95
95
95
95
90
90
90
90
95
95
60
50
90


Med.

60
90
95
50
95
62
85
75
90
85
90
60
50
90
Unacc.

% Low%

50
90
95
30
95
50
80
7)
90
80
90
60
50
90







Table
Constituents









5.17
in Activated


Percent Volatilized to Air


Ace. % Med

0 0
19 54
0 5
0 3
0 5
0 0
0 0
0 4
0 0
0 0
0 0
0 0
0 0
0 0
Unacc.

% Low%

0
S
5
2
5.
0
0
4
0
0
0
0
0
0


^ =
























=






_t - ,

= , s



Sludge Systems




Percent Partitioned to


Sludge




Percent Biodegraded
Unacc.

Ace. %

7
9
10
W
10
9
9
9
33
10
52
60
50
90

Med.%

5
8
10
5
10
6
9
8
33
9
50
60
50
90

Low%

4
8
10
3
10
5
8
7
33
8
50
60
50
90

Acc.%

83
67
85
85
85
81
81
81
57
86
43
0
0
0
Unacc.

0
S
c
.3
a
€
o^
CD
~ CD
a
"--•-3
(D
f
(D
Med. % Low % 3

55
28
81
43
81
55
77
64
57
77
41
0
0
0
co
46
28
81
26
81 : ,
45
72
60
57
72
41
0 :
0 - i
0 - '
-------
O
Benzal Chloride
Benzene
Benzotrichloride
Benzyl Chloride
Bis-2-Chloroethyl Ether
Bis-2-Ethylhexyl Phthalate
Bromomethane
Butyl Benzyl Phthalate
Cadmium
Captan
Carbon Disfulide
Carbon Tetrachloride
Chlordane
Chlorobenzene
90 55
95 90
90 45
90 90
90 50
90 90
95 95
95 90
27 27
90 50
95 85
90 85
90 90
90 90
Ace = Acclimated sludge with constant loading Unacc =






50
90
40
90
30
90
95
90
27
30
80
SO
90
90
0
24
18
23
0
0
86
0
0
0
T6
72
9
27
17
72
14
45
3
0
90
0
0
0
77
77
9
45
15
72
12
45
2
0
90
0
0
0
72
72
9
45
7
2
7
7
9
65
0
43
27
7
1
12
33
14
4
2
4
7
5
65
0
41
27
4
1
9
33
14
4
2
3
7
3
65
0
41
27
2
1
8
33
M
83
69
65
60
81
24
10
52
0
83
18
6
48
50
34
16
28
38
43
24
5
50
0
45
8
0
48
32
31
16
25
68
26
at
5
50
0
23
7
0
48
32
Unacclimated sludge with transient loads



























O
Q
"S-
01
-------







Pollutant
Chlorobenzilate
•£ Chloroethane
= , Chloroform
Chloromethane
2-Chlorophenol
	 : Chromium
Cresols
Cumene
Cyanide
Cyclohexane
Di-N-Butyl Phthalate
Di-N-Octyl Phthalate
Dibromomethane
i 1,2-Dichlorobenzene






Overall

Acc.%
90
95
90
95
95
70
95
95
60
95
90
90
85
90




Fate






Table


5.17 (cont.)
of Organic Constituents in Activated

Percent Removal

Med.
63
90
80
90
65
70
50
95
60
95
90
90
80
87

Unacc.
% Low%
50
90
80
90
60
70
40
95
63
95
90
90
80
85



Percent Volatilized to Air

Unacc.















3
c
|
Sludge Systems Z,




Percent Partitioned to Sludge

Ace. % Med. % Low % Ace. %
9
76
63
86
0
0
0
38
0
10
0
0
43
45

6 5
81 81
72 72
35 85
0 0
0 0
0 0
57 57
3 3
86 86
0 0
0 0
64 64
78 77

7
1
2
1
8
70 .
8
4
57
4
20
7
13
32

Unacc.

Med. % Low %
5
I
2
1
5
70
4
4
57
4
20
7
12
9

4
1
2
1
5
70
3
4
57
4
20
7
12
9


Ace.
74
18
25
0
87
0
87
53
3
82
70
83
30
14


Percent Biodegraded
Unacc.
(D
1
3-
&
£1
% Med. % Low % 
-------

1,3-Dichlorobenzcne . 90
1,4-DicMeiebenzeHe 90
1,2-DichloroethaHe 90
1,1-DicUoroethylene 95
2,4-Dichlorophenol 95
1,2-Dichloropropane 90
Dichlorvos 90
Dicofol 90
Diethyl Phthalate 90
3,3-Dimethoxy Benzidine 9)
2,4-Dimethyl Phenol 95
Dimethyl Phthalate 95
2,4-Dinitrophenol 90
1,4-Dioxane 90
Ace = Acclimated sludge with constant loading



87
87
50
90
as
73
50
90
75
30
85
65
75
50
Unacc



85
85
30
90
50
TO
30
93
70
20
83
60
70
43
45
45
45
76
0
45
0
45
0
0
0
0
0
0
78
78
45
81
0
63
0
45
0
0
0
0
0
0
77
77
27
81
0
63
0
45
0
0
0
0
0
0
3
23
5
0
8
0
9
8
1
8
8
0
9
9
3
9
3
0
4
0
5
8
1
3
7
0
8
5
3
9
2
0
4
0
3
8
1
2
6
0
7
4
42
23
41
19
87
45
81
37
»
72
87
95
81
81
6
0
3
9
51
7
48
37
74
27
78
65
63
45
6
0
2 .
9
46
7
27
37
69
18 .
74
60
63
36
= Unacclirnated sludge with transient toads



























g
f
Oi
-------
                                                                                                      t :  ME
j§









Pollutant
Epichiorohydrin
Ethyl Benzene
Ethylene Thiourea
Formaldehyde
Hexachloro 1,3-Butadiene
Hexachloroethane
Hydrazine
Lead
Maleic Hydrazide
Mercury
Methanol
Methoxychlor
Methyl Ethyl Ketone
Methyl Isobutyl Ketone







Overall

Ace. %
87
95
85
85
95
95
95
70
90
50
100
90
95
90






Fate



-







Table





5.17 (cont.)
of Organic Constituents in Activated
Percent Removal

Med.
59
90
67
85
90
9°
85
70
75
SO
95
90
50
50
Unacc.
% Low%
25
90
60
80
90
90
80
70
70
50
95
90
30
30
Percent Volatilized to Air

Unacc.






























Sludge Systems
Percent

Ace. % Med. % Low % Ace. %
0
24
0
0
0
0
0
0
0
0
1
54
0
0
0 0
72 72
0 0
4 4
5 5
5 5
4 4
0 0
0 0
3 3
5 5
54 54
3 2
0 0
9
6
9
9
9
9
10
70
9
48
10
8
10
9
Partitioned to Sludge
Unacc.

Med. % Low %
6
5
7
9
8
8
9
70
8
48
ro
8
5
5
3
5
6
8
8
8
8
70
7
48
10
8
3
3

Ace.
78
66
77
76
86
86
85
0
81
2
90
28
85
81
Percent Biodegraded
Unacc.
s
c
D
Q.
1
-------
Meihylene Chloride 95
N-Butyl Alcohol 95
N-Nitrosodimethyl Amine 90
Naphthalene 95
Nickel 35
Nitrobenzene 90
2-Nitropropane 95
P-Benzoquinone 95
Parathion 0
PCB 92
Pentachlorophenol 95
Phenol 95
^ Phenylene Diamine 90
Phosgene 100
Ace = Acclimated sludge with constant loading



87
90
75
75
35
25
95
30
55
92
25
85
75
100
Unacc =



85
90
70
70
35
20
95
40
40
92
20
80
70
100
38
0
0
0
0
0
86
0
0
9
0
0
0
1
52
0
0
4
0
0
90
0
0
9
0
0
0
5
51
0
0
4
0
0
90
0
0
9
0
0
0
5
13
10
9
27
35
9
1
8
0
22
17
14
9
10
12
9
8
21
35
3
1
4
4
22
5
13
8
K>
12
9
7
20
35
2
1
3
3
22
4
12
7
10
44
86
81
68
0
81
9
87
0
61
78
81
81
90
23
81
63
50
o :
23
4
46
51
61
21
72
68
85
22
81
63
47
0
18
4
37
37
61
16
€8
63
85
Unacclimated sludge with transient loads



























O
1
$
cn
-------




:






Pollutant
Phthalic Anhydride
Pyridine
Selenium
Silver
Styrene
1,1,2,2-Tetrachloroethane
Tetrachloroethylene
Thionrea
Toluene
Toleune Diamine
Toxaphene
Trans-l,2-Dichloroethylene
Tribromomethane
1 ,2,4-Trichlorobenzene







Overall


Ace. %
90
15
50
SO
SO
90
90
90
90
90
95
90
65
85






Fate



,,,


of Organic
Percent Removal


Med.
90
15
50
90
90
25
85
75
90
75
90
80
35
85
Unacc.

% Low%
SO
10
50
SO
SO
20
80
70
SO
TO
90
80
30
85





Table





5.17 (cont.)
Constituents in Activated
Percent Volatilized to Air


Unacc.































Sludge Systems
Percent Partitioned to Sludge


Ace. % Med. % Low % Ace. %
G
0
0
0
23
36
45
0
23
0
57
63
36
43
0 0
1 1
0 0
0 0
72 72
15 12
68 64
0 0
72 72
0 0
72 72
72 72
21 18
51 51
9
2
50
90
14
4
3
9
25
9
4
27
5
8
Unacc.



Med. % Low %
9
2
50
SO
14
1
3
8
18
8
4
8
3
8
9
1
50
SO
14
1
2
7
18
7
4
8
2
8
Percent


Biodegraded
Unacc.

0
— t
O
c
Q.
(D
i
(D
3-
£
(D
Ace. % Med. % Low % 3
81
13
0
0
54
50
42
81
52
81
34
0
24
35
SI
13
0
0
5
9
14
68
0
63
14
0
11
26
81
9
0
0
5
7
M
63
0
63
M
0 '.-
10 :::
26
-------
1,1,1-Trichloroethane
1,1,2-TricMoroethane
Trichloroethylene
2,4,6-Trichlorophenol
1,1,2-TC 1,2,2-TF Ethane
Trifluraline
Vinyl Chloride
Xylenes
95
80
95
95
90
90
95
95
Ace = Acclimated sludge with constant loading
Source: US EPA 1996

90
25
87
55
85
90
95
87
Unacc =

85
20
85
50
80
90
95
85
16
40
67
0
63
0
86
21
81
2)
TO
0
68
0
90
TO
77
16
68
0
64
0
90
68
1
0
6
8
4
33
2
M
1
0
5
4
3
33
2
13
1
0
5
4
3
33
2
13
18
40
23
87
23
57
8
57
8 8
5 4
12 12'
51 46
14 13
57 57
3 3
4 4
Unacclimated sludge with transient loads









Oi
'o
Oi
.9.
Q
•a
Oi
-------
          Groundwater Treatment Systems
            Stripping is also a concern with many slurry reactor designs as most use
          aeration to provide all or part of the energy required for solids suspension.
          Where heavy PAHs or other relatively nori-volatile compounds constitute
          most of the contaminant mass, volatilization may not be a concern. Where
          solvents or low molecular weight hydrocarbons make up a substantial frac-
          tion of the organic loading, system design will normally include provisions
          for minimizing air emissions. This could include providing a reactor cover
          fitted with an air capture and treatment system. At the French Limited
          Superfund Site in Crosby, Texas, pure oxygen was introduced into a recycled
          liquid stream to provide sufficient oxygen to meet biological treatment re-
          quirements, but avoid the stripping action resulting from forced air systems.
            The FBR concept is different from other aerobic processes in that strip-
          ping of brganics is essentially eliminated due to" trie method of oxygen disso-
          lution employed. The FBR process uses a pure oxygen bubble contactor
          which provides complete oxygen dissolution without release of any bubbles
          or off-gas. As a result, stripping is minimized and occurs only at the
          air-liquid interface at the reactor surface through diffusion.
,'  	""                                        	I     ,    ,          	

          5.4.2.3 Sorption
            Although not a primary removal mechanism for most organic compounds,
          sorption on biological solids does occur, can contribute to organics removal,
          and may also affect waste sludge characteristics. As an example, pesticides
          (e.g., lindane) have been shown to be non-degradable with sorption provid-
          ing the primary removal mechanism in biological reactors. Heavy metals
          also tend to sorb onto biological solids.  Thus, concentrations of such com-
          pounds can accumulate in the waste biomass and can potentially produce a
          waste sludge that is classified as a hazardous waste. An evaluation of the
          potential for the accumulation of non-degradable, toxic constituents within
          the waste stream should be carefully made to make sure that complications
          related to handling and disposal of a hazardous solid waste stream  are not
          unanticipated.
            For slurry reactors, desorption rather  than sorption generally controls the
          applicability and success of solid-phase  treatment.  The rate of biodegrada-
         tion in many cases is limited by bioavailability, especially for heavier, less
          soluble compounds, such as five- and six-ring PAHs.  The longer organic
         contaminants have been in contact with solids, especially silts and clays with
         high organic carbon content, the less available the compounds are for bio-
         degradation, and the less successful biological treatment systems will be in
         yielding treated solids with low residual contaminant concentrations. Some
                                          5.96
-------
                                                            Chapter 5
slurry reactor designs incorporate the addition of surfactants into a slurry
preparation phase to improve contaminant desorption and bioavailability.

5.4.2.4 Biodegradation
   Biodegradation is the primary removal mechanism for most organic com-
pounds found hi contaminated solids and groundwaters. Most organic com-
pounds are biodegradable, although many  of the constituents typically found
at contaminated sites require an acclimation period before significant bio-
degradation occurs. For example, acclimation periods of 4 to 6 weeks have
been reported for ethylbenzene and benzedrine, respectively, before munici-
pal activated sludge achieved maximum removal rates for these compounds.
Acclimation is generally performed by gradually increasing the concentra-
tion of a constituent over time while decreasing the base food source.  This
procedure is necessary to prevent possible shock or toxic effects to the biom-
ass from the full-strength waste stream before complete acclimation of the
biomass to the waste stream has been accomplished.
   Addition of co-substrates, such as methanol or glucose must be added to
increase the biodegradation rate of some compounds, chlorinated, organics in
particular.  Studies have shown that the addition of methane, which promotes
the growth of methanotrophic organisms, can significantly improve the re-
moval of chloroform, trichloroethene, and other chlorinated organics through
the stimulation of methanemonooxygenase (MEMO) by methane oxidizing
organisms. MMO is an extracelluar enzyme which has non-specific activity
toward chlorinated solvents, resulting in their fortuitous oxidation under
specific operating conditions.

5.4.2.5 Treatability and Pilot Studies
   When first evaluating the potential feasibility of using ex-situ biological
treatment, it is necessary to understand the biodegradability of the contami-
nants present. The reported biodegradability of typical groundwater and soil
contaminants is provided in Table 5.17. Site characterization data should be
used to estimate the nature and projected concentrations of contaminants that
will exist in the extracted groundwater and excavated contaminated soil that
must be treated from a site.  If the waste composition, or the application of a
particular reactor configuration have not been validated based on docu-
 mented field-scale studies, treatability and pilot studies may be warranted.
 This is particularly true for slurry-phase reactors where only limited data
 currently exists defining  their performance in the treatment of highly con-
 taminated, complex waste constituent soil streams.
                                  5.97
-------

 Groundwater Treatment Systems
    The first step in determining if biological treatment is feasible for the
 specific contaminant and matrix is to perform a biotreatability evaluation
 using representative samples of the material.  Based on the data generated
 during the treatability studies, a pilot-scale study may be needed to further
 define operating conditions and parameters to optimize the cost-effectiveness
 of the proposed treatment scheme.  Pilot-studies can be critical to demon-
 strating the technical and economic feasibility of slurry reactors before field
 implementation.
   "iii,/  ;  "v • .'»	:	>;'f1 	  ' ":.•   •  <  ,|11:i|ii,»'(•,*''  15, •'!|,'''Jf *•'••. •  :-:'  ':  •  •    ' 4•:AW *' 	•
    After it is determined that the constituents, or the majority of the constitu-
           111   ', •"' 'T !•' '   ,  ' • i:1!1 i. "1 i.J!l r1 . ii' i'" :" '"I! „''". 'It'" ' i., ,i ,l"'l|'! 'i';ii|', ."K  '.'. "'i J ...  J       	
 ehts, are biodegradable, it is necessary to identify candidate technologies for
 further consideration.  The technologies to be selected for further evaluation
 will depend on the characteristics of the contaminated media, most notably,
 the overall organic content as measured by BOD or COD and whether the
 media to be treated is groundwater or soil.  Generally, for ground water treat-
 ment applications, an aerobic fixed-growth process is preferred due to the
 relatively low organic content of the expected wastewater stream.  Sus-
 pended growth reactors are required for treatment of contaminated solids in
 a slurry-phase application with  high organic content, up to 250,000 mg/kg
 concentrations.

 5.4.2.6  Oxygen and Nutrient Requirements
   •'."•.      ,    ";  •	,  •. '  -.' - .', ,«;' ',:••. ::!;L .i .;:;,' • . . ,  :..;•'  -,  	i-,  ::E ,:;,"
   Aerobic treatment processes  use oxygen as an electron acceptor and re-
 quire adequate nutrients, in the form of nitrogen and phosphorus, to promote
 biological growth. Oxygen can be supplied passively as in an RBC or trick-
 ling filter process or actively as  in an activated sludge or FBR process. The
 total amount of oxygen required depends on the organic content of the waste
 stream being treated and the endogenous respiration requirements of the
 biomass as follows:
                 ,: •  ". • ; ,; •  ">  '  i, '. .; • *.** * ' • ; ''if;!'!,;1'!./,'1 < '•••'•••  '"Hi, ''   \  *iT- v,\"v • \	.'
                02 = a'Sr + Xdb'Xvt                              (5.32)
  .  '   '	 •"   -•"•".    •'•  -•-•••••"'•iv—      •	,'	•:!:	•..:;-.
where:
                                             I                   "'
        O2  =   oxygen demand (ML'3);
        ai   =   oxygen consumption for biodegradation (M) O2/(M)
                 COD removal;
        Sr  =   COD removal (ML"3)f
        Xd   =   degradable fraction of biomass;
        b"   =   endogenous oxygen consumption (T1);
        Xd   =   volatile suspended solids concentration (ML'3); and
          t   =   hydraulic retention time (T).

                                5.98
-------
                                                            Chapters
   The oxygen supply requirements can be estimated using the above
formula or by empirical methods during treatability studies by measur-
ing oxygen uptake rates. In any event, sufficient oxygen must be sup-
plied to meet the maximum oxygen demand for biodegradatiori and en-
dogenous respiration.
   Nutrients generally must be added to groundwater and slurry-phase reactors
to meet biological growth requirements because contaminated media are gener-
ally deficient in nitrogen and phosphorus. The nutrients may be added as indi-
vidual compounds using aqueous ammonia, anhydrous ammonia, or ammo-
nium salts to supply nitrogen and phosphoric acid to supply phosphorus.  The
use of individual compound addition is advantageous in that a specific addition
rate for each nutrient can be controlled independently, thus minimizing costs.
Alternatively, diammonium phosphate can be used to supply both the nitrogen
and phosphorus in one step. Nitrogen and phosphorus requirements can be
estimated based on a COD:N:P ratio of 100:5:1. The actual nutrient require-
ments should be determined during treatability studies and during actual opera-
tion by monitoring effluent nutrient levels.  Nutrient addition should be adjusted
to maintain effluent residual concentrations of nitrogen and phosphorus of ap-
proximately 0.5 and 0.1 mg/L, respectively.
   Oxygen is typically provided to slurry reactors using submerged or float-
ing aerators, by compressors and spargers, or by saturation of a recycled
water stream with oxygen, which is then introduced into the reactor. Oxy-
gen transfer and fugitive emissions from gas-sparged slurry systems can
significantly limiteffective aerobic slurry bioremediation. Volumetric gas
flows can be reduced by using pure oxygen rather than air to reduce volatile
emissions associated with gas introduction.  Evaporative cooling, foam for-
mation, and energy costs attributable to mechanical aeration can also be
reduced through the use of pure oxygen.  More efficient mixing technologies
with higher oxygen transfer efficiencies are required to offset the increased
cost of pure oxygen addition (Storms 1993).
   Nutrient addition is typically required as is pH control. Surfactants, dis-
persants, and cometabolites are sometimes added to improve substrate avail-
ability and to improve the physical characteristics of the solids.
   Because slurry reactors are typically run in a batch mode, the more
readily degraded compounds are removed relatively rapidly during the initial
period of treatment. The more recalcitrant compounds subsequently degrade
at a slower rate. Cell growth rates behave similarly, and as early formed
cells die, they lyse, releasing internal cellular material, and recycling  any
nutrients they contain. Because of this nutrient cycling, nutrient addition
rates during the initial phase of operation are relatively large and then taper

                                  5.99
-------
 Groundwater Treatment Systems
 off over time during treatment. Typical cell densities may be 109 cells/mL or
 1 g of carbon/]L  This equates to an initial nitrogen requirement of approxi-
 mately 100 mg/L.
                            ;•         '       i  •       •            : .......
 5.4.2.7 Reactor Temperature?
    Biological reaction rates increase with increasing temperature, following
 the Arrhenius relationship of an approximate doubling of rate constants for
 each 10°C temperature increase up to a maximum tolerable temperature of
 approximately 35°C. Heating full-scale reactors  is particularly important
 during cold season operation when biological degradation rates will fall.
 The decrease in treatment times achieved at higher reactor temperatures has
 obvious benefits, but requires additional capital and operating costs to insu-
 late reaction vessels and heat incoming waste flows. One additional draw-
 back to reactor heating is that elevated temperatures may also increase vola-
 tile emissions, requiring additional off-gas treatment that may not be needed
 at lower temperatures.
 ..... '      lf   "       ' •       •'•„•;,''  V    ! j,':.|  J *!• > , .   '. ,'«.    '. ,  •-. ....... /lilSt"1!"' I
   Reactor heating can be advantageous, however, as indicated by Wbodhull
 and Jerger (1994) in the slurry reactor treatment of soils containing approxi-
 mately 10,000 mg/kg of PAHs. Nine days were required to reach clean-up
 levels at 25 to 27°C versus only 6 days at 35 to 37"C.  Overall treatment
 costs were 10% less at the elevated temperatures, despite increased operating
 costs for reactor heating because of the increase in reaction rate and decrease
 in treatment time.
5.4.3 Process-Flow Diagrams
   Suspended growth processes that are applicable for treatment of contami-
nated groundwaters include the activated sludge process and several of its
variations. Activated sludge consists of an aeration tank, clarifier, and return
sludge pumps. The process can operate in a plug flow or complete mix con-
figuration. A typical plug flow activated sludge system is shown in Figure
5.4.2. Bio mass grows hi the aeration basin where it contacts the substrate
and removes organic constituents. Oxygen is supplied to the biomass in the
aeration basin by diffused or surface mechanical aeration.  The biomass is
separated from the treated substrate in a clarifier by gravity sedimentation.
Excess sludge is removed from the process to maintain the desired ratio of
biomass to substrate. The majority of the biomass separated in the clarifier
is returned to the aeration basin for reuse
                  1    '     ' ii1 " i •' ..... • „'" 'ill ', 'inl'ii1: ,"' „:   !, ;l ', l|'»"! , ''in, "iii   " ' • i"  ,  "    it  I j ',',![! . „' !
   Fixed-film processes that are applicable to groundwater treatment include
fluidized-bed reactors, biological activated carbon, rotating biological
                                 5.100
-------
                                                            Chapter 5
contactors, and submerged biological filters. Trickling filters and biotowers
have generally not been employed for groundwater treatment due to the poten-
tial for significant stripping of organics and associated air emissions concerns.
In fixed-film processes, the biomass is attached to a media of some type. Media
can be plastic in the case of submerged biofilter, plastic discs in the case of
RBCs, GAG in the case of BAG, and sand or GAC in the case of an FBR pro-
cess. The medium allows the biomass to be retained in the treatment process
without relying on solids separation processes (e.g., sedimentation) as does a
suspended growth process. The attached growth concept allows the system to
operate effectively even at low organic concentrations where insufficient new
biomass is produced to provide for flocculant settling conditions which are
necessary for effective operation of suspended growth systems. Figure 5.21
shows a schematic of the FBR process, while Fig;ure 5.22 shows a schematic of
a suspended growth system containing fixed-film media in the form of pow-
dered activated carbon, the (PACT®) process.
   A process  schematic of a commercial slurry-phase treatment system is
shown in Figure 5.23. As indicated in this figure, the slurry-phase treatment
process is similar in configuration and reactor layout to the PACT® process,
but differs from it significantly in that in the slurry reactor the media in sus-
pension is the contaminated soil to be treated, while in the PACT® process it
is the treatment media itself.

5.4.4 Process Modifications

5.4.4.1  Suspended Growth Processes
   Suspended growth processes applicable for treatment of contaminated
groundwaters consist of the activated sludge process and several of its varia-
tions. Suspended growth processes, such as aerated or facultative lagoons
are generally not applied due to concerns regarding air emissions, costs for
lined lagoon construction, and the inability to develop and sustain an accli-
mated biomass on a one-pass system where the sludge is not retained and
recycled in the process.
   Conventional Activated Sludge. The standard activated sludge process
may be applicable for treatment of comparatively high-strength organic con-
centrations, e.g., >500 mg/L as BOD, where air emissions are not of concern
and the contaminants are readily biodegradable;.  Due to the aeration process,
stripping may be a significant removal mechanism  for many organics using
this technology.  In addition, BOD concentrations less than  100 mg/L are
difficult to treat effectively in conventional activated sludge systems due to
difficulties with solids-liquid separation that occur  at the low aeration basin

                                 5.101
-------
                                                                    Figure 5.21
                                                   Fluidized-Bed Reactor Process Schematic
                                                                                                                                 0
                                                                                                                                 3
                                                                                                                                 •3
pi,
:o
N)
                              Separation   Straineis
                               Tank
Recovery
 Wells
                                                                                                             Biomass Control
                                                                                                                  $







1
< —







*-
1 F
rs


Kecycie

1
1
I
L_
Eductor
~tE/~] o
— 3 1 Fluidization
-r3J Flow
^=i Control
luidization Valve

r-&-



waier usyei


A

O2 Recycle
^
- >
/
L
t


t
>

:V
f
/L
"T ^
I



r\

Bubble
Trap
Rotometer





- S -
Bed

-
Level










AAA








Fluidized
Reactor
r*fn

* 	 1 	 s.
cm

                                                                         Modulating
                                                                            Valve
                                             Compressor

                                                   Oxygen Generation System
Nutrient
 Feed
                                                                                                                                                    C^

                                                                                                                                                    (D
                                                                                                                                                    I
                                                                                                                                               !
-------
                                                                          ?& 5.22
                                                                PACT® Flow Diagram
01

§
Wastewater
                                                                                                                 Filtration
                                                                                                                (Optional)
                                                                  To Regeneration x
                                                                  or Solids Disposal
                                                                                                                              *• Effluent
                                                                                                                                                         o
                                                                                                                                                         Q
                                                                                                                                                         en
-------
i
              Soil Screening/
            Waste Preparation
                                                                         Figure 5.23
                                                Full-Scale Slurry-Phase Biological Remediation Process
                                 Soil
                                 Water
                              Nutrients,
                               Sluny-
                             Conditipning
                              Chemicals
   Slurry
Preparation/
Soil Washing
                                                                 Slurry
                                                                    Air
Slurry-Phase
 Biological
  Reactors
                                                  Slurry
  Slurry
Dewatering
                                                              Make-up Water,
                                                               As Required
                                                                                        Water
                                                                                   Management lank
                                                                                                               y Discharge Water,
                                                                                                                   As Required
                                                                                                                                     -»• Oversized Solids


                                                                                                                                     -> Solids
                                                                                                                   0
                                                                                                                   s
                                                                                                                   Q.

                                                                                                                   i
                                                                                                                   t
                                                                                                                   a-
                                                                                                                   (D
-------
                                                            Chapter 5
biomass concentrations that are generated from low strength waste. At low
BOD concentrations, the biomass growth rate is low and only a dilute biom-
ass concentration can be maintained in the aeration basin. At low biomass
concentrations, flocculant settling does not occur and poor solids separation
results in biomass carryover in the secondary clarifier, resulting in an inef-
fective treatment system.
   The conventional activated sludge process can be configured to minimize
air emissions by including an off-gas recycle system. The aeration basin is
covered and the off-gas is collected, is recycled through the aeration blowers,
and diffuses into the aeration basin. A small percentage, typically 10 to 15
percent by volume, of make-up air is added to provide adequate oxygen and
to purge carbon dioxide. This approach effectively reduces off-gas emis-
sions by increasing the contact of the stripped organic constituents with the
liquid and biomass in the aeration tank where they can be biodegraded.
Table 5.18 summarizes results of off-gas analyses from an activated sludge
system equipped with off-gas recycle treating a contaminated groundwater
consisting of Site A influent from Table 5.16. Performance data for this
system is provided in Tables 5.19 and  5.20.
   Powdered Activated Carbon Treatment.  A variation of the activated
sludge process is the powdered activated carbon treatment (PACT®) process
(Figure 5.22)  in which powdered activated carbon (PAC) is added to the
aeration basin. This technology has been successfully applied to high
strength and complex wastewaters, such as landfill leachate.
   PACT® is a proprietary process offered by Zimpro that involves the addi-
tion of PAC to an activated sludge system. The PAC provides a medium for
the biomass to attach to while also adsorbing refractory or toxic constituents.
As a result, a higher level of effluent quality may be achieved in instances
where there are refractory organics. Other benefits include improved sludge
settleability, greater resistance to shock loads and potential for reduced efflu-
ent aquatic toxicity due to removal of refractory organics, and to some extent
heavy metals. In cases where nitrification may be required, e.g., with a land-
fill leachate, the PACT® process can provide conditions suitable for sustain-
ing nitrification where this would not be possible in a conventional activated
sludge system due to inhibitory or toxic compounds.
   PACT® can also reduce off-gas emissions by adsorbing volatile com-
pounds.  Table 5.21 presents off-gas emissions data from a PACT® system
treating the same groundwater as described in 5.19. A comparison of the
data indicates the PACT® process, with off-gas recycle, achieved comparable
air emissions control. Effluent data and operational data for this PACT®
system are provided in Tables 5.22 and 5.23, respectively.

                                 5.105
-------
=st, tit ,:  «=-      -r,
        2~r    ;:::
• - - ; - s — .











Table 5.18

Mass
Balance for
VOCs in Activated
Sludge
Mean Concentration


Volatile Organic


He

Influent

(ug/L)(ug/L) (ug/L)
Methyl methacrylate
cn
'_, Benzene
O 1,4-Dichlorobenzene
Cis-l,2-Dichloroethene
	 Trans-1, 2-Dichloroethene
Ethylbenzene
Styrene
Trichloroethene
Toluene
Vinyl chloride
0.0102
0.1456
0.1080
0.1096
03328
0.1341
0.0500
0.2463
0.1364
0.9040
1,010,000
18
18
15
23
129
38
28
35
2

Bio Effluent
Liquid Off-Gas
Oig/L) (ug/L)
03 0.0
0.0 OX)
0.0 0.0
105 1.15
0.0 0.0
02 0.03
0.1 0.01
15 0.47
0.1 OX)1
0.1 0.09

Influent

(ug/day)
12,120,000
216
216
180
276
1,548
456
336
420
24
Process
Mean

Effluent

With Off-Gas Recycle
Mass Rate

Stripped

(Ug/day) (ug/day)
3.6
0
0
126
0
24
12
22.8
12
12
1
0
0
249
0
6
1
101
3
20
Mean Fraction of Influent Mass

Biodeg.

(ug/day)
12,119,996
216
216
-195
276
1,540
454
212
416
3

Effluent


0.000
0.000
0.000
0.400
0.000
0.002
0.003
0.065
0.003
0.050

Stripped


0.000
0.000
0.000
1381
0.000
0.004
0.002
0301
0.007
0.814

Biodeg.


1.000 ; '_i
i.ooo ; „
1.000
-1.08
1000
1 AAA> — 	 	
0.995
0.995 ; ;'
0.631
0590 :
0.136
Source: USEPA1996C
- ;,
-------
                        Chapter 5

Table 5.19
Example Activated Sludge with Off-Gas Recycle Performance Data
Parameter
VOLATILE ORGANICS
Methyl raethacrylate
Acetone
Benzene
Carbon disulfide
1 ,4-Dichlorobenzene
Cis-l,2-Dichloroethene
Trans- 1 ,2-DichIoroethene
Ethylbenzene
Styrene
Trichloroethene
Toluene
Vinyl chloride
Methylene chloride
1,1-Dichloroethane
1 ,2-Dichloroethane
SEMI-VOLATILE ORGANICS
Dimethyl phthalate
Diethyl phthalate
Di-n-butyl phthalate
Phenol
2-Methylphenol
Naphthalene
Benzyl alcohol
2-Methylnaphthalene
Bis (2-ethylhexyl) phthalate
4-Methylphenol
2,4-Dimethylphenol
ND = Not detected
Source: USEPA1996C
Influent (|ig/L)

1,010,000
1,000
18
125
18
15
23
129
38
28
35
15
2,500
2,500
2,500

36
48
11
160
47
126
135
251
6,4
82
102


Effluent (jlg/L)

03
8.0
NJ
03
ND
105
ND
02
0.1
1.9
0.1
0.1
5.4
5.0
5.0

0.5
0.5
0.7
05
1.0
1.0
1.0
1.0
3.3
05
1.0


5.107
-------
Groundwater Treatment Systems
                                  Table 5.2d
                 Summary of Operational and Performance
                     Data for Conventional Parameters
                      Parameter
Operating Data
              Process Parameters
               HRT (days)
               SRT (days)
               Temp CC)
               F/M(day •')
    1.6
    12
    23
    0.24
              Aeration Basin Parameters
               MLSS (mg/L)
               MLVSS (mg/L)
               DO (mg/L)
               pH(s.u.)
               OUR [mg/(L-hr)]
               SOUR [mg/(gVSS«hr)]
               Air Recycle Ratio3
   6,300
   5,000
    53
   7.0-8.1
  I  243
    4.8
    0.87
Influent
Flow (mL/min)
TBOD (mg/L)

TCQD(mg/L)
SCOD (mg/L)
TSS (mg/L)
VSS (mg/L)
TDS (mg/L)
pH(s.u.)
NH4-N(mg/L)
P04-P (mg/L)

Effluent
TBOD (mg/L)
SBOD (mg/L)

TCOD (mg/L)
SCOD (mg/L)
TSS (mg/L)
VSS (mg/L)
TDS (mg/L)

NH4-N(mg/L)
P04-P(mg/L)
	 ' ''
|
7.6
1,460

:' 3,320 ; 	 "
	 2,920
46
9
2,245
	 63-7.2
42
' 	 ' 21 ' " " i""
„ , • • • :, 'j :, '„",::•

28
15
•I
432
288
""85 ' .
' 	 80
1,660

10
"' 3.4 "
1 • " l|-' '! '' " !l • • '
* Recycle air flow/total air flow.
Source: USEPA1996C
                                    5.108
-------
Cn


Mass
Balance for
VOCs in
Table 5.21
a PACT® System
Mean Concentration
Volatile Organic
Methyl methacrylate
Benzene
1 ,4-Dichlorobenzene
Cis-l,2-Dichloroethene
Trans-1 ,2-Dichloroethene
Ethylbenzene
Styrene
Trichloroethene
Toluene
Vinyl chloride
He
(|ig/L)(ng/L)
0.0102
0.1456
0.1080
0.1096
0.3328
0.1341
0.0500
02463
0.1364
0.9040
Influent
(Hg/L)
1.010,000
18
18
15
23
129
38
28
35
2
Bio Effluent Influent
Liquid Off-Gas
(Hg/L) - (|ig/L) (Hg/day)
122
03
0.0
6.7
60
02
0,1
108
02
0.1
0.12 6.060,000
0.01 108
0.00 108
0.43 90
0.00 138
0.03 774
0.02 228
0.44 168
0.03 210
0.09 12
without
Off-Gas
Recycle
Mean Mass Rate
Effluent
(Hg/day)
732
1.8
0
402
0
12
Z4
10.8
12
0.6
Stripped
(Hg/day)
27
9
0
159
0
6
4
95
6
20
Biodeg.
(Hg/day)
6,059,900
97
108
-109
138
767
221
61
203
•8



Mean Fraction of Influent Mass
Effluent
0.000
0.017
0.000
0.447
0.000
0.002
0.011
0.064
0.006
0.050
Stripped
0.000
0.087
0.000
7.762
0.000
0.007
0.019
0.570
0.028
1.627
Biodeg.
1.000
0.898
1.000
-1211
1.000
0.991
0.969
0.363
0.967
-0.667
Source: USEPA1996C










Chapter 5
-------
Groundwater Treatment Systems
                             Table 5.22
                Example of PACT® Performance Data
Parameter
, i. ,'"! I.1 • , • , , ii» • '"• ,!' "'. ' i , ' '
VOLATILE ORGANICS
Methyl methacrylate
Acetone
Benzene
Carbon disulfide

1,4-Dichlorobenzene
Cis-1 ,2-Dichloroethene
Trans-l,2-Dichloroethene
Ethylbenzene
Styrene
Illnl' ' •' ' '"' ,ll •' ''
Trichloroethene
Toluene
Vinyl chloride

Methylene chloride
1 , 1 -Dichloroethane
1 ,2-Dichloroethane
SEMI-VOLATILE ORGANICS
Dimethyl phthalate
Diethyl phthalate
"" " ,, , '' , . 	
Di-n-butyl phthalate
Phenol
2-Methylphenol
Naphthalene
Benzyl alcohol
2-Methyinaphthalene
Bis (2-ethylhexyl) phthalate
4-Methylphenol
, , „„ ' ' .
2,4-DimethyIphenol
Influent (ng/L)
iiiiii",, 	 ; ' i, ,1 i >
1 : :
1,010,000
1,000
18
125
• ,, ••!' i, '" i :'l 'i| ,i, •.
	 • : 'i
,
15
,,
„, J „
129
	 '" 38 ' ' " "
ni>';, j.i 	 I • , •
28 '
	 35"" '
	 is '"
- 	 'i •'.. '
2,500
2,500
2,500
' J
36
i
< < ,.r, " „ ,; i J ,
48
"' ' 	 ' '' '" '
' 	 ii 	
	 usoj
47
. ; 	 il
	 " " '" ' 	
126
1
135
|
251 	 ""
6.4
	 !"'
82
- 	 • i •• H -
102
1
Effluent Qig/L)
.IP, ., 	 	 i 	 	 	 ...

12
2.8
0.3
03
!; ' '"'i'1, " , ''I1 1!;!1,"11'!!1111 ' ::'' ,„'

vs>
6.7
, I " ,
	 ' 	 '"• !" ' ' 'T"1"" ' 	
0.2
0.4
;, ' > : '.ii'i1 H1" 'H'T , i, ,,,(, , /jiiiiiji.
' 1.8. 	 " " '^\ "_
" 	 02
0.1

7.0 =" '':'"'' ' "' *"
5.0
5.0

05
0.4
" 	 ' '' 	 " '' '
0.6
0.5
1.0
1.0' 	
1.0
i.o 	 : 	
20
05
	 	 i:>, 	 ,.< i 	 MI in, -„ 	 / i
1.0
ND - Not detected

Source: USEPA1996C
                                                          'i'» .WSJli:1 Ill'iiU, ..... ' id ...... i,!" '
                               5.110
-------
                                                                      Chapter 5
                                   Table 5.23
                 Summary of Operational and Performance
                     Data for Conventional Parameters
                      Parameter                 Operating Data

              Process Parameters                        1-6
               HRT(days)                            12
               SRT (days)                            23
               Temp (°C)                             0.5
               F/MCday'1)

              Aeration Basin Parameters
               MLSS(mg/L)                          6,580
               BIOMLVSS (mg/L)                     4,300
               Carbon TSS (mg/L)                     1,558
               DO (mg/L)                            4.8
               pH (s.u.)                              75-8.1
               OUR [mg/(L-hr)]                        23.4
               SOUR [mg/(gVSS'hr)]                    53
               Air Recycle Ratio8                      O-87
Influent
Flow (mL/min)
TBOD (mg/L)
TCOD (mg/L)
SCOD (mg/L)
TSS (mg/L)
VSS (mg/L)
TDS (mg/L)
pH (s.u.)
NH4-N (mg/L)
. P04-P(mg/L)
Effluent
TBOD (mg/L)
SBOD (mg/L)
TCOD (mg/L)
SCOD (mg/L)
TSS (mg/L)
VSS (mg/L)
TDS (mg/L)
NH4-N (mg/L)
P04-P (mg/L)

3.8
1,460
3,320
2,920
46
9
2,245
63-7.2
42
21

, 43
23
682
305
215
196
1,797
6.0
0.8
° Recycle air flow/total air flow.
Source: US EPA 1996C
                                      5.111
-------
                                    Groundwater Treatment Systems
•'iiiiJi,
5.4.4.2  Fixed-Film Systems
   Media for biomass attachment includes random or structured plastic me-
dia, granular activated carbon, sand, wood, or rock, depending on the type of
process and application. As the liquid flows across the substrate with the
attached biomass, organics, nutrients, and oxygen diffuse from the liquid to
the biomass. As metabolism of the organics occurs, byproducts and carbon
dioxide diffuse put of the biomass into the liquid phase. As oxygen flows
into the biological film, it is consumed through biological respiration.  The
zpne in which oxygen is present is aerobic while deeper into the biofilm
layer where oxygen does not penetrate, the metabolism becomes anaerobic.
   Fluidized-Bed Reactor. The fluidized-bed reactor process is a highly
robust treatment technology that is particularly applicable to groundwater
treatment due to its broad operating range. The FBR process consists of a
fluidized bed containing granular activated carbon or sand as the biomass
support. The medium is fluidized using influent groundwater and, in cases
of high organic concentrations, a recycle stream of treated effluent. A por-
tion of the fluidization flow is passed through a bubble contact unit where
pure oxygen is added under pressure causing complete dissolution of the
oxygen.  The oxygenated influent passes through the FBR where contami-
nants are removed by biodegradation and sorption.  Stripping does not occur
as the oxygen addition process eliminates the release of bubbles that could
cause stripping.  If GAC is used as the support mechanism, adsorption of
recalcitrant organics will also occur.  The recalcitrant organics are then bio-
degraded over time due to their increased contact'time with the biomass
attached to the media.  Thus, the GAC is auto-regenerated and generally
does not require replacement, except as makeup for attrition. Table 5.24
shows  performance data for treatment of a contaminated groundwater.  Typi-
cal design criteria are shown in Table 5.25.
   Biological Activated Carbon.  Biological activated carbon (BAG) is simi-
lar in concept to the FBR process except that the BAG process operates in a
downflow mode. Biomass grows on the granular activated carbon and si-
multaneously biodegrades organics from the liquid phase and those that
adsorb onto the carbon. Due to the potential for plugging of the GAC bed,
this process is only applicable where the influent BOD concentration is less
than approximately 10 mg/L and individual toxic organic concentrations are
less than approximately 200 mg/L. Additionally, provisions for periodically
backwashing the BAG column must be made to prevent complete plugging
of the bed due to biological growth. Performance data are shown in Table
5.26. Design criteria are generally the same as for a GAC unit. The design
                                                                    5.112
-------
                                                          Chapters
should provide for backwashing, for aerating the influent prior to the BAG
unit, and for nutrient addition.
                             Table 5.24
          Example Fluidized-Bed Reactor Performance Data
Parameter
VOLATILE ORGANICS
Methyl methacrylate
Acetone
Benzene
Carbon disulfide
1 ,4-Dichlorobenzene
Cis-1 ,2-Dichloroethene
Trans-1 ,2-Dichloroethene
Ethylbenzene
Styrene
Trichloroethene
Toluene
Vinyl chloride
Methylene chloride
1 , 1 -Dichloroe thane
1 ,2-Dichloroethane
SEMI-VOLATILE ORGANICS
Dimethyl phthalate
Diethyl phthalate
Di-n-butyl phthalate
Phenol
2-MethyIphenol
Naphthalene
Benzyl alcohol
2-Methylnaphthalene
Bis (2-ethylhexyl) phthalate
4-Methylphenol
2,4-Dimethylphenol
Influent (fig/L)

3,240
80
25?.
17
0.9'
61
U
41
2.0
23
ffl
M
131
131
131

23
13
1.1
OS
5.0
4.4
1.0
0.8
1.4
0.5
34
Effluent (ng/L)

0.4
34 :
15
05
0.0
3.4
0.0
0.1 ]
0.1 :
02
0.0
27.0
4.0
42
4.4

05
0.4
1.6
05
1.0
1.0
0.8
1.0
95
05
1.0
Source: US EPA 1996c
                               5.113
-------
H	
            Groundwater Treatment Systems
                                            table 5.25
               Example Fluidized-Bed Reactor Operational and Performance Data
                                     Parameter
Operating Data
                             Process Parameters
                              HRT(hr)
                              DO(mg/L)
                              pH(s.u.)
                              Temperature (*C)
                              COD Loading (lb/1,000 ft3)
                             Influent
                              Influent How (mL/min)
                              TBOD (mg/L)
                              TCOD (mg^L)
                              SCOD (mg/L)
                              TSS (mg/L)
                              VSS(mg/L)
                              TDS (mg/L)
 I-  '	J'
  •. i	
   |
    Zl
  6.6 to 7.3
   1 22
    160
 ';	i''
    16
 r:L®
    193
  	; y •
    363
pH(s.u.)
NH4-N(mg/L)
Effluent . '" ,/' 	 " ' ", '
r ^tBOEXmg/L)
SBOD(mg/L)
SCOD(mg/L)
• . ; ' .. , ' . fss(mgnj) ' ' ;;' ^ ;;;'
VSS(mg/L)
TDS (mg/L)
NH4-N(mg^)
Pb4-P(mgfl.)
Source: US EPA 1996c
•"•\ ', -:r • i ••"' ';• •! 	 !.^'''iii.;i«1 - j l|;;"'": ;;> *
6.4 to 7.2
6.8 ''" ':" '";"" "i;"': '
12
9
•••^ 	 '35;; 	 :" •_• 	 •• • ; ; ;; ;(
' 24'
295
3
..,.,,.,... , .. • ; . t.^- .,
.;; 	 •.;:• '..--I... ;
'i M , " ••• • ,V: " • , .
., ' ' 	 '
                                               5.114
-------
                          Chapter 5
i
Example BAG
Parameter
VOLATILE ORGANICS
Methyl methacrylate
Acetone
Benzene
Carbon disulfide
1 ,4-Dichlorobenzene
Cis- 1 ,2-Dichloroethene
Trans-l,2-Dichloroethene
Ethylbenzene
Styrene
Trichloroethene
Toluene
Vinyl chloride
Methylene chloride
1,1-Dichloroethane
1 ,2-Diehloroe thane
Table 5.26
Process Performance
Data
Influent Qjg/L) Effluent Oig/L)

2,6(50
23
35
1.4
07
2.6
02
2.4
O.S!
161
62:
34
24
24
24

0.6
3.6
ND
0.9
ND
03
ND
0.1
JO
0.1
0.1
17
5.0
4.9
4.7
SEMI-VOLATILE ORGANICS
Dimethyl phthalate
Diethyl phthalate
Di-n-butyl phthalate
Phenol
2-Methylphenol
Naphthalene
Benzyl alcohol
2-Methylnaphthalene
05
1.0
15
5.7
1.4
0.8
0.9
1.0
Bis (2-ethylhexyl) phthalate 4.0
4-Methylphenol
2,4-Dimethylphenol
ND = Not detected
Source: US EPA 1996C
03
26


05
0.4
12
2.8
1.0
1.0
1.0
1.0
33
05
1.0


5.115
-------
 Groundwater Treatment Systems
     '.,•,!,             ' ,'  'u, '  ,i ,.,'":i" :' , niinij  '"iinii'ij '  • '"  it"1',1"i| !',;,•'", . 1'  • ill:, •      	• , n<  •! r niii'i'i , .si'i'1,
   Submerged Aerobic Filters. Submerged aerobic filters consist of a vessel
 with a random or structured plastic packing. Liquid flows through the vessel
 while oxygen is provided through diffused          Biomass attaches to the
 media which is submerged in the vessel. This process can operate over a rela-
 tively broad range of influent concentrations and offers flexibility due to the
 attached growth concept.  Excess biomass periodically sloughs from the media
 and must be removed by sedimentation, flotation, or filtration processes.
   Rotating Biological Contactors. RBCs consist of a series of plastic
 wheels (drums) attached in parallel. The drums rotate slowly with typically
 25 to 50 percent of the drum submerged at any one time in the influent being
 treated. Biomass attaches to the plastic drums and sorbs contaminants and
 nutrients from me influent while submerged. When not submerged, oxygen
 "•:,!,,. ";,,|.iiii."   ,   ', „, •   ''in i'1: ,' •    ,  HI	 • ' '. |!i'l»|iii!J.JI!I|l|l|'H ii:,!";,!!!1 •  ,• • uv^i" L  •  j  . ' •'  |M |||! • «	*•   ''' '• j.	
 diffuses from the atmosphere into the biomass.  Excess biomass is sheared
 from the media and is removed in a clarffier or filter. RBCs can treat
 groundwater over a broad range of concentrations by using a recycle stream
 to dilute high influent concentrations and by using multiple pass systems.
 However, due to mechanical  problems with the shafts encountered in the
 past, this process has not been used significantly in groundwater treatment
 applications. An advantage of this process, however, is that volatilization of
 organics is minimized due to its passive mode of aeration.

 5.4.4.3 Slurry Reactor Systems
   In slurry reactors, soils or sediments are mixed with water to form as high
 a solid contqnt as can be managed, up to 35% to 40% solids. The limiting
 factor is usually the density of the solids and the ability of the mixing equip-
 ment to maintain a well-mixed system. Sediments are generally less dense
 than soils and can be maintained in suspension more easily and at a lower
 energy requirement, ranging from 0.1 to 1 horsepower per  1,000 gallons.
 Mixing is provided by mechanical means, by aeration, or a combination of
 mechanical means and aeration. Although mixing is generally considered
	necessary,	Remediation Technologies InZ (^M^lctiS jgjJ^ i990cynas	
 reported a case where degradation rates were actually faster in an unmixed
 system than in a mixed system.
 '"  '"i" •, •	! .•.  " ).. "i1, 	 '•  	 I'.ir'Siii	 If	 '""'ii! • . ; Ii 	Ml  111' 1-   • •  ' ,.,;••* ' ," , ; •„•"()(. |"J	* ,:'!!,
   Oxygen can be provided through aeration below the surface which en-
 hances mixing.  However, this also increases emissions of volatiles which
 may require capture and treatment. Use of oxygen instead of air reduces the
 volatile emissions but also minimizes contributions towards suspension of
 solids, requiring more substantial mechanical mixing.  Recovery of an aque-
 ous stream, with subsequent  saturation with oxygen prior to recycling of the
                                  5.116
-------
                                                            Chapter 5
aqueous stream to the reactor, as was the case at the French Limited
Superfund Site, nearly eliminates emissions.
   Nutrients are added and chemicals are added for pH adjustment to the
initial slurry reactor feed. During treatment, further nutrient addition and pH
adjustment are typically added automatically as needed.
   Slurry reactors are operated in one of two processing modes:  batch or
continuous/semi-continuous. Slurry reactors are operated in the batch mode
by filling the reactor, treating for a fixed length of time, and discharging the
slurry at the completion of treatment. Continuous/semi-continuous mode
reactors have a constant influent and effluent stream with the slurry remain-
ing within the reactor based on the design hydraulic residence time.
   Batch operation is the most common configuration for slurry reactor op-
eration and offers several advantages. Batch operation generally requires
less sophisticated equipment, instrumentation,  and control than a continu-
ously-operated reactor, and is flexible in terms of solids processing rates it
can achieve. The process can be sampled and monitored with relative ease
and material can be held in the reactor until analytical results for treatment
criteria are complete and desirable treatment endpoints are reached. Batch
reactors also yield the highest conversion for a given reactor volume. The
main disadvantages of batch reactors are the "down-time" to drain the reac-
tor and refill it with slurry, and the fact that it does not operate under pseudo-
steady-state conditions as does a continuous feed reactor that has stablized.
   Continuous flow reactors require a greater degree of operational and
monitoring sophistication.  Both upstream and  downstream process opera-
tions must be matched with the processing rate of the reactor. Depending on
regulatory requirements for demonstration of treatment, the treated slurry
may require out-of-reactor storage until laboratory data confirm treatment
requirements have been met.

5.4.5 Pretreatment Process

5.4.5.1  Groundwater Treatment Systems
   Pretreatment may be necessary prior to treatment in aboveground reactors
to prevent toxicity to the biomass or accumulation of inert materials in the
biomass that is generated during treatment.  Typical pretreatment processes
for groundwater include oil/water separation where free product may be
encountered and chemical coagulation/flocculation/sedimentation for pre-
cipitation of toxic levels of heavy metals.
                                5.117
-------
                      Groundwater Treatment Systems
liilf
fill
                         Oil/water separation typically consists of a corrugated plate or coalescing
                      oil/water separator between the extraction well(s) and the treatment system.
                      Product, either LNAPL or DNAPL, is separated based on differences in
                      specific gravity between it and the groundwater to be biologically treated.
                      Free product is removed either as a supernatant (LNAPL) or subnatant
                      (DNAPL) and collected for off-site disposal.  The aqueous phase is then
                      pumped into the bioreactor for removal of dissolved phase organics.
                         Groundwater containing elevated concentrations of heavy metals gener-
                      ally must be treated either to meet final surface water discharge or ground-
                      water reinjectiori requirementsTor to"prevent biomass toxicity and accumula-
                      tion in the biological sludge which would affect treatment plant operation,
                      performance, and sludge handling and disposal (Table 5.27). In addition,
                      .elevated iron groundwater concentration (e.g., >1 mg/L)can cause fouling of
                      aeration equipment and other downstream processesT Common technologies
                      for metals removal include:
                          •...„••  hydroxide precipitation using lime or caustic soda,
                                                                 i|: ,"i| ,   .    ' '       , '    '
                              •  iron co-precipitation using naturally occurring iron or ferrous
                                 sulfate,
                                                                   i
                              •  sulfide precipitation,
                              •  electrochemical precipitation, and
                              •  ion exchange.
                         Achievable effluent concentrations for various metals using applicable
                      technologies are summarized in Table 5.28.
                                                                         I	"i;: ;, MI 	 ,•{>	u
        5.4.5.2  Soil Treatment Systems
           Like many solids treatment processes, the first step in the application of a
        slurry reactor is screening of the contaminated soil to remove material
        greater than a specific size, in general about"5 cm (2 hi), that is not effec-
        tively tteated in me reactor.  This soil screening equipment can also reduce
        me size of large soil clumps. Tneremovaf of large objects and the reduction
:;:;!!i!  ; -;; -;, in.sjze of others" enhances' soil"and'sediment mixing and enables the con-	
• H!   ":i ' tarriinated'material to be suspended and'efiiciently'"mixed" within the slurry
        reactor. Further size reduction._ will occur by mixing to expose more soil/
        sediment surface to the aqueous phase.
           Soil screening equipment is typically capable of processing 80  to 100 tons
        of excavated material per hour.  The equipment is designed to reduce soil
        "clumps" prior to screening, remove material larger than 5 cm (2 in.) (e.g.,
        concrete, rubble^ timbers, etc.),  and convey it into a debris stockpile.  The

                        '	    •   ••    ^ 5.1 is  "' "  '    ^   ;;	:	;	_

        ,,l        ,    "T  , i ,	   ,',    '   " , ;,,],,' i',,, ;,,  ||  , ill i, i',!! iiin	'«, iir.""»!'hi' ,'" , ' il'ni	' • i. "i! "I'lir iiin!i,ii   	i „ "  'i	il™	ii'i	i
               ' '   ',, '               '   	, '  "',,,"     , ' ' ,j , |[   • ', ',„,', '"A • '' "    " ,  • ' ',  ;' , ', ' '• ,",,, „,,,
-------
                                                               Chapter 5
material less than 5 cm (2 in.) is conveyed into a separate stockpile. The soil
screening equipment typically screens all of the soil  independent of the
downstream processing schedule.
                                Table 5.27
    Threshold Inhibition Concentrations to Activated Sludge Processes
                    Constituent
Threshold Inhbition Concentration
         (mg/L)
             Arsenic
             Cadmium
             Chromium (+6)
             Chromium (+3)
             Copper
             Cyanide
             Lead
             Mercury
             Nickel
             Silver
             Zinc
          0.1

          1-10
          1-10

          15-50

          1.0
          0.1-5
          1.0-5
          0.1-1
          1-Z5

         0.25-5
          0.3-5
 Source: USEPA1996C
   In soils which contain a range of particle sizes (e.g., sand through clay), a
soil washing system should be considered because organic contaminants
tend to concentrate in the humic and fine size fractions of the soil (i.e., or-
ganic matter, silt, and clay), while the contaminants associated with the
coarse size fraction (i.e., sand and gravel) are: primarily surficial.  Soil wash-
ing, which separates soils by particle size, can separate the soils into several
size fractions. The objective of soil washing then is to reduce the volume of
soil requiring further treatment by concentrating the waste stream into a
smaller volume of fine-grained material that can be effectively treated in the
slurry-phase bioreactor while producing a washed, large particle size soil
fraction which meets the site treatment criteria.  In some instances, the sand
fraction may represent 80 to 90% of the total contaminated soil mass, and
can be treated to clean-up requirements by soil washing alone. This greatly
                                  5.119
-------
                                    Groundwater Treatment Systems
": iiit
                                                                   Table 5.28
                                               Achievable Heavy Metals Removal Concentrations
Metal

	 . ,, " 	 ::, Arsenic (As)i
,, ,p
Barium (Ba)
Cadmium (Cd)
• "'';•; 	 "
«ii 	 " ' ',','' "' '' ,
Copper (Cu)

Lead (Pb)
Mercury (Hg)
Nickel (Ni)
Zinc (ZN)
Source: USEPA1996C
Achievable
Concentration
(mg/L)

0.05
0.06
0.005
0.5
0.05
0.05
0.008
0.02
0.01
0.01
,0.02 "
0.01
0.001
0.005
0.12
0.1
1 	 |li|r 	 ,lll|n , , , , 	
•: „ , ' 	 i j .• . , i • 	 • v.
Technology

Sulfide precipitation (pH 6-7)
Carbon adsorption
1 II ,,' ' '1 ," i ,l> ' I,'
Co-precipitation (ferric hydroxide)
III J , . llj'll! ' II,,,, I,,' ' , ' il 1,1 	 >
Sulfate precipitation
Hydroxide precipitation (pH 10-11)
Co-precipitation (ferric hydroxide)
Sulfide precipitation (pH 6-7)
Hydroxide precipitation
Sulfide precipitation (pH 8.5)
Carbonate precipitation (pH 9-9.5)
Hydroxide precipitation (pH 11.5)
Sulfide precipitation
Co-precipitation (alum)
lini" i, i ?> ,;,it- , l| ,:", ' , ' :,'" ',„ ' • , ! ::f ',•„,; '
Co-precipitation (ferric hydroxide)
Hydroxide precipitation (pH 10-11)
Hydroxide precipitation (pH 10)
in 	 ,i ,"; , ' ,,!• j1 ' : " i l; ,, , „ I 	 iiiii;,,, "ii
                            i!	I!11'i
reduces the volume of solids to be treated in the slurry reactor, and thus the
size of the reactor required or the number of batches required for complete
site treatment.
   A slurry preparation/soil washing system processes the screened material at
an average rate of 8 to 12 tons per hour. The process separates the stockpiled
material into several size fractions utilizing several separation techniques. The
unit operations include wet- and dry-screening, hydrocyclones, sand screws,
and hydro classifiers  The slurry" is treated with conditioning chemicals and the
largef size fractions wasfied arSTseparated" from the slurry. The remaining
slurry is generally amended with nutrients and pH adjusted prior to transfer to
the bioreactors.  (Refer to Liquid Extraction Technologies: Soil Washing, Soil
Flushing, Solvent/Chemical, Mann et al. 1998 for details).
                                                                      5.120
-------
                                                            Chapter 5


   In some cases, chemical pretreatment has been used on contaminated
soils prior to slurry reactor treatment to partially degrade relatively recalci-
trant chemicals.  Hydrogen peroxide has been used in conjunction with
slurry treatment to pre-oxidize biologically recalcitrant targeted contami-
nants, forming oxidized intermediate compounds that are more susceptible
to biological treatment than the parent compounds. Applications of this
strategy have been predominantly targeted alt polynuclear aromatic hydrocar-
bon contaminants found in gas station site pollution. There are some con-
cerns with the use of hydrogen peroxide.  High concentration can effectively
disinfect the contaminated solids, likely requiring culture addition following
pretreatment. Iron in the soil also limits the effectiveness of this pretreat-
ment process because trace amounts of iron can autocatalytically decompose
hydrogen peroxide.  The oxygen formed by decomposition will be lost from
the liquid phase and will have little effect on the treatment. Peroxide mix-
tures are available to avoid rapid decomposition, but notably these mixtures
are less reactive than the pure hydrogen peroxide.
   An integrated chemical/biological treatment process (MGP-REM) has
been applied to manufactured gas plant (MGP) contaminated soil in New
Jersey (Liu et al. 1994). The MGP-REM process combines chemical oxida-
tion as a pretreatment for the difficult-to-degrade organics followed by aero-
bic slurry processing.  Mild chemical oxidation with Fenton's reagent (hy-
drogen peroxide and ferrous ion) produces hydroxyl radicals which are ex-
pected to propagate a chain reaction of oxidation which leads to structural
modification of the targeted organics. Bench-scale studies have shown that
application of the MGP-REM process can enhance the rate and extent of
treatment for the PAHs found in contaminated MGP soil.  Field-scale imple-
mentation of this technology  uses excavated soil which has been screened in
two stages to the minus 20 mesh fraction.  Conditions for the pilot treatment
study are presented in Table 5.29, along with results of the study showing
significant removals of 2- through 6-ring PAH constituents during treatment.

5.4.5  Posttreatment Process

5.4.5.1 Groundwater Treatment Systems
   Effluent polishing from aboveground reactors treating contaminated
groundwater may be required to comply with final discharge limits. Typical
posttreatment processes include air stripping for removal of residual volatile
organics, and granular activated carbon (GAG) for removal of volatile and
semi-volatile organics, pesticides and herbicides, as well as biological pro-
cess degradation products. Filtration is typically required prior to any

                                5.121
-------
is;1'
IIP1;;; . fi

Ilii,
 "W! '!' .
 ":„.!" i I' ."', '
                "
               -if!
                      Groundwater Treatment Systems
                      posttreatrrient process to prevent fouling of the treatment equipment or to pre-
                      vent clogging of reinjection wells (if employed). Filtration may consist of sand
                      filters or multi-media (e.g., anthracite and garnet) filters to reduce effluent total
                      suspended solids concentrations below 1 to"5 mg/L. If greater filtration effi-
                      ciency is required (e.g., for groundwater reinjection), bag filters, cartridge fil-
                      ters, or membrane filters can be used to provide essentially complete solids
                      removal for solids with diameters as low as 1 |om. Chemical oxidation (e.g.,
                      Uy/peroxide or UV/6zone) can be usectas an effluent polishing step to remove
                      recalcitrant organics in treated groundwater via chemical oxidation.
                                     .'    '. •    .      .:-  •'.  >'»•'  -	"-,:	i	-  './•	:..-
                                           •  ;  	   Table 5,29	
                                  MGP-REM Treatment Conditions and Performance
                                                 Factor
                                                                      Operational Range
                                   % Solids in Reactor

                                   Slurry Residence Time (days)

                                   % Hydrogen Peroxide Concentation (vol/vol)
                                        " i^'",''  j.   	:;» '	i,.11 'Nit,, , ' „;; ' 	;   u	j	"^i '^
                                   Initial Soil PAH Concentrations

                                   Unit ProcessScheme
                                         1	10 to 35

                                          ' 8 to 20
                                          0.5 to 2.0
                                         100 to 2,000
                                           4 Stage
                                      Biological/Chemical
                                                                                           l!'"i|',, " 1 '':V  ' ''',,,11'"
                                            Pollutant Category
                                                                        % Removal
                                   Total
                                                                           95
                                                    1 ''"'"I   i"" -!	;!"	 M 'I!"V, * ', ' i I' '.,.'• I  iM", •' '"•	 !
                                                     1 i? .", : .' HI  w .si.,. • , 'i'li'i,,'!"',,	i  i1- ' -Hi i i n v i* _i|ii * '" '
                      Source: Liu at at. 1994
iK1   "iandi-RingPAHs"'

          I 6-Ring PAHa
           _/il..|;;,   j '__,  ' ...i	i ,;.;-i!( ,»,, j., 1^1 ;H*J•*,/." .^ i 'i'1''ftii"'.,'ij« '^	
                                                                                              !;l	JliC
                                                                                             •• ~ j	
                                   4, 5, and 6-Ring PAHs
                         Final pH adjustment and removal of residual dissolved oxygen should be
                      strongly considered for applications that use groundwater reinjection to pre-
                      vent chemical incompatibility or induced biological growth in the aquifer
                      which can lead to premature fouling and plugging of injection wells. Dis-
                      solved oxygen can be removed through chemical means (e.g., sodium
                      sulfite) or by sparging nitrogen or carbon dioxide. Adjustment of pH can be
                      accomplished with carbon dioxide or hydrochloric acid addition.
                                                         5.122
-------
                                                             Chapter 5
5.4.5.2 Soil Treatment Systems
   At the completion of treatment in a slurry bioreactor, the soil and water
making up the slurry must be separated for further handling. A number of
options exist for liquid-solid separation including drying beds, gravity filtra-
tion, filter presses, centrifuges, and thermal dryers. Depending on project
requirements, the aqueous or solid fractions may require further treatment,
i.e., stabilization of metals in the solids, prior to discharge or disposal. Pol-
ishing processes for the aqueous effluent are the same as for groundwater
treatment systems discussed above. The treated aqueous phase is typically
transferred to a storage tank for re-use in the slurry preparation process, or
can be discharged provided it meets permit limits for release into an appro-
priate receptor. The treated solids can be placed back into the ground or
disposed at an appropriate landfill provided the requirements of the Land
Disposal Restrictions (LDRs) have been met  On-site disposal will require
that moisture be adjusted to allow compaction requirements to be met.
Some drying may be required to prevent water loss during transportation.
   Off-gas treatment may also be necessary, especially for soils contaminated
with a mixture of chlorinated solvents which degrade poorly or not at all under
aerobic conditions, or with highly volatile petroleum hydrocarbons. Depending
upon the concentrations of volatile species, the reactor temperature, the mode of
oxygen supply, applicable local, state, and federal regulations, as well as nui-
sance considerations, off-gas treatment may be a significant factor in the design
and cost of aboveground reactors. Reactors may be designed as enclosed ves-
sels to facilitate the collection and treatment of gaseous emissions and may even
be operated under reduced pressure conditions,

5.4.7  Process  Instrumentation and Control
   The instrumentation and control for biological treatment processes will
vary depending on the type of process being used. For example, an anaero-
bic process requires careful monitoring of pH, alkalinity, methane produc-
tion, and volatile fatty acids. An aerobic process requires monitoring of pH,
dissolved oxygen, oxygen uptake rate, and nutrient (nitrogen and phospho-
rus) levels. Furthermore, different types of aerobic processes have different
monitoring requirements. An activated sludge process requires measurement
of return activated sludge (RAS) flow rate; however, this is not required for
fixed-film processes, such as trickling filters or rotating biological
contactors. Fluidized-bed reactor systems require monitoring of residual
dissolved oxygen levels to control the oxygen dissolution system.  The
height of the fluidized-bed is also monitored to control the biomass wasting
and fluidization pump flow rate.

                                 5.123
-------
                                       	!'!,'•!!!'.'! M	|	!	 i"
                                           r
Groundwater Treatment Systems
   In general, process monitoring and control can be as simple or as sophisti-
cated as desired. Minimal automatic monitoring and control may be appro-
priate for smaller systems or where full-time operator coverage is available.
Where minimal on-site operator coverage is desired, fully automated sys-
tems using a supervisory control and data acquisition (isC ABA) process can
be implemented!. These systems typically use flow 'monitoring' at multiple
locations, level control algorithms, pH control, DO monitoring and control,
and flow control valves.  The computer control system can be programmed
to achieve certain operating conditions with alarms to notify the operator
(remotely)  of problems.  Set points for tank liquid or bed levels, chemical
addition rates,  pH, DO concentrations, arid flow can be adjusted remotely
using a computer interface connection.  Data from selected monitoring
points are collected and archived automatically.

5.4.1	Safety Requirements      .'.'."".,,'!'.,'.''.	    	'	"
  The safety requirements for aboveground reactors are typical of any reme-
dial or process industry system. System design must consider worker health
and safety and meet all applicable OSHA requirements.  Design, fabrication,
installation, and operation codes and guidelines are summarized in Section
5.4.9. Large quantities of chemicals (i.e., nutrients, caustics, acids, surfac-
tants, etc.) may also be present on a site so adequate worker and environ-
mental protection must be provided during storage and handling of these
potentially  hazardous chemicals. Process control philosophy and instrumen-
tation design is critical for monitoring process arid equipment performance
and providing adequate alarms or automatic shutdown to minimize endan-
gering workers, discharging contaminants to the environment, or damage to
equipment.

5.4.9  Specifications Development
  In general, the applicable requirements of the following codes and stan-
dards will govern the design, fabrication, installation, and operation of
aboveground biological treatment systems for bioremediation applications:
       •  Uniform Building Code (or governing local code),
       •  National Electric Code,
       •  National Electric Safety Code,
       •  Uniform Fire Code,
                                            J          '           '
       •  National Electrical Manufacturers Association,
       •  Occupational Safety and Health Act,

                                5.124
-------
                                                            Chapter 5
        • Instrument Society of America,
        • Institute of Electrical and Electronics Engineers,
        • American Society for Testing Material,
        • ANSI Standard B31.1 - Petroleum Refinery Piping,
        • ANSI Standards 31.6-Chemical Plant Piping,
        • American Society of Mechanical Engineers,
        • American Welding Society, and
        • American Petroleum Institute.

5.4.10 Costs
   Costs associated with the treatment of groundwater contaminants in aque-
ous phase, aboveground bioreactors depend on the contaminants of concern
and their concentrations in the influent stream. Full-scale bioreactors have
been used for many years for the cost-effective treatment of domestic and
many industrial waste streams at less than  $0.26/m3 ($1.00/1,000 gal). For
contaminated groundwater systems designed to remove hazardous constitu-
ents in the waste stream, treatment is generally provided at less than $0.797
m3 ($3.00/1,000 gal), including pumping costs. Off-gas treatment, if re-
quired, is expected to cost less than $3.17/kg ($7.00/lb) VOC removed (US
EPA 1993b).
   Cost estimates for slurry-phase reactors are not as well developed, so the
following discussion provides a detailed cost estimate for a hypothetical
slurry reactor treating 8,100 m3 (10,000 yd3) of contaminated soil.  The costs
were developed based on data collected from full-scale operating treatment
systems using the following assumptions:
        •  4,050 m3 (5,000 yd3) of contaminated soil are removed in the soil
          preparation/soil washing process.
        •  Four 720 m3 (180,000 gal) (operating volume) slurry-phase reac-
          tors are operated at a 25% solids concentration.
        •  The reactors are operated in a batch mode with each batch requir-
          ing 30 to 35 days for biological treatment.
        •  The treated slurry is dewatered in a filter press and the recovered
          water is recycled to the slurry preparation/soil washing process.
        •  The treatment system is operated 7 days a week for the duration
          of the project.
                                5.125
-------
                                                                     .(' .W'J"!1!!	IB'if11'11' ''	'
                                                                              „';	;,	\	1	:,	
            '"mi
                       Groundwater Treatment Systems
'It,1.,
,!&,
m	i'<
                            HI11  	,'In!
                           	tl   It
                           . 	•!. i ;  •         ,|            I                      .
              A cost summary for this hypothetical treatment process is presented in
            Table 5.30. The system costs and capacity reflect actual full-scale systems.
            The labor costs include direct costs,, benefits, overhead, and per diem for
            field personnel.  Equipment costs include direct costs, depreciation, and
            operation and maintenance requirements^  Materials, supplies, arid utilities
            include direct costs plus a markup.  The total estimated cost for a commer-
            cial slurry-phase biological process, operating under the above-mentioned
            conditions, is  approximately $2SQ/m* ($20Q/yd*) of material as indicated in
            Table 5.30.
•	i*': „ ,-iiii
 II1' II   '"'I i
                                                         Table 5.30
                                    CpsfAnalysis for Slurry-Phase Biological Treatment
                                       of ClOO m3 (io,000 Yd3) of Contaminated Soil
                                                 Cost Element
                                                                 Cost ($)
                         Engineering Design and Procurement

                         Treatment Costs

                           Site Preparation and Equipment Setup

                           Soil Screening

                           Slurry Preparation
                          ':" '  " ,!  , !' Ill /'.I  , ' „ 'I ........ II"'1!:, ,  « "I, I,  'IPIUjll ..... I" "' jllill
                           Slurry Biological Treatment

                           Decontamination and Demobilization
                              i: n;,;^,  , fi  ....... :, ,„ ..... , |, r ..... ||N|1| , ........  | "i!^ [hl| Jj"',
                           Subtotal, Treatment Costs
                            i" ," ' , TI '  , I II, ill'1' " I!' '""  '
                           Project Administration

                         Grand Total
                                                                                 180,000
1
1
ll.nlli.yUii '»' j| • 1 ,. I!!1"!!: Hi in, 	 " ;
il"! 	 i ,, 	 ,|f ||-.| ,
if! 	 !, • --;,;" jj!
420,000
80,000
250,000
ii !ii,'"iii'i'i • ' ,. '•! i " 	 : "" il 	 "!O|i,.: ;: 'i,™!1 r
650,000
150,000
1,550,000
                                                          ' ..... !i'" ..... ...... ,11 "I I , '"I i!lll! "i1! ,'
                                                                               $2,000,000
                          Three process variables have a significant impact on the total project cost
                       of a slurry-phase treatment system:  (1) reactor solids concentration, (2)
                       residence time in the reactors, and (3) the percentage of material removed in
                         ,	ii,,1.',,  •   ',,in» •'; , 	'"• '   ,    * 'i", i,	V .7 ,	*•	   „  ?rr,,    ,   M „„       ,    „  ,,,,,,
                       the slurry preparation/soil washing system.
                          The effect of solids loading in the slurry reactors on treatment costs is
                       shown in Figure 5.24. In general, for a given reactor configuration, the
                       greater the slurry solids concentration, the lower the unit cost for the con-
                       taminated material.  The upper solids concentration that can be effectively
                       handled in a conventional slurry reactor configuration is limited to
                                                           5.126
IK!!,' , '111'I' ii|| |, ',„!  || 	 I'!!!!;1'1  ,,'  .11, <>,,, '||l <
.llliilllHl	I!!!!;1!,!,,	IE]:; Jliiiii. Ill;iJlililli	ft!:1	,!,,;!;: Ilillliili.iiilillilli'.iJii'H
IPi,' i 	i!;|j if • '     .,": „   • i;,   i,,    "i   ,'i •, 	i,.     ',','."	
Ill	.JiillliiBHlmlliili'i:,,:,!	m 'lii ,	in	"". i''i;;!3iiill!!ll!	^A^ JH. '. .I.!1;1'!1 Hi'",'	ii: L .'il!: I'-ii'	J^frtilii':.,.!':..''':'.;.^!,!	• /iii"' 'ill"!'!'!!1:!" !l!liiiiilili:!l!i '•
                                                                     , ',:', , d j,, il I,;:;"n,;, 	    i1,,'™"", ;:","; i, ;;, i;,; n, ,;i, ,'
                                                                     a.,, ilillg IF i, Biii,; i.1 it fill	»: '.iiJIiigi:	•!ll.,!»	i litii.	i'ig!'Si.'
                                                                                                |
-------
                                                            Chapters
approximately 30% to 35% solids, resulting in a minimum cost of approxi-
mately $56/tonne ($50/ton) of solids. The effect of reactor residence time on
project costs is shown in Figure 5.25. An increased residence time reduces
the through-put of the system, requiring additional labor and equipment
costs to treat the same amount of material per batch. The effect on project
cost of soil washing for removal of material from the feedstock is shown in
Figure 5.26. The greater the quantity of material removed in the soil wash-
ing process, the less material requiring treatment and dewatering in the ac-
tual slurry-phase reactor.
                              Figure 5.24
                Slurry-Phase Biological Treatment Effect of
                Solids Concentration on Treatment Costs
      200
    o
    a
    3
    I
      150
      100
       50
                              _L
_L
                       10      15    .  20     25
                            Percent Solids (weight/weight)
                                                   30
                                                          35
                                                                 40
   Modifications of operating temperatures can also affect slurry reactor
economics. Using a full-scale slurry reactor treating PAH-contaminated soil
as an example, the total capital and operating costs for the boiler system to
heat the reactors were approximately $150/day/reactor. Using an average
solids loading of approximately 152 tonnes/reactor (170 tons/reactor)
(Woodhull et'al. 1993), the cost of heating the reactors equates to approxi-
mately $0.96/tonne/day ($0.86/ton/day). Based on the data collected from
the field reactors, heating the reactors from approximately 25 to 35°C
                                 5.127
-------
Grouridwater Treatment Systems
increased the kinetics for PAH removal by a factor of 1.6. For an initial
concentration of 10,000 mg/kg PAHs, slurry-phase biological treatment at
25°C requires approximately 9 days to achieve the treatment criteria of 950
mg/kg, whereas biodegradation at 35SC requires only approximately 6 days
to reach the same level of treatment. _TJie cost Jo operate the unheated treat-
ment system, adjusted for a 4-month shutdown period during the winter, is
approximately $230/tonne ($205/ton)  for a 9-day batch time (Woodhull and
Jerger 1994; Woodhull et al. 1993). Operation of the treatment system with-
out a winter shutdown at a 6-day batch time yields a total cost of approxi-
mately $202/tonne ($l80/ton). Heating the reactors for 6 days requires an
additional $5.60/tonne ($5/ton) in operating costs, raising the total costs to
approximately $207/tonne ($185/fon); still lower than the $230/tonne ($2057
ton) for the extended, unheated operation. This impact would be even
greater as the ambient and slurry temperatures decrease to 15°C, causing
even longer batch operating times to achieve treatment criteria. In this ex-
ample, heating the reactor allows  for continuous operation of the treatment
system, reducing required operating times, increasing equipment utilization,
and lowering overall life-cycle treatment costs.
               •'  ••  '  ;	''';' •;,":';;': Figure 5125
               Slurry-Phase Biological Treatment Effect of
               Solids Residence Time on Treatment Costs
      200
           111111111111111111111111111111111111111111111111111111
         0       7       14     21     28     35     42     49     56
                              . Residence Time (days)
                                5.128
-------
                                                            Chapter 5
                              Figure 5.26
      Slurry-Phase Biological Treatment Effect of Percent of Material
       Removed in Slurry Preparation System on Total Project Costs
      250
      100
                10
 20     30     40     50     60
Material Removed in Slurry Preparation System (%)
                                                                 80
5.4.11  Design Validation
   The applicability of a biological treatment process should be evaluated by
considering the types of constituents to be removed, the initial and long-term
concentrations expected, the required effluent quality, and cost comparisons
to other technologies.  If the constituents present are readily biodegradable,
the type of biological process to be employed will depend primarily on the
contaminant concentrations. Groundwater or leachate with organic concen-
trations (as BOD or COD) up to 200 mg/L can be treated with FBR or fixed
growth processes. Higher concentrations can be treated with FBR, activated
sludge, or PACT® processes. For highly-contaminated soils and sludges that
have contaminant concentrations ranging from 2,500 to 250,000 mg/kg,
slurry-phase reactors should be considered.  Specific design selection and
validation are best determined by treatability situdies.
   Treatability testing for both aqueous and slurry-phase reactors is generally
recommended as each individual groundwater and soil matrix will have spe-
cific organic removal rates dependent upon the specific constituents and
concentrations in the matrix. For aqueous-phase reactors, testing should
consist of operation for at least 8 weeks of bench- or pilot-scale systems so
                                 5.129
-------
                                     Groundwater Treatment Systems

•i, ins!1:"1!!:,; : ':""•, ,k i, •
              I'M	     • !  "llll'J1,  	I'll
              ,	,       ••• •  i  • • • -i
                       that multiple sludge ages are achieved.  Data collection during the treatabil-
                       ity phase should focus on influent and effluent quality, as well as operational
                       parameters, such as oxygen uptake rate, sludge yield,.sludge settleability,
                       etc, that will be important for design development. It is generally advisable
                       to evaluate at least two operating conditions (e.g., organic loading rate, hy-
                       draulic residence time) for each process under corisideration. For slurry-
                       phase applications, bench-scale reactor studies can identify potential mixing
                       problems withthe contaminated solids and can aid in determining optimal
                       solids concentrations for field-scale reactor bperation.  Temperature effects,
                       bioaugmentation options, foaming control options, and surfactant enhance-
                       ments can all be evaluated effectively at the bench-scale prior to initiating
                       full-scale bioslurry remediation.

•if.  '   ii"1'"   ''I!  "iiifi  .:..   ,  • • •,!'",-I .',„      '.'  ';, I1,!1":1 •!  •              I        ''; i/:'.  'I,':1, i ."'•'»•••   • ',•„,",•!	i: ;>:,, ill'liliiiis „{; i
                       5.4.12 Permitting  Requirements
                          In general, any unit that treats a hazardous waste is subject to RCRA
                       operating permit requirements as a treatment, storage, and disposal facility,
                       unless  the system meets one or more RCRA permit exemptions.  Some types
                       of hazardous waste management units are exempt from RCRA. permitting
                       requirements because their operation is  either regulated under federal arid"
                       state programs or the operation of the unit should result in a minimal release.
                       In additionj the system may meet thei totally enclosed treatment facility
                       (TETF) permit exemption criteria if the system is integrated into the produc-
                       tion process to treat routinely-generated wastes. Enclosed aqueous-and
                       slurry-phase reactors typically qualify for these exemptions.
                          Depending oni the  project location, air discharge permits may be re-
                       quired for bperation of the reactors and any integral off-gas treatment
                       systems that are used. Direct discharge of excess treated water  from the
                       process will typically require a National Pollutant Discharge Elimina-
                       tion System (NPDES) permit,  or an agreement from the local  wastewater
                       treatment facility if it is discharged into a sanitary sewer system.
                       Groundwater reinjection may require  state  or local injection permits.
                       State and local regulations vary and need to be investigated as part of
                       system design to ensure that all required permits are obtained without
                       interfering with system cfesigri or operation.
                            	'	'	;	:	|" ;	[l  	
                       5.4.13 Design Checklist
                     , ,'I1'"   'III' ,!,'	|	I i|' ,;,| „''„",'||,,"|, „";||'I" |	!„!',;	|,|'|||,'"I'I'lli,,, ,' '",!"" ,l	I',,,,':	\	''.".Ill'l,',,!"!!"!!!"",.'!»!,.\	' ',„. '!«'!!	11.1!	',i	1, .,' 	"	!	 ' 	 ',!	',"",
                          Prior to implementation of aboveground reactor design, the items listed in
                       Table 5.31 should be reviewed.
                                                                     5.130
-------
                                                                                  Chapter 5
                                         Table  5.31
              Aboveground Biological Treatment Design Checklist

                             Aqueous-Phase Reactor                  Slurry-Phase Reactor i

Site Features                                        Property lines
                                                Nearest power source
                                           Building/confined space locations
                                                Underground utilities
                                            Normal use and traffic patterns
Site Characterization               .       Type and distribution of contaminants            '
                        Contaminated groundwater volume          Contaminated soil volume
                             Aquifer characteristics               Infrastructure near excavation
                             Groundwater chemistry               Space for soil processing and
                                                                        stockpiling
Soil Characteristics              Aquifer properties              Soil type (particle-size distribution)
                                                                   Soil moisture content >
                                                                 Volume/extent excavation
Test Results                               Laboratory treatability study results
                         Aquifer hydraulic testing results         Processing/handling requirements
Design/O&M                                      Off-gas treatment
                                                  Electrical system
                                                  Monitoring system
                                               Piping,  valves, controls
                                                Nutrient levels/source                   ;
                                              Remedial goals achievable
                                  Well design                    Soil pre- and post-processing
                           Groundwater transfer system                Soil transfer system
Health and Safety                              HASP complete/kept on site
                                                  Personnel trained                     !
                                                    Ground faults
                                              Electrical  system inspection
                                                       Security
                                            Emergency shutdown procedures
Miscellaneous                                         Pennits
                                               O&M manual reviewed
                                                 Monitoring schedule
                                                    Site restoration                     ;
                             Treated water disposal              Treated soil storage & disposal
                                            5.131
-------
::	i	!««!,'	!!	r:,	i
                 Groundwater Treatment Systems
                 5.4.14 Implementation
                    Implementation of treatment using aboveground treatment reactors re-
                 quires the coordination of a number of activities! For all aboveround sys-
                 tems, equipment support pads, treatment buildings, storage buildings, electri-
                 cal services, security fencingarid lighting, access roadways and paths, if
                 needed, and site grading should be completed before any equipment is deliv-
                 ered. Equipment arid tanks should be placed on support pads or in buildings
                 as specified. The systems should be inspected to ensure that all construction
                 specifications have been met. Notification should be made to all appropriate
                 local agencies.
                    For aqueous-phase systems, wells must be installed using appropriate drill
                 rigs and installation techniques. Wells must be completed  using proper pro-
                 cedures to ensure their integrity over the duration of the project. Under-
                 ground transfer lines must be installed from well manifolds to connect to all
                 requisite tanks and pumps.
                    For slurry-phase systems, an important aspect of system implementation
                 is management of soil Curing pre- and post-processing. Stockpile locations,
                 as well as soil pre- and posttreatment areas, needl to be located to minimize
                 soil handling requirements. Equipment selection, both typearid size, are
                 important to the efficiency of the operation. Where rain may be heavy dur-
                 ing pre- and post-processing,  provisions need to be made to cover the con-
                 taminated soils within the pretreatment, stockpile and posttreatment areas.
                 All soil processing and stockpiling areas should be located so that transpor-
                 tation requirements to the slurry reactor are minimized, and provisions
                 should be made to minimize suspension and transport of contaminated soils
                 away from the treatment area during process and transport.
                                          i. y.'n, :••	ii;:"!!1:,),1,; ilk! '•<•; • •, n , |. .• .., ., ,  . ,  ,,    , ,  st , . j im •, ,.

                 5.4.15 Start-up Procedures
                   Start-up procedures vary with the design and the nature (i.e., suspended
                 growth, fixed-film, slurry-phase reactor) of the aboveground reactor being
                 used in a particular application.  Startup begins with checking the system to
                 ensure that no problems have  developed since installation.

                 5.4H 5. t  Groundwater Treatment Systems
                   For an aqueous-phase system, start-up procedures are identical to those
                 described in Section 5.2 for the Raymond Process. Baseline measurements
                 , '.ii, "y,,,'1!!1 'iif /i  iifi1 '  ,,i ,','!' 'iii •, in1 I1"*	' /."'I 
-------
                                                            Chapter 5
prepared, reviewed, and followed. A similar list will have been prepared for
the start-up period and will include measurements of temperatures, flow
rates, and pressures across the system. Prior to loading of the groundwater
into the bioreactor, it normally is seeded with activated sludge from a local
wastewater treatment plant. If contaminants to be removed from the con-
taminated groundwater are recalcitrant or inhibitory, seed organisms accli-
mated in pilot-scale reactors are added to the full-scale system.
   Groundwater recovery is initiated with 100% of the treated water being
polished with activated carbon and discharged to a sewer line or surface
water body until water quality can be demonstrated to meet regulatory re-
quirements. Once this condition has been met, activated carbon treatment is
no longer required.  If groundwater reinjection is to take place at the site, a
portion of the groundwater will be diverted to a surge tank. When the level
in the surge tank exceeds the "low level", the transfer pump is turned on and
water is introduced to the injection wells through the manifold or header.
   During the start-up period, the schedule for measuring and recording flow
rates, meter readings, and collecting samples for analysis is followed.
Samples of the influent and effluent to the groundwater treatment system are
collected for analysis. Water level measurements are made for evaluation of
the groundwater recovery system design so that adjustment of recovery and
injection rates from individual wells can be carried out to achieve the desired
groundwater flow patterns. During this period, modifications to flow rates
and other parameters are made to meet specifications and/or to optimize
performance. Because the system will not behave entirely as expected,
modifications to the O&M manual should be anticipated. Monitoring and
documentation procedures are also evaluated and changes are made to these
procedures as necessary.

5.4.15.2 Soil Treatment Systems
   For slurry-phase treatment systems, prepared soils are diluted and the
slurry is transferred from the slurry preparation system to the reactor (nor-
mally multiple reactors are used with operating volumes typically 380 to 680
m3 (100,000 to 180,000 gal) each).  Feeding the reactor is initially done in
small increments.  Influent and effluent slurry and effluent water phase con-
centrations are monitored as the  reactor solids content is raised to design
levels to ensure that system performance is achieved during startup.
                                5.133
-------
nil niiiiniinii i in niinij i n nun nn i
in i nn i i n in i
I1 III H !
ii iiiniii i nn nn nili in in i
i in i i
I 	 I1 1
iiinnin nn
ii
'l
1 II 111 Illl 1 II 1 II II ;!'"' Ill 1 1
II , 1
i ' !
i i in i null n i

l|»
	 nn In
i
i i

i
1 Hi I Hill mill II II III Illllil Mill 1 111 Mill 1 II
HI n
i
iiiiiiiinjiiiiii iiiiini i
nn
i
1 IIM III III III •
1
1 ll
                                     Groundwater Treatment Systems
 I' 11,11 i '"' ,: ,„: i II
 I1" Will1!' ' , ,  |i
it 1111 i "if:'I.!,!!11!"'
S k 11!' I  , i"! !,: ,';'.
                        	«",	I
S jit'., i, "•,11
III, j 1'" III '»" I' ',''"
ili
Sii,
    5.4.16 Performance Evaluation
      Performance evaluation includes determining contaminant removal effi-
    ciency in groundwater arid soil slurries7Hquid/solid separation efficiency in
    clarifier units, oxygen transfer efficiency and dissolved oxygen levels in
    bioreactors, and throughput rates of the full scale systems. It is highly un-
    likely In all but the smallest systems in relatively homogeneous formations
    or treating relatively uniform soil slurries that the system as originally de-
    signed will provide the best praclicafremeHra'tTon'.1 Th'e 6&M plan shoul3
    incorporate proceduresto evaluate performance and to modify operations as
f;;vt necessary to achieve overall optimal treatment. This requires that the moni-
    toring plan be designed to identify optimization refinements as well as to
    satisfy regulatory requirements and to measure treatment progress. Potential
    changes in operational procedures include modifications to reduce operator
    time, to minimize monthly operating costs, to better distribute nutrients and
    oxygen, or to improve slurry mixing to reduce tlie required treatment time.
    5,4.16.1 Operations Practices
      Operations activities for abovegrouncl bioreactors include: maintenance
    of the treatment system; management of nutrient and electron acceptor addi-
    tion; evaluation and response to monitoring data; and routine maintenance of
    equipment, controls, and monitoring equipment, as well as housekeeping.
    For aqueous-phase reactors treating groundwater, additional operations ac-
    tivities are required for the groundwater recovery system that include: injec-
    tion arid recovery well maintenance and balancing groundwater recovery,
    discharge, and reinjection as appropriate. Slurry pre- and post-processing
    systems operations are needed in the preparation of contaminated soils prior
    to . [treatmentand for dewateririg soil slurries following treatment.
      Maintaining flow in the injection wells is frequently the most time-
    consuming activity in an aqueous-phase treatment system.  The rate of
    remediation is closely linked to the rate of introduction of the electron
    acceptor. If the rate of water injection decreases, the number of months
    over i^hich O&ll, repbrting,  and management costs are incurred in-
    creases. Design and well completion procedures for injection wells are
    important, as is scheduling well redevelopment at a frequency based on
    experience with different types of aquifers, the mass of organics to be
    degraded, and the design flow rate.  Performance and anticipated prob-
    lems with injection wells can be evaluated from changes in the cycling
    frequency of the high/low controllers for gravity feed wells and by pres-
    sure changes in pressure fed wells. Wells can be treated by surging to
    remove fines. Biological growth  and precipitation of calcium or iron
                                                                    5.134

-------
                                                            Chapter 5
can be addressed by adding dilute hydrochloric acid to the well and sub-
sequent recovery of the spent acid after a few hours.
   Treatment operational practices and monitoring are critical because; of the
potential to discharge water that is out of regulatory compliance to surface
water, groundwater, or sewers, all of which can result in fines and poor rela-
tions with the regulatory agency or the public.

5.4.16.2  Operations Monitoring
   As with most remedial systems, monitoring includes baseline sampling
and analysis. Monitoring is most intense during startup, and decreases in
frequency as the biological reactors reach steady-state operating conditions.
To the extent practical, monitoring should be conducted with instrumentation
and automatic recording devices. For remote sites, the ability to interface
with monitoring equipment from an off-site location is particularly important
as this can significantly reduce travel and labor costs.
   Initially, the most critical parameters to be monitored are the influent and
effluent water and/or slurry quality in the reactor treatment system. If the
treatment system continuously does not meet performance criteria suggested
from treatability studies during startup, the system should be shut down and
the reason for poor treatment performance identified. If a source of micro-
bial inhibition or toxicity is identified during foil-scale system startup, addi-
tional treatability studies may be necessary to identify ways of reducing this
inhibition via additional pre-treatment steps, through increased acclimation
periods, or by decreased influent loading rates.
   In aqueous-treatment systems, monitoring groundwater quality changes
is necessary to meet regulatory requirements and to evaluate remedial,
progress.  Interpretation of these data requires a detailed understanding of
the process. Groundwater composition will change  over time, and to some
extent, the more degradable compounds will be treated first, as the more
soluble compounds will be removed through groundwater extraction. Solu-
bilization effects from biosurfactants will frequently result in increased
dissolved-phase concentrations. Thus, interpretation of these data needs to
be made in context of the mechanisms of remediation and the sequence of
their occurrence along the groundwater flow path. It is important that the
client and agencies understand that increased concentrations of various
constituents in the groundwater may occur before improvements to ground-
water quality are achieved.
   The biodegradation parameters, especially nutrient and electron acceptor
concentrations,  are initially intensely monitored in the bioreactors to refine
                                5.135
-------
                                      Groundwater Treatment Systems
I"",. I!,,!,!1    l!  i',
 nutrient and oxygen addition rates. DO, pH, and temperature can easily be
 measured on-site using readily available meters. Phosphate, ammonium ion,
 nitrate, nitrite, carbonate, sulfate, magnesium, manganese, chloride, and iron
 caii be; measured on-site using test kits  During startup, this rapid access to
 data can be beneficial. For routine operations, however, it is not always cost
 effective to use test kits and the results are not often as defensible as data
 obtained from an analytical laboratory.
   Nutrients and electron acceptor concentrations are measured in the concen-
 trate tanks following deliveries and in headers at a location downflow of the
 point of mixing in the bioreactors. These values are compared to tank level
 records for consistency arid compared to the 6&M schedule.  Solids content
 measurements are also routinely monitored to ensure control of microbial biom-
 ass concentrations in aqueous-phase reactors, and slurry levels in slurry-phase
 reactors. As in aqueous-phase reactors, physical parameters are analyze! and
 evaluated as an indicator of process operation in slurry-phase systems. Specific
 parameters include slurry temperature, pH, total solids, slurry density, dissolved
 oxygen content, nutrient concentrations, etc.  These parameters are used to
 operateand optimize slurry reactor performance.
   	;    •	,	[••;	        t 	, !'"
   For aqueous-phase reactors, it is important to gauge monitoring wells
 frequently during the first several months of operation to evaluate
 grburidwater recovery system performance.  Groundwater modelling is
 normally used to design the well layout and select groundwater recovery
 rates. Measured groundwater elevations are used to evaluate groundwa-
 ter flow patterns and to adjust flows from individual wells using a model
 calibrated to site conditions.
   Because of the nature of the contapinated material being treated in a
 slurry-phase reactor, to effectively and accurately monitor itsperformance it
 is particularly important that care be taken to collect representative samples
 at various points in the process and that these samples be analyzed for the
 appropriate chemical, physical, and biological parameters. Sample collec-
 tion procedures are needed for collection from both the slurry reactors and
 the process equipment  and piping. To collect representative samples, a sta-
 tistically-based plan must account for the effect of variability in the feed
 characteristics, particle size distribution, and contaminant distribution on
 soli3s content. Statistically representative sampies must be  collected arid
 must be tested for particle size density, total solids, and slurry density to
 determine how representative:thesample is of thei entire batch.  Analytical
results will be biased if samples are riot representative, i.e., higher fraction of
fines in sample, increased total solids, etc. In general, composite samples
are preferable to discrete, grab samples. Once the samples are collected,
                                                                       5.136
                             	nil
-------
                                                             Chapter 5
specific, repeatable procedures are required for sample handling, prepara-
tion, and extraction.

5.4.16.3  Quality Assurance/Quality Control                :
   QA/QC procedures include practices common to the other bioremediation
processes, including the use of blanks, blind duplicates, and spiked samples
for laboratory and field measurements.  Quality practices specific to both
aqueous- and slurry-phase aboveground bioreaictor processes include the
following:
        •  control of composition of the nutrient and electron acceptor con-
          centrations in the bioreactor systems;
        •  review of health and safety practices;
        •  review of operating practices and training of new field per-
          sonnel; and
        •  routine evaluation  of monitoring and metering equipment,
          valves, etc.
   QA/QC considerations specific to aqueous-phase bioreactors treating
contaminated groundwater include:
        •  comparison of groundwater recovery rate and totals with the rates
          and total from the injection header and discharge line; and
        •  consistency of biological parameter data with changes in ground-
          water quality and mass removed from groundwater extraction.
   QA/QC considerations specific to slurry-phase bioreactors treating con-
taminated soils include:
        •  conducting solids mass balance calculations to compare influent
          soil and water rates  with effluent oversize reject, slurry, treated
          solids, and treated liquid rates (Because the slurry density is not
          necessarily equal through out the reactor, sampling requirements
          and data interpretation are critical.; and
        •  consistency of biological parameter data with changes in soil
          slurry quality and mass removal during slurry treatment.
   The key performance criteria  is reduction in constituents of concern in the
groundwater and contaminated solid  matrix. Other measurements include pH,
dissolved oxygen, and nutrient concentrations in addition to the appropriate
analysis for the constituents of concern. Where applicable, surrogate analyses
may be used for interim sampling events to reduce the total analytical costs.
                                 5.137
-------
Groundwater Treatment Systems
 5.5  Biosparging
5.5.1  Principles of Operation
   Biosparging is an in situ remediation technology used for the remediation
of groundwater contaminants. The process is sometimes referred to as in
situ aeration or in situ air sparging (IAS). IAS can foster the removal of
contaminants through a variety of physical, chemical, and biological pro-
cesses.  To promote the mass transfer of VOCs out of groundwater and the
mass transfer of oxygen into groundwater, pressurized air is injected beneath
the water table. Soil vapor extraction "(S VE) is often used in conjunction
with IAS to control the movement of contaminant-laden vapors migrating
                !'       '   ,   ,  ! •  »!:  ,„',„;•: „' II '    ;• 911, „,' "i  ',;p  	, ',,. 	  	 i,» , ,|	li' ' „',";,»„" •:
from beneath the groundwater table (Brown and Jasiulewicz 1992).
lAS-enhanced aerobic biodegradation of contaminants can occur in both the
saturated and unsaturated zones.
   When pressurized gas is injected into water-saturated porous media, its
behavior dependent upon media particle size and particle size distribution.
Observations of injected air morphology and movement in a
two-dimensional model aquifer, packed with various sizes and mixtures of
glass beads, were made by Ji et al. (1993). In 4-mm beads, which corre-
spond to medium or coarse gravel, "bubbly flow" was observed; air bubbles
of one to three bead diameters in size migrated upward through the pores in
a "stumbling" motion. In 0.75-m.m beads or smaller, which correspond to
sands, silts, and clays, "channeling flow" occurred, and the plume resulting
from sparging was estimated to be 50% water and 50% air. In 2-mm beads,
air was in the form of both bubbles and channels and it was concluded that
this size is where the "transition between the flow regimes occurs."
   In experiments using mixtures  of bead sizes, small variations in the media
resulted in distorted plume shapes. In experiments where confining layers of
fine glass beads were placed above coarser beads, rising air migrated later-
ally upon encountering confining  layers. These experiments demonstrated
that air channels are sensitive to media heterogeneities.
                    ',",'' MI" i, i , 	 ; JPiiili Ni1'1 'ill,,;™  i;1 'i , '  ' i*-'	   I,,     ,„,    „,,      	  ., .
   Therefore, in soils with particle sizes equivalent to sands or  smaller, air
flow is restricted to discrete continuous air channels, and in natural soils,
distorted air channels will predominate  (Ji  et al. 1993). In natural soils, the
diameter of these air channels is estimated to be equivalent to a few grain
diameters (Johnson et al. 1993).
   It has been estimated that soil hydraulic conductivity of 0.001 cm/sec or
greater is necessary for successful air sparging (Middleton and Hiller 1990;
                                5.138

-------
                                                            Chapters
Loden and Fan 1992). IAS appears to be best suited for remediation of uni-
form coarse-grained sands and gravel, free of confining layers, where air
flow will be relatively uniform and more predictable (Marley, Hazelbrouck,
and Walsh 1992; Marley and Bruell 1995).  However, in highly-permeable
soils, air flow will be primarily vertical, which limits the radius of influence
(Nyer and Suthersan 1993).
   In soils containing high percentages of silts and clays, often only minimal
air flow rates can be achieved. In natural sediments, horizontal permeability
can be several  orders of magnitude greater than vertical permeability (Freeze
and Cherry 1979).  If the vertical movement of air is restricted by a confin-
ing layer,  air will migrate in a horizontal direction, possibly spreading con-
taminants (Marley, Hazelbrouck, and Walsh 1992; Martin, Sarnelli, and
Walsh  1992; Nyer and Suthersan 1993). If the injected air pressure should
exceed the overburden pressure,  then soil rupture and cracking can occur,
short-circuiting air flow and reducing system effectiveness (Marley,
Hazelbrouck, and Walsh 1992; Johnson et al. 1993). Therefore, IAS may
not be  appropriate'where confining layers are present.
   IAS has been used for the remediation of groundwater contaminants, such
as BTEX, resulting from petroleum products amd chlorinated solvents, such
as TCE (Bass and Brown  1995).  Biosparging is most often used to add dis-
solved oxygen (DO) to groundwater to facilitate the aerobic biodegradation
of dissolved biodegradable contaminants. IAS can also be used to promote
the mass transfer of VOCs from the groundwater to a vapor phase. Once
volatile contaminants have entered the vapor pihase, they may be transported
to the vadose zone where biodegradation may take place. Alternatively,
contaminant-laden  vapors can be removed from the aquifer via S VE for
subsequent treatment.

5.5.2  Process Design Principles
   The potential for mass transfer of a contaminant from an aqueous phase to
a vapor phase depends on the interfacial area between the phases and the
contaminant's distribution at equilibrium between the phases. The interfa-
cial area available for mass transfer is dependent upon the distribution and
radial extent of air-filled channels during the sparging process.  One method
of estimating a dissolved contaminant's distribution at equilibrium is with
the Henry's Law coefficient.  Henry's Constant (H) describes the ratio of a
contaminant's vapor pressure to its aqueous solubility as follows:

                H = atm / (mol / m3) = atm • m3 / mol               (5.33)
                                5.139
-------
                                                             	I	!	
Groundwater Treatment Systems
   A contaminant must have a Henry's coefficient of greater than 10"5
atm»m3/mole to be stripped from an aqueous phase via IAS (Brown,
Herman, and Henry 1991). Henry's Law coefficients of various groundwater
contaminants commonly found at Superfund sites are listed in Table A-l.
Additionally, the Henry's constants of selected gasoline additives at 25 °C are
as follows: tert-butyl alcohol (TEA), H = 1.20 • 10'5 amrmVmol (Montgom-
ery 1991); di-isopropylether (DIPE), H = 9.97 • 10'3 atm«m3/mol (Montgom-
ery 1991); and methyl tertiary butyl ether (MTfiE), H = 5.72 • 10"4 atm«m3/
mol (Merck & Co. 1983).
   The efficacy of IAS is highly dependent upon (1) the extent of contact of
the injected air and the contaminated soil and groundwater, and (2) the mag-
nitude of aquifer mixing (Bass and Brown 1995). One measure  of the extent
of contact is known as the radius of influence (ROI). ROI can be defined as
the distance from an IAS well to a point where air flow can be detected or
where the effects of air contact, groundwater mixing, or groundwater oxy-
genation are detectable and consistent (Marley arid Bruell 1995). Radially
symmetric air flow is unlikely in IAS system operation.  Figure 5.27 shows a
typical R6l obtained under field conditions.
                             Figure 5.27
           Asymmetric "Real-World" Radius of Influence (ROI)
                                           Sparge Well
   A field pilot-scale evaluation is usually conducted to determine an
LAS-well ROI. However, no standard method exists to deduce the ROI from
field data. In practice, the ROI is estimated from one or more experimental
measurements. Several widely-used ROI monitoring parameters are de-
scribed below.
                       ,     •      .      '!• 4                    i .,•
                                          1
                                5.140
-------
                                                            Chapter 5
   Groundwater mounding is the upward movement of the water table in
response to the injection of air into the saturated zone.  Mounding indicates
bulk water displacement and is often used as an indication of ROI because it
is easily measured. However, caution should be used when employing
groundwater mounding as a measure of ROI because mounding is transient,
generally negligible under steady-state conditions, and extends beyond the
region of air flow in the saturated zone (Lundegard 1995). New technolo-
gies that are being used to measure ROI  on an experimental basis include the
use of neutron probes and electrical resistance tomography (ERT). Neutron
probes have been successfully used to measure changes in the percent of air
saturation in a saturated sand (Acomb et al.  1995).  ERT uses
cross-bore-hole resistivity surveys to yield a multidimensional image of air
distribution in the saturated zone (Schima, LaBrecque, and Lundegard 1994;
Lundegard 1994; Lundegard and LaBrecque 1995).
   Tracer gases, such as helium (He) or sulfur hexa-fluoride (SF6) have been
used to determine IAS ROIs. Generally, tracer tests are run with SVE wells
in operation.  A typical test consists of injecting a slug of tracer gas into the
pressurized line connected  to the sparging well. Tracer gas content is then
measured in vadose zone monitoring points to determine the gas-phase tracer
content in wells surrounding the IAS well. Portable instruments, such as the
Mark Model 9820 He detector or TIP Model 5550 halogen detector, can be
used to quantify gaseous He and SF6, respectively (Johnson et al. 1995;
Baker, Hayes, and Frisbie 1995).  In a study using an He tracer, it was dem-
onstrated that the ROI of a  sparging well in both groundwater and the vadose
zone were comparable (Javanmardian et al.  1995).  Helium tracer has also
been used to monitor the vapor capture capability of a combined IAS/S VE
system (Johnson et al. 1995).
   Measurements of pressure distributions in both saturated and unsaturated
zones have been  used as possible indicators of ROI. Pressures measured
within the saturated zone are partially due to water table mounding and may
result in overestimates of ROI (Acomb et al. 1995). Due to the nature of
pressure propagation from  an air source, the use of unsaturated zone pres-
sures may also result in overestimates of the ROI.
   Depleted groundwater DO content is often found in the vicinity of hydro-
carbon spills as a result of biodegradation activity by naturally-occurring
aerobic bacteria. After air sparging is  initiated., DO levels may rise substan-
tially at several monitoring wells. Oxygen transport may be occurring as a
result of a combination of advection, dispersion, and diffusion. Increases in
groundwater DO resulting from sparging activity is the most popular
                                5.141
-------
Groundwater Treatment Systems


indicator of air sparging well ROI and is a more direct measurement of
whether oxygen is being supplied to the bacteria.
   Figure 5.28 shows experimentally-measured ROI values from 37 sites
(Marley and Bruell 1995). ROI values greater than 12 m (40 ft) were re-
ported at only a limited number of sites; note that the testing procedures and
analyses of data from sites exhibiting high ROI values were considered to be
of questionable reliability. The majority of sites reported ROIs between 3
arid $ m (10 and 26 ft).
5.5.3 Process-Flow Diagram
   The biosparging process is depicted schematically in Figure 5.29. The
components of a typical biosparging process include the wells, a manifold,
and a compressor system.

5.5.4 Process Modification

5.5.4.1 Air Injection Without Air Extraction
   IAS systems are generally used in conjunction with SVE to prevent the
migration of contaminated IAS gases. However, use of IAS without an SVE
system has been documented at one research site (Beausoleil et al. 1993).
Here, a low-flow (i.e., 4.25 m3/hr [2.5 scfm]) IAS system used indigenous
microbial populations in the adjacent vadose zone to degrade all contami-
nants. No significant downward or lateral dispersion of BTEX in the aquifer
was observed, and no BTEX was volatilized to the atmosphere during the
study period.
                                          •i
   IAS systems can be designed and operated to optimize oxygen mass
transfer into an aquifer or contaminant volatilization from an aquifer.  Selec-
tion of the operational mode often depends on the relative biodegradability
of the contaminant. In laboratory studies where an IAS system was operated
primarily to add oxygen to a model aquifer, there was an optimum air injec-
tion rate above which further increases caused decreases in the rate of oxy-
gen mass transfer (Rutherford and Johnson 1995).  Oxygen mass transfer
appeared to be affected by both air injection rate and groundwater flow.At
low air injection rates, increases in air flow resulted in a higher density of air
channels.  A higher density of air channels, which constitutes an increase in
interfacial area, increased mass transfer rates. At low injection rates, dye
studies demonstrated that water still would flow through the zone of air
travel. At air flows above the optimum flow rate, further increases in air
flow rates did not cause further increases in air channel densities but,
                                5.142
-------
                                                                            Figure 5.28
                                                              IAS ROIs Found at Numerous Sites
Oi

§
                                     Oto2m     2 to 3m     3 to 4m
4to5m     5to6m    6to7m     7to8m     StolOm     >10m

       Radius of Influence
      Total Sites = 37

      Source: API Publication 4609, In Situ Air Sparging: Evaluation of Petroleum Industry Sites and Considerations for Applicability, Design and Operation, 1995. Reprinted courtesy of the American Petroleum
      Institute.                ~             ~                           ~                      ~     .......                       -
                                                                                              o
                                                                                              Q
-------
Groundwater Treatment Systems
                                Figure 5.29
          Biosparging Process-Flow Diagram Showing Biosparging
         Well Operating in Parallel with a Soil Vapor Extraction Well
        Vadose Zone
      (Unsaturated Zone)
      In Situ Air Sparging Well
\
  Idealized Air Channel
                                                  Saturated Zone


                                        Hydrocarbon "Smear Zone"

                                       Water and Soil Containing Hydrocarbons
Source: API Publication 4609, In Situ Air Sparging: Evaluation of Petroleum industry Sites and Considerations tor Applicability,
Design and Operation, 1995. Reprinted courtesy of tho American Petroleum Institute.
instead, caused reductions in relative permeability to the flow of water.  In
this case, water was forced to flow around the zone of air travel, and a de-
crease of mass transfer of oxygen into the water was observed.  It is expected
that the same qualitative trends will be observed in a field setting; however,
specific correlations between laboratory results and field studies do not cur-
rently exist.
                                             !
5.5.4.2  Pulsed Operation
   Generally, an SVE system is used for several weeks to remove any re-
sidual hydrocarbons above the water table prior to initiating IAS activities.
Figure 5.30 (top) shows an idealized response of hydrocarbon content in
SVE stack gas for a system where SVE was initially conducted alone fol-
lowed by continuous combined operation of SVE and IAS (Marley and
                                  5.144
-------
                                                             Chapter 5
 Bruell 1995). During SVE-only operation, the decrease in effluent hydrocar-
 bon content follows a typical first-order decay. When IAS is initiated, hy-
 drocarbon content increases as VOCs trapped beneath the water table are
 mobilized by the migrating air.  However, a gradual first-order decay ito as-
 ymptotic levels is again observed. It is theorized that this response to IAS in
 fine-grained soils is because the air-filled channels constitute select preferen-
 tial pathways within the aquifer that directly Impact limited portions of the
 aquifer.  Therefore, contaminants contained ini soils not directly impacted by
 the air-filled channels must diffuse or advect to these areas (Wilson, Norris,
 and Clarke 1996a).
   Cycling of an IAS system on and off for periods of time ranging from 12
 hours to several days is known as "pulsed" operation. An idealized response
 of hydrocarbon content in stack gas for an IAS system which incorporates a
 continuous SVE system combined with pulsed IAS well operation is shown
 in Figure 5.30 (bottom)(Marley and Bruell 19135). Field data suggest that
 pulsed sparging greatly enhances groundwater mixing, which is necessary to
 overcome the diffusion limitations of sparging caused by air channeling
 (Clayton, Brown, and Bass  1995). Additionally, it is theorized that trapped
 air can be induced to dissolve by the action of sparging.  The enhancement
 of dispersion resulting from pulsed IAS has been modeled (Wilson, Norris,
 and Clarke 1996b).
   It is also theorized that selection of a pulse frequency depends on the
 desired mechanism of contaminant removal (i.e., volatilization versus bio-
 degradation) (Rutherford and Johnson  1995). Trapped air remains after
 injection stops and continues to supply oxygen for biodegradation, while
 volatilization ceases when air injection stops.

 5.5.4.3  In-Well Aeration Systems
   In-well aeration is the process of injecting & gas, usually air, into a well,
 resulting in an in-well airlift pump effect. In a typical application, illustrated
 in Figure 5.31, air is injected into the bottom of a well. The air travels up-
 ward, stripping volatiles  and aerating the water. This upward movement of
 air results in an airlift pump effect causing water to flow into the well from
 the deeper screened portion of the well and out of the well from the shal-
 lower screened portion establishing a circulation pattern within the aquifer.
If hydrogeologic conditions allow, in-well aeration creates a circulation cell
that treats and aerates the water as it passes through the well. The airstream
may also provide oxygen for biodegradation in both the saturated and unsat-
urated zones.
                                5.145
-------
Groundwater Treatment Systems
                                  Figure 5.30
   Idealized Hydrocarbon Removal Data Resulting from a Continuously
          Operated System (Top) vs. A Pulsed Operation (Bottom)
                               60      80     100    120

                                   Time Period (days)
                 20
                                   Time Period (days)
Source: API Publication 4609. In Situ Air Sparging: Evaluation of Petroleum Industry Sites and Considerations for Applicability,
Design and Operation, 1995. Reprinled courtesy of the American Petroleum Institute.
                                                                      ..•in;
                                    5.146
-------
                                                              Chapter 5
                               Figure 5.31
                     Typical In-Well Aeration System
             Air Compressor
               or Blower
 Reprinted with permission from Air Sparging Site Remediation, R.E. Hinchee, "Air Sparging State of the Art," 1994.
 Copyright CRC Press, Boca Raton, Florida.
   The operating principle of in-well aeration is relatively straightforward.
Air is used to strip and/or oxygenate water by establishing an "in-well
pump-and-treat" system. The advantage of this approach, compared to tradi-
tional pump-and-treat systems, is that it avoids  removing water from the
aquifer for aboveground treatment.

5.5.4,4 Sparging Gate-Wells,Trenches,, and Curtains
   A number of additional air sparging techniques have been used to contain
and remediate VOC-contaminated groundwater, including the application of
sparging gate-wells and trenches or "curtains" (Pankow, Johnson, and
Cherry 1993; Marley et al. 1994).  The concepts of sparging gate-wells and
trenches are illustrated in Figure 5.32.        ;-.....-..
   The sparging gate-well uses hydraulic barriers to direct contaminated
groundwater flow through a treatment zone. The sparging trench is con-
structed perpendicular to the contaminated groundwater plume flow direc-
tion. The contaminants in the groundwater may be remediated while passing
                                 5.147
-------
Groundwater Treatment Systems

Figure 5.
Sparging Gate-Well an
X- 	 N.
1 Upgradient
Zone of
Remediation
/ Flow \
| * Jr Jr |
a . 1
CutoffWall liiiliiia CutoffWall
Sparge
Gate Well
/\
\
T •*
Downgradient
Zone of
Remediation
Source: Pankow, Johnson, arid Cherry 1993

32
d Trench Systems
/ \
/ Upgradient
/ Zone of
/ Remediation
Flow
\ I
\ i I I
Sparge Trench
I
I I I
Downgradient
Zone of
Remediation
1 ' ' 1 L
11 ' ''''
through the sparging gate-well or trench, via volatilization, biodegradation,
or other physical/chemical processes.
                                           j
   A sparging curtain resembles a sparging trench in that it is installed per-
pendicular to the flow of the contaminated grouridwater plume.  However,
vertical sparging wells are generally spaced equally along the length of the
curtain to emulate the performance of the sparging trench.
   The main considerations in the design of a flow-through sparging treat-
ment system are to ensure that:                                      '
        •  groundwater will flow through, not around the system;
        •  target VOCs can be removed to predetermined clean-up lev-
           els; and
        •  that relatively uniform air flow is provided over the entire system.
                                5.148
                                           i!"
-------
                                                             Chapter 5
   To achieve these criteria, the system design parameters (e.g., trench di-
 mensions and fill material properties) and operating parameters (air injection
 flow rates, sparging manifold radius, port size, and manifold pipe lengths)
 must be established. In many cases, an associated passive or active SVE
 system may be required in conjunction with these systems.

 5.5.4.5 Pure Oxygen                                        :
   Delivery of oxygen is often the rate-limiting step controlling biodegrada-
 tion. Air contains approximately 20% oxygen by volume. When using air as
 a sparging gas, the maximum DO concentration that may be obtained within
 an aquifer is  8 mg/L, based on partitioning described by Henry's Law at
 typical groundwater temperatures. Soils with low permeabilities may se-
 verely restrict the rate at which sparge gas, and therefore oxygen, can be
 introduced into an aquifer formation. When sparge gas flows are restricted
 to less than 3.4/m3/hr (2 scfm), the use of pure oxygen as a sparging gas
 should be considered. With 100% oxygen as a sparge gas, the resulting DO
 level is 40 mg/L.  Therefore, the amount of DO delivered and rate of biodeg-
 radation can be as much as five times faster than air when using pure oxygen
 as a sparge gas. This benefit may offset the lower sparge gas flow rates.
 Additionally, higher DO concentrations result in greater concentration gradi-
 ents and higher rates of mass transfer to areas not directly contacted by
 sparging gas. Furthermore, in biosparging applications, the injection of pure
 oxygen can provide a means of effective sparging in geologic conditions not
 suited to traditional air sparging,
                                         i
   As an example, a biosparging pilot study Avas conducted for groundwater
 and soils contaminated with semivolatile organic compounds at a facility in
 Texas used to store wastes and wastewaters containing elevated levels of
 nitroaromatic and aromatic compounds. Site operations led to release of
 these compounds into the groundwater which was located in a confined
 sandy aquifer underlying a clay aquiclude. These site conditions prevented
 using a cost-effective SVE system for sparging gas capture. The pilot study
 demonstrated the successful application of pure oxygen into the aquifer.  At
 an oxygen flow rate of less than 1.7 m3/hr (1  scfm), a zone of influence in
excess of 9.1 m (30 ft) was observed.

5.5.4.6 Addition of Methane to Sparge Air
   Industrial solvents such as TCE, PCE, 1,1,1-trichloroethane (TCA), car-
bon tetrachloride,  and chloroform are common environmental pollutants.
Bacteria found in groundwater can biologically transform these chlorinated
compounds via several pathways. Indigenous methanotrophic organisms can

                                5.149
-------
Groundwater Treatment Systems
be biostimulated with the addition of methane as an electron donor and oxy-
gen as an electron acceptor.  Methanotrophs produce the enzyme methane
monooxygenase (MMO), which initiates the first step of methane oxidation
when methane is used as the sole carbon source for energy and growth
(Semprini et al. 1990). Under aerobic conditions", MMO can epoxidize alk-
enes. Aerobic TCE oxidation can be accomplished by mixed cultures of
methanotrophic and heterotrophic organisms. TCE oxidation first involves
the epoxidation of TCE by methanotrophs, an abiotic hydrolysis of the ep-
oxide to nonvolatile products, followed by heterotrophic degradation of the
products to CO2, chloride, and water (Semprini et al. 1991).
   Laboratory studies have shown that this process can be conducted aerobi-
cally with an air phase that contains as little as 0.6% natural gas (i.e., meth-
ane) by volume (Wilson and Wilson 1985). In microcosms, optimum
gas-phase oxygen and methane  content to promote TCE degradation were
between 7.7% to 8.7% and 1.7% to 2.7%, respectively, which correspond to
aqueous concentrations of 3.2 to 3.7 mg/L and 0.'4 to 0.6 mg/L, respectively
(Kane, Fischer, and Wilson 1996).
   The addition of methane to simulate the growth of methanotrophs in the
field has been investigated at the US DOE Savannah River Site, South Caro-
lina  (DOE 1991).  '     "   '	"'	'	  ,	\	".	
   Methane can be added to sparging air by piping a methane line equipped
with a check valve, isolation valves, and flow meter to a sparging well. The
methane supply must produce sufficient line pressure to overcome pressure
resulting from the air sparging compressor or blower.  Methane content of
sparging air should be maintained below the lower explosive limit (LEL) of
5% to prevent explosive conditions. The methane addition must occur only
when a sparging blower is operating; this can be accomplished with an inter-
locked valve rated for natural gas service.
                          "... •'"  ',!!•, '','," •"" " ,,i,    ','"' ,,        "'•'           , il i •'
5.5,5 Pretreafment Processes
   Pretreatment processes for IAS systems are related  to the supply of air or
oxygen. There are a number of methods of supplying  oxygen  for
biosparging applications including liquid oxygen and oxygen concentrators.
Oxygen concentrators use a dual-bed molecular sieve design to remove the
nitrogen (and other non-oxygen components of air) from a compressed air
stream. For example, a model AS 4000 oxygen concentrator can generate
11.3 m3/hr(67cfm) of 90%  to95%pureoxygenat310.5kPa(45psi). The
compressed oxygen from the concentrator is generally fed into a steel oxy-
gen receiver tank, which can then be plumbed to the oxygen flow control
                                5.150
-------
                                                            Ghapt©r5


 manifold. The oxygen flow control manifold system generally consists of
 pressure regulators, electric ball valves, electronic pressure and temperature
 transducers, and mass flow meters that are used to monitor and direct the
 oxygen flow to each of the biosparging manifold headers.
   For air injection, only oil-less compressors or compressors outfitted with
 oil removal devices should be used to introduce contaminant-free air into the
 aquifer (see Section 5.5.11 for compressor selection).

 5.5.6 Posttreatment Processes
   Depending on site-specific conditions it may be necessary to capture and
 treat the gas emanating from the saturated zone as a result of the biosparging
 process. Ideally, if it can be demonstrated that the emanating gas is being
 degraded in the vadose zone and poses no significant risk, no collection or
 further treatment will be required. If collection is required, an SVE system
 is typically incorporated into the biosparging system design. Treatment of
 contaminants in the SVE offgas may consist of one or a combination of the
 following treatment technologies:
        • activated or impregnated carbon sorption;
        • thermal or catalytic oxidation;
        • non-carbon based sorption (e.g., Zeolite);
        • biofiltration; and
        • membrane/separation technologies.
   For thermal/catalytic options, the use of internal combustion engines
 (ICE) has shown promise   However, in some states a catalytic  converter on
 an engine exhaust is considered an air-control device, and thus  requires a
 permit. If target  destruction efficiency can be achieved before the exhaust
 passes through the catalytic converter then a permit is generally waived. As
 an additional control, the S VE/ICE system must shut itself off if the ICE
 goes down. This is generally achieved by using the vacuum created by the
 engine for the SVE. If additional blowers/compressors are used, then there
 must be a control system which shuts off these blowers/compressors if the
 ICE goes down.
   The preferred technology for offgas treatment will be based on an
 evaluation of the most cost-effective process for the contaminants of
concern.  See Section 6.2 of this monograph for a detailed discussion of
biofiltration for  offgas treatment. See Vapor Extraction and Air
Sparging, Holbrook et al. (1998).
                                5.151
-------
ITS!'
       Groundwater Treatment Systems
       5.5.7  Process Instrumentafibn and Control
         IAS system instrumentation and controls monitor and regulate the flow of
       air (oxygen) at the source and at the individual injection wells.
                                     '    ".   "'   : , ''"!'  !     .    '      '    '  I.1'"
       5.5.7.1 Wells
         In some cases, sparging wells can be driven from the surface to save in-
       stallation costs. Limited experience indicates that driven points perform as
       well as drilled wells (Droste et al. 1994). However, the probability of short-
       circuiting along a driven well casing is higher than with a drilled well instal-
       lation. In addition, special drive point techniques may be needed in
       fine-grained soils (clays) due to the potential for well screen smearing.
         For drilled wells, the average grain size of the filter pack should be as
       close to the native soils as practical.  Filter packs that have an average grain
       size larger than the native geologic materials may be more permeable than
       the native soil. While a highly-permeable filter pack is an advantage in wells
       constructed for other uses (e.g., monitoring or extraction), a filter pack that
       has a higher permeability than the surrounding formation enables the applied
       air to short-circuit up the borehole. A bentonite seal is used to seal the bore-
       hole and prevent short-circuiting of the air supplied. This seal should be
       placed from approximately 0.3 m (1 ft) above the IAS well screen filter pack
       to approximately 0.3 m (1 ft) above the seasonally high water table level and
       then hydrated. The annular space above the bentonite seal should be filled
       with a 5% (by weight) bentonite/cement grout mixi
         Well-screen slot size should complement the filter pack design.  Since air
       readily passes through well screens, a small slot size usually is sufficient;
       underestimating the slot size  (by a small margin) relative to the filter pack is
       usually acceptable. In most cases, a 0.5-mm (6.020 in.) or 0.25-mm (0.010"
       in.) slot size is used.
         A relatively short length of screen, such as 6.3 to  1 m (1 to 3 ft) is suffi-
       cient.  Shorter screen lengths are not recommended.  The well screen typi-
       cally is flush-threaded slotted Schedule 40 or 80 PVC or CPVC pipe. A
       schematic of a typical drilled air sparging well is shown in Figure 5.33.
         It is recommended that wells be developed to minimize accumulation of
       fines in the screened section and/or filterpack. Air sparging wells should be
       developed prior to operation because pulsed operation produces an effect
       essentially the same as well development, but without the removal of accu-
       mulated fines. In some cases, the reverse gradient created between pulses is
       sufficient to cause the fines to migrate into the well! and the filterpack, causr
       ing clogging problems.

                                       5.152
-------
                                                                 Chapter 5
                                Figure 5.33
               Typical Sparging Well with Grouting and Seal
               Surface Grade
                    Filter Pack
                                            in. Schedule-40 PVC Riser
                                           1 in. Schedule-40 PVC
                                           Well Screen (0.01 in. Slot)
Source: API Publication 4609, In Situ Air Sparging: Evaluation of Petroleum Industry Sites and Considerations forApplicabiSty,
Design and Operation, 1995. Reprinted courtesy of the American Petroleum Institute.
5.5.7.2  Manifold
   The construction of an IAS manifold typically includes the following
components:  check valve, throttle valve, manifold piping or hose,
quick-connect couplers, and plugs and sampling port(s) at the wellhead.
   The manifold is typically buried underground below the frost level.  If it
is within the frost zone, it may need to be protected from frost with insula-
tion and/or heat tape, and flexible connections may be needed to prevent
damage from frost heaving.  Once the main manifold run has been installed
in the vicinity of a group of wells, hard piping or a high-pressure air hose
equipped with couplers and plugs can be used for attachment to the well.
   Check valves are recommended at the well (between each well and the
manifold) to prevent temporary high pressure in the screened interval of the
aquifer from forcing ah- and water back into the manifold after the IAS
                                  5.153
-------
Groundwater Treatment Systems
system is shut down. A throttling valve should also be installed at each well
to allow the isolation of the well from the system or for adjustments to the
well air flow rate; gate valves and globe valves are effective throttle valves.
   A manual or automatic pressure relief valve should be installed immedi-
ately downstream of the air compressor outlet. This valve exhausts excess
air from the manifold to either the atmosphere or the compressor air inlet
and acts to prevent excessive pressure from damaging the manifold or frac-
turing the aquifer soils in the event of a system blockage.
                 1   .    '   '      ;'";::/*;/,!. ..•! !'"":  : "   •;/'•.  .'  , "    i,;,"
5.5.7.3 Compressor System
   The selection of a suitable air compressor is typically based on the results
of an in situ pilot study. The results of the pilot study are used to determine
the optimal pressure and flow for a well installed within a specific geologic
setting. The pressure capacity and flow rate of the air compressor should be
designed based upon the maximum expected pressure and flow for any one
group of wells and must consider manifold system head losses.
   Compression of air can generate a significant amount of heat and noise.
As part of the design, the air compressor exhaust temperature should be
calculated based on  manufacturer's data. Piping and manifold materials
must be compatible  with compressor discharge temperature and pressure.
   The process of air compression can cause production of moisture in the
air compressor and/or manifold line.  In the winter months, precipitation in
the manifold can freeze, restricting or blocking the flow from the compres-
sor.  Heat tracing can be used to winterize the piping/manifold. A receiver
(air tank) with a manual or automatic drain to remove condensate from the
receiver is recommended.  For larger systems, moisture removal equipment
may be installed upstream of the air inlet to the compressed air source.
Common air compressor types are described in Section 5.5.11.

5.5.8  Process and Instrumentation Diagram
   Figure 5.34 presents a typical P&I diagram for a biosparging application
using" a concentrator to provide pure oxygen.  The concentrator can be elimi-
nated when using air injection or replaced (including the compressor) with
liquid oxygen tanks.
     ,,  ,  '  ,      ,    |       J ' "  4  •] , "  i'"      | ji  .  '  n    „        	i  'iiJ, '
5.5.9 Sample Calculations
   The use of excessive sparging pressures may cause soil fractures which
short-circuit the air injected and compromise the effectiveness of the IAS.
                                5.154
-------
                                                         Figure 5.34
                             Process and Instrumentation Diagram for a Typical Air Sparging System
Oi
                   Ambient Air
                   Intake Filter
                     n-
                                 Air
Oil-Less
     Pressure
      Gauge
-a—n-
                                                   Pressure
                                                   Relief
                                                   Valve
                                                                      Gate
                                                                      Valve
                                                                    -HgH
                                            Flow
                                            Meter
                                       Solenoid
                                       Valve
                                            o-
                                                  IAS
                                                  Wells
                                                                      Gate
                                                                      Valve
                                                                               -Q
                                                                   IAS
                                                                  Wells
                                            Flow
                                            Meter
                                       Solenoid
                                       Valve
                                                                                                                             9
                                                                                                                             Q
                                                                                                                             TJ
                                                                                                                             cn
-------
Groundwater Treatment Systems
                                               ij               '      I "
The maximum air pressure that can be safely applied without producing soil
fractures can be calculated based on an estimate of the overlaying pressure
from the weight of the soil and water column above the top of the IAS well
screen.  Then, as a safety factor, only 60% to 80% of the overlying pressure
should be applied to avoid soil rupture. This calculation is only a first ap-
proximation estimate because field conditions may vary (Marley and Bruell
1995). A sample calculation follows:
           Assume that the soil is uricpnsolidated medium-size sand with a
           33% porosity (0.33) and a particle specific gravity of 2.65.  The
           water table is located 5.49 m (18 ft) below the soil surface and
           the top of the sparging well screen is located 1.52 m (5 ft) below
           the water table. (Note:  all calculations use gauge pressure.)

In SI units
                                               i
                                 ...        ii              .       , „ .

      water pressure = (1.52m) • 0.33 •        =       or 4-92 kPa   (5.33)
soil pressure = (5.49 +1.52) • 2.65 • (1 - 0.33) •        =
                                             m,         m
or 122.10 kPa
                                               I
                                ..     	    .  j              ,       . .
                total pressure = 4.92 + 122.10 = 127.02 kPa          (5.35)
                                   	   •••.]'       -.      I i      i •.••.
                Maximum pressure to avoid soil fractures =          (536)
                0.6 • 127.02 = 76.2 kPa to 0.8 • 127.02 = 101.6 kPa
                                    ''   '   ' !;  |     '   :"       '   ';  '"''
In English units
                                ••   ;      :  • • :  !    ;,   '- •'            i .'•
           water pressure = 5 ft • 0.33'• 62.4 Ibs/ ft3 = 103 Ibs/ ft2     (5.37)

soil pressure = (18 + 5) • 2.65 • (1 - 0.33) • 62.4  Ibs / ft3 = 2,548 Ibs / ft2  (5.38)

total pressure = 103 + 2,548 = 2,651 Ibs / ft2 • 1 ft2 /144 in2 = 18.4 psi  (5.39)

                maximum pressure to avoid soil fractures =
                                                                 (5.40)
                0.6«18.4 = 11.0psito0.8«18.4 = 14.7psi
                                 5.156
-------
                                                            Chapters


   When sparging is initiated, the pressure required to establish flow is al-
 ways greater than the pressure required to maintain air flow. This is because
 sparging air will displace a portion of the overlying water from the sparge
 well and soil column, and subsequent air flow will require lower pressures.
 Over-pressure can be defined as the pressure in excess of that required to
 overcome hydrostatic head at the top of the \vell screen, which is measured
 once flow has been established.
   Assuming an operating over-pressure of 34.5 kPa (5.0 psi) for 8.49 m3/hr
 (5 cfm) of air flow in a medium  sand, the pressure required to maintain flow
 in the system can be estimated as follows:
 In SI units

                              9 81  kN   14 9 kN
      hydrostatic head = 1.52 m»    .   = -    ,    or 14.9 kPa      (5.41)
                                m        m

 pressure required to maintain flow = 34.5 kPa: +14.9 kPa = 49.4 kPa   (5.42)

 In English units

        hydrostatic head = 5 ft • 62.4 Ibs / ft3.1 ft2 /144 in2 = 2.2 psi  (5.43)

        pressure required to maintain flow = 5.0 psi + 2.2 psi = 7.2 psi (5.45)

   This pressure is much less than the maximum pressure to avoid soil frac-
 tures. Pressure required to establish flow may be somewhat higher (i.e., 10
 to 20 kPa[ 1.5 to 2.9 psi]).
   Dissolved oxygen must be supplied in sufficient quantities to satisfy sto-
ichiometric requirements for the aerobic biodegradation of contaminants.  In
the case of gasoline, which contains hundreds of hydrocarbons, benzene can
be used as a surrogate for approximate calculation purposes. The oxygen
requirement for the complete aerobic biodegradation and mineralization of
benzene to carbon dioxide and water can be calculated as follows:

                C6H6+7.502-»6C02+3H20                     (5.45)

                78kgC6H6 + 240kgO2 -»264kgCO2 + 54kgH2O      (5.46)
                          =3.1kg02/kgC6H6                     (5.47)
                      66
                                5.157
-------
Groundwater Treatment Systems
   Therefore, every 1 kg of benzene will require 3.1 kg of DO for biodegra-
dation. The time required for the delivery of oxygen is often the
rate-limiting step that controls the fate of biodegradation.  It should be noted
that approximately 1% of the available oxygen from injected air is actually
used in biodegradation with 99% not transferred to the groundwater during
sparging (Newman et al. 1994).
        ,    '         ,        "l" ' '"    ' ': 'I1'1 „"'„,  •!  ' |
     „,'             '  ' , ,   i!'   : '",' "f s:   '  ,, •',."  , if ' i
5.5.10  Safety Requirements
                                             ,i
   IAS system technology health and safety considerations are primarily
mechanical, electrical, and chemical.
   Air compressors often have the capability of producing in excess of 621
kPa (90 psi) air pressure. This pressure can easily cause injury to personnel
and equipment damage if not properly controlled. IAS systems should be
designed using only suitable materials that are rated for expected operating
pressures and should have properly operating pressure relief valves located
at appropriate locations. Proper grouting is essential to avoid the launching
of pressurized wells from the ground^putting personnel in the vicinity at
risk; this phenomenon has been observed at one site. Air flow and pressure
Should be gradually applied to avoid pressure surges that could lead to soil
fupture (see Section 5.5.9 for calculation of maximum sparging pressures).
   Electrical hazards are always present when using electrical equipment
under field conditions.  All equipment should be wired by qualified techni-
cians according to local electrical codes. Proper grounding of all equipment,
such as compressors, vacuum pumps, catalytic converters, monitoring equip-
ment, etc., is required to prevent the possibility of electrocution. Equipment
that is outfitted with automatic  ground fault protection should be used when-
ever available.
   Movement of chemical contaminants resulting from IAS activity must be
strictly controlled. Proper IAS/SVE system design, layout, installation and
operation are required to prevent spreading of contaminants, migration of
contaminants into buildings, or fugitive emissions to the atmosphere. All
NAPLs should be removed prior to  the startup of an IAS system. A properly
designed S VE system with vapor treatment is generally operated prior to the
startup of the IAS system. A comprehensive site survey should be conducted
to identify any buried utilities that are present that could serve as possible
conduits for gas migration.  If present, such utilities should be isolated from
the IAS/SVE system via interceptor wells or trenches.
                  1 •' • • ''•    '  "       '"' I1!:1',:' 1;!i  "'  : '  j         ',: ,     ''''    ,| '• h|l"!ii ',„"":
                                 5.158
-------
                                                             Chapters
   If flammable gases are being vented from the soil by the S VE system, it is
 possible to approach LELs in the pipe network without dilution. Care
 should be taken to provide dilution valving at the wellheads so vapor mix-
 tures within the piping above the surface is well below the LEL,

 5.5.11  Specification Development
   The following information on the selection of air compressors and piping
 materials is provided only as an initial guide.  To prevent injury to personnel
 and damage to equipment, always consult manufacturers concerning the
proper application of their products.
   A pilot test using a portable air compressor is generally conducted to
 determine site-specific pressure and flows. Atypical over-pressure and flows
 are 34.5 kPa (5 psi) and 1.7 mVhr to 17m3/hr (1 to 10 cfm), respectively (see
 Section 5.5.9 for calculation  of maximum pressure to avoid soil rupture and
Table 5.32 for sizing information).                               i
   Air compressors are typically quite noisy and if they are to be near resi-
dential areas they should be located in enclosures outfitted with noise abate-
ment equipment and insulation.  Air compressors can also generate signifi-
cant heat; therefore, it is necessary to use piping material that is compatible
with expected discharge pressures and temperatures. This is often accom-
plished by using several lengths of metal piping to allow for heat transfer
and system cooling before coupling to piping made of polymeric materials.
   Air compression leads to the production of water in the compressor receiver
tank and manifold lines. Therefore, air tanks should be drained regularly to
prevent condensate buildup. It may be necessairy to winterize the compressor
system and heat trace exposed piping to avoid system icing and blockage.
   Only continuous-duty,  oil-less air compressors should be used. Most
commonly available compressors do not meet these specifications. Oil-less
compressors are necessary to avoid introducing hydrocarbons to the aquifer.
Rotary-vane pumps or regenerative blowers can be used only when low air
pressures (i.e., up to 69 kPa [10 psig]) are required. Rotary-lobe blowers can
be used for sparging sites  when air pressures do not exceed 103.5 kPa (15
psig).  Reciprocating compressors are generally required for IAS pressures
in excess of 103.5 kPa (15 psig).  Reciprocating compressors can generally
achieve over 621 kPa (90 psig) pressures and often use Teflon® components
to avoid the use of lubricants. Other types of compressors (i.e., rotary screw)
can be used if provisions are made to keep hydrocarbon lubricants from
entering the air stream.
                                5.159
-------
                                        Groundwater Treatment Systems
.;;• . ,; ; • . •
' ' 	 ...table 5.32*
Typical Air Sparging System Design and
Operational Parameter Specifications
Parameter and range
Screen length
0.15 to 3.05 m
(0.5 to 10 ft)
Well diameter
2.54 to 10. 16 cm
(0.35 to 18.2 psi)
Overpressure
2.41 to 125.67 kPa
(0.35 to 18.2 psi)
Well screen depth
below water table
0.61 to 8.08 m
(2 to 26.5 ft)
In situ sparging flow
rate
2.21 to 67.96 m3/hr
(1.3 to 40 cfm)
In situ sparging
pressure
24.11 to 172.25 kPa
(3.5 to 25 psi)
(SVEROI)
(IAS ROD ratio
0.16 to 7.42
Source: Martey and Bruell
Most often used
value
(no. of sites)
0.61 m
(2ft)
16 sites
5.08 cm
(2 in.)
17 sites
2.41 to 34.45 kPa
(0.35 to 5 psi)
14 sites
1.52 to 3.05 m
(5 to 10 ft)
10 sites
2.21 to 8.50 m3/hr
(J. 3 to 5 cfm)
16 sites
34.45 to 68.90 kPa
(5 to 10 psi)
17 sites
1 to2
12 sites
1995
	 	 	 i..
Second-most often
used value
(no. of sites)
0.91m
(3 ft)
8 sites
10.16 cm
(4 in.)
Ir 	 ' "'
7 sites
34.5 to 68.90 kPa
(5 to 10 psi)
9 sites
3.05 to 4.57 m
(10 to 15 ft)
8 sites
8.50 to 1 6.99 m3te
(5 to 10 cfm)
9 sites
68.90 to 103.35 kPa
(10 to 15 psi)
8 sites
o.ietoi
6 sites
H-P •,.!" !"!! 	 ,„. 	 :'.',.
" • ' 	
Third-most often used Total
value number
(no. of sites) of sites
1.52m
(5ft)
7 sites «
2.54cm
(1 in.)
5 sites 37
68.90 to 103.35 kPa
(10 to 15 psi)
5 sites 31
0.61 to 1.52 m
(2 to 5 ft)
6 sites 31
25.48 to 33.98 rn3/hr
(15 to 20 cfm)
5 sites 49
137.80 to 172.X) kPa
(20 to 25 psi)
6 sites 40
3to4
3 sites 26
• . :. . r t 1 ., 
-------
                                                            Chapter 5
   In all cases, compressor air inlets should be located to avoid the introduc-
tion of airborne contaminants. Therefore, inlets should not be located within
service garages or in close proximity to S VE stacks.
   The selection of piping materials depends upon the site-specific operating
conditions, including gas pressures, gas temperatures, ambient temperatures,
potential for shock, and chemical compatibility. Pipes that are 2.5 to 5 cm (1
to 2 in.) in diameter are generally more than sufficient to carry air flow rates
of 3.4 to 17 m3/hr (2 to 10 scfm). While smaller pipes may be used, head
losses associated with smaller pipe diameters and smaller fittings should be
carefully considered.
   Technically, Schedule 40 PVC is often rated for operating temperatures
up to 60°C (140°F) and nonshock operating pressures for liquids in excess of
690 kPa (100 psig). However, air is a compressible fluid, and striking a
blow to any pressurized PVC pipe could potentially cause the pipe to burst,
possibly resulting in injury. Therefore, when a sparging air compressor dis-
charge is at pressures greater than 276 kPa (40 psig) or at temperatures
greater than 38°C (100°F), the use of steel pipe that has been protected from
corrosion or galvanized steel piping is recommended. When a sparging air
compressor discharge is  at pressures less than 276 kPa (40 psig) and at tem-
peratures less than 38°C (100°F), the use of Sichedule 40 or 80 PVC pipe
should be investigated. Threaded fittings sealed with Teflon® tape are pre-
ferred over glued fittings on pressurized IAS plastic pipe lines to prevent the
introduction of solvents associated with glues into the groundwater system.
   Well screens used for IAS systems generally use 0.25- to 0.5-mm (0.01-
to 0.02-in.) slots. Microporous bubblers have also been used at IAS sites
(Kerfoot 1995). Typical design parameters from the API-IAS Database are
presented in Table 5.32.                    !

5.5.12 CostDatq
   Based on pilot- and full-scale applications of this technology, the total
cost of source area biosparging generally ranges between $13 and $55 per
m3 ($10 and $42.30 per yd3) of soil remediated. The upper range of general
cost can increase to $80  per m3 ($61.50 per yd3) when an SVE vapor capture/
treatment system is required. On larger sites (greater than 10,000 m3 [13,080
yd3] of impacted soil), costs of less than $13 per m3 ($10 per yd3) can be
achieved. On small sites (less than a few hundred m3of impacted sojl), costs
in excess of $80 per m3 ($61.50 per. yd3) are common because of the fixed
costs involved in project planning, permitting,  drilling wells, and installing
the required system equipment.  Table 5.33 provides typical costs for
                                 5,161
-------
ill   i   .;




Pilot Testing & System Design
Full-Scale Installation
Oi
.rfo *;

Posttreatment Cost

Total Project Cost'
Table 5.33
Typical Full-Scale Sparging System
Cost Element
Conduct Pilot Testing
Data Analysis/Full-Scale Design
Install SVE Blower and Off-Gas Treatment Systems (300 scfin)
In situ Sparging System (150 scftn)
Electrical
Manifold & System Installation
Start-up Report
Annual O&M Expenses
Posttreatment Soil and Groundwater Sampling
Site Closure Report


Cost°-b
Unit Cost ($)
20,000
20,000
100,000
20,000
10,000
125,000
8,000
32,500
12,000
3,000
333 (24.6)d


No. of Units
Lump Sum,
Lump Sum
Lump Sum
Lump Sum
Lump Sum
Lump Sum
Lump Sum
2 Years
Lump Sum
Lump Sum
11,500m3


Cost ($)
20,000
20,000
100,000
20,000
10,000
125,000
8,000
65,000
12,000
3,000
383,000
"Costs are for large sparging site with >11,000 m3 of impacted soil.
System includes an SVE component with off-gas controls. Significant SVE component costs have been separately labelled.
°CERCLA sites will generally cost up to three times the projected expenditures due mainly to the additional deliverables and analytical and QA/QC measures required.
Cost in parentheses is without major SVE components.
:- ' '
"\
dwater Treatment .

CO


» -


-------
                                                             Chapter 5
planning, design, installing, and operating a full-scale sparging system for a
large site (11,000 m3 [14,387 yd3]) impacted by petroleum hydrocarbons.

5.5.13  Design Validation
   Over the first few months of operation, the performance of the IAS sys-
tem should be monitored to validate the design parameters. It is important to
detect, quantify, and, if necessary, correct flaws in the system that may arise
from unforeseen environmental factors, limited historical/background infor-
mation, etc.
   Discussions on many of the monitoring techniques used to validate the
design are presented in Section 5.5.2.  Monitoring individual well perfor-
mance is recommended.  Comparisons with the design operating pressures
and flows will provide the necessary data to balance the system and ensure
optimal distribution of flow across the target remediation area.
   Further, monitoring of groundwater quality (as described in Section 5.5.18)
will allow validation of the design ROI. It is important to ensure that monitor-
ing points are located at discrete vertical intervals (it is recommended that verti-
cal intervals be 0.3 to 0.6 m [1 to 2 ft] in length) and in areas most likely to be
least impacted by the sparging system (e.g., midway between sparge points or
in a lower permeability strata). The groundwater quality data in conjunction
with the SVE discharge data (or vadose zone monitoring data if no SVE system
is used) should be used to evaluate the projected rates of volatilization/biodegra-
dation of the contaminants at the site.

5.5.14  Permitting Requirements
   Permitting requirements vary significantly on a state-by-state basis. Gener-
ally, permits may be required for construction and operation of a remedial sys-
tem, for discharge of any offgases from a remedial system, and where specialty
gas injection (e.g., methane) is proposed. Local authorities should be consulted
for information regarding permitting requirements at a specific site.

5.5.15  Design Checklist
   The following is a list of items that need to be addressed hi the design and
implementation of sparging technology for aquifer remediation at a given site:
        •  develop a geologic cross-section with contaminant distribution
           overlay;
        •  determine radius of influence of sparge well (provide reasoning);
                                 5.163
-------
                                           '„ "!/,
 Groundwater Treatment Systems
         •  determine number of wells required;
                 !  '        'i "    "  '  '», "   i" , ! -!l '•'    :   "       ,   :, '• ,! ',:'!"•
         •  determine injection pressures and flow rates;
         •  determine pulsed or continuous injection (pulse interval?);
         •  define well details (superimposed on geologic cross-section):
                 "'	       '. „',•'• .   .  ....      ,|
              •  screen length and diameter;
                           i r	    M  ;	;,       i
              •  slot size/filterpack; and
              *  screened interval;
         •  design Manifold:
   -   ,; •    • " 	;';•	/ .,   :.	'"i-   • r. •• .     • :n I     "   '•'•  •••    ;     i -  ',:
           ,.;,». type and size;
              •  headloss calculations;
              •  valving and instrumentation;
              •  P&I diagram;
                                             i
         •  develop equipment specifications;
         •  complete required permits; and
         •  develop design validation/system monitoring plan.
                                            1
 5.5.16  Design Implementation
   Implementation considerations for in situ biosparging systems and associ-
 ated components are provided in Sections 5.5.3,5.5.7, 5.5.8, and 5.5.18.

 5.5.17  Start-up Procedures
11 •   , ,  ,.„          •' ,,n  i ,   i,   ,,  "   , "''ill „, ij'., ",„,,!,  . "i 'I ' I 'f|., , i,,1	||| .',,.!,. 'I,, ,  ... , •'   i.i , I „!!"• . ,
   Extensive site definition and baseline measurements should, be made to
 determine the extent and distribution of contaminants. All NAPL should be
 removed prior to the initiation of IAS to avoid spreading of contaminants.
 SVE systems, where required, should be operated first to remove any prod-
 uct and vadose zone contamination. These systems should be interlocked
 with the IAS system.
                                             j
   If any chemical adhesives were used in constructing the system, the
 volatiles from the manifold system should be purged by opening IAS well-
 heads and injecting air into the manifold lines and running the air compres-
 sor for a minimum of 10 minutes and up to 2 hours. A portable hydrocarbon
 "sniffer" should be used to determine when the lines have been purged.  Af-
:"•.    i'".  P     ' '    •    • i '     ,.i.. " : „„.  "MI,  nil!1;, .... ' i ,i ' i :| i1 ....'   ' •' '••  	    i •   I...   H 	'i • H
 ter purging the manifold lines (if appropriate), he following procedures
 should be implemented for each of the IAS well groups:
                                   ! "hi' i1. ,   ,!' '[ ''       '"     '       ' ':,':' ' "
                                             i
                                 5.164
-------
                                                            Chapter 5
        •  Turn on the air source. Adjust the throttling valve from a lower
          pressure to the necessary pressure to attain the design air flow
          rate for the chosen well group.  (Do not exceed the maximum
          recommended air pressure);
        •  Balance the flow to each well since each well may behave
          differently;
        •  Develop a flow versus pressure (F/P) curve for each well. The
          generated F/P curve allows determination of well flow rate based
          upon wellhead pressure measurements. This approach reduces
          the effort required during routine site measurements;
        •  After balancing the wells, verify the; air compressor and manifold
          line pressure and total injection flow rate. Also, determine the
          agreement between total air compressor flow and the cumulative
          flow as measured at each of the wells. Any design deficiencies
          will be apparent at this time;
        •  Sample the S VE system inlet and exhaust streams and analyze
          each over the start-up period;
        •  Check for bubbling in piezometers at the site. If bubbling is
          observed, operators  should install air-tight caps on these wells to
          prevent fugitive VOC emissions;
        •  Record periodic groundwater table measurements to document
          the site-specific impacts of operating the IAS well group on
          groundwater mounding/mixing; and
        •  If any positive subsurface air pressure readings and/or high levels
          of vapor-phase contaminants are measured in vadose zone moni-
          toring points adjacent to buildings or other structures that may
          accumulate potentially hazardous vapors, system operators
          should immediately re-evaluate the operational parameters of the
          sparging system. Discontinue operation of the air sparging sys-
          tem if conditions are deemed unsafe.

5.5.18  Performance Evaluation

5.5.18.1 Operation Practices
   Following system startup and balancing, two simple methods are avail-
able for responding to changing conditions:                        ,
        •  vary air injection rates per well; or
                                5.165
-------
Ground water Treatment Systems
        •  vary pulse frequency and pulsed well groupings.
   As part of the system evaluation, these two optimization methods should
be explored to provide site-specific system enhancements over the first six
months of system operation.  If monitoring of the system suggests that cer-
tain impacted areas of the site are not being effectively remediated, incorpo-
rating additional sparge wells into the system — potentially, the most effec-
tive optimization strategy available once the system is installed — should be
considered.
                                           i  '   "     "         '
5.5.18.2 Operation Monitoring
   Monitoring of various physical parameters associated with the IAS arid
SVE processes is necessary to ensure optimal system performance.  Typi-
cally, IAS and SVE pressures and flow rates for the system are measured and
recorded. Individual wells are also monitored to balance me system.
    : •      "  ,,,'„, ,,, ,", ,"'',''   , ' :"', iif  , i, ,, : "" „, ' 'Hil'i 'i ' n, , "i • ! ,ill,  i,j|i hi!1!!, '' ,    •""	'•'  '   •      ' "I I'll .'Hill" '
Groundwater levels should be monitored to  ensure that the SVE well screens
are not submerged and that the IAS well screens are submerged.  Addition-
ally, groundwater levels can serve as an indicator of possible impacts of I AS
and SVE activities on regional groundwater flows.
   Previously, monitoring of various chemical parameters (e.g., DO levels)
was used to determine ROI in relatively short evaluation periods (i.e., <1
day).  Over longer time periods, monitoring of parameters, such as relative
abundance  of hydrocarbon-utilizing bacteria and measurements of hydrocar-
bon content in water samples is warranted. Typically, BTEX measurements
in groundwater samples are conducted on a quarterly basis. However, trie
minimum monitoring frequency depends upon local regulatory agency re-
quirements.
   Contaminant removal rates can be .monitored by observing the hydrocar-
bon content of the gases captured" by the SVE system. When contaminant
levels have dropped to non-detectable {f^)'m tne vapor phase, groundwater
BTEX levels should then be monitored.  Following several months of IAS,
groundwater BTEX levels will also often show at non-detect levels.  This is
especially likely if measurements are made while an IAS system is still in
operation or immediately after IAS activities have been terminated.  How-
ever, after extended periods, a rebound in BTEX levels may occur. This may
coincide with high groundwater levels in the spring and probably results
from the mobilization of trapped NAPLS that were not within the zone of
influence of the sparging system.
   A review of 21 IAS sites revealed that rebouncl generally occurs within 6
to 12 months and in some cases can take as long as 16 months to develop
                                5.166
-------
                                                          Chapter 5
(Bass and Brown 1995). This study showed that IAS was especially effective at
sites where there was a dissolved phase alone (i.e., an absence of NAPLs).
When petroleum-based LNAPLs were present, such as in source areas, the
highest incidence of rebound was observed. Petroleum- contaminated sites also
exhibited a higher level of rebound than chlorinated solvent sites.
   The efficacy of IAS is highly dependent upon: (1) the extent of contact of
the injected air and the contaminated soil and groundwater; and (2) the mag-
nitude of aquifer mixing. Therefore, rebound was minimized by adequate
treatment times (>10 months for source areas and 4 to 6 months for dis-
solved plumes),  high air flows (>17 m3/hr/well [>10 scfm/well]), close well
spacing (<6 m [<20 feet]), and a high sparge well density covering the entire
area (Bass and Brown 1995).

5.5.18.3 Quality Assurance/Quality Control
   See Section 5.5.13, Design Validation.
5.6  Emerging Technologies — Permeable
Migration Barriers

  Traditional methods of source area remediation are increasingly being
replaced or augmented by emerging and recently-established technologies.
In the last several years, alternatives for addressing downgradient migration
or controlling contaminant plumes have also gained attention. Traditional
methods of preventing migration (i.e., slurry walls and pump-and-treat sys-
tems) have been effective, but are expensive.  Alternative approaches of
plume containment include intrinsic bioremediation (Section 5.3) and migra-
tion barriers. Migration barriers can consist of air sparging wells, wells
containing slow-release oxygen compounds (or other electron acceptors),
and interceptor trenches.

5.6.1  Biological Barriers
  The primary goal of biological migration barriers is to reduce the concentra-
tions of groundwater contaminants at locations downgradient of the barrier.
The barrier is placed to intercept the groundwater contaminant plume as shown
in Figure 5.35. Placing the barrier along transect A-A' increases the potential
for preventing contaminants from passing through the barrier as compared to
placing the barrier further upgradient (e.g., along Transect B-B').

                               5.167      i
-------
Groundwater Treatment Systems
                               ....       ..
                               Figure 5.35
                        Migration Barrier Concept
   In one sense, biological barriers have been used longer than any other
method. As demonstrated several years ago by studies at a wood treatment
facility in Conroe, Texas, naturalbiodegradation can be the controlling factor
with respect to how large a plume grows (Wilson et al. 1985). At many sites,
natural attenuation has controlled or limited migration during the period
prior to discovery of the contamination and during the often lengthy period
of site investigation through remedy implementation.
   Recently, intrinsic bioremediation (as discussed in Section 5.3 of this mono-
graph) has been adopted as the solution for many sites. Several well-docu-
mented studies identifying the factors controlling natural attenuation, pressure
on responsible parties to control costs, and increased awareness of regulatory
agencies concerning costs associated with UST reimbursement programs have
all contributed to the rapid increase in use of this approach.
  :- ' -„  ...'*'.'	        , :;•,.;':,::,";;„ :,. .'!::•:.[:;..;,:••.    „•.  '      ;;:.! ',:i "
   Intrinsic bioremediation will always play an important and appropriate
role in site remediation. However, as with other technologie;s, it will not
always be the most appropriate technology after all factors are considered.
                                 5.168
-------
                                                             Chapter 5
 For instance, at many service stations, the annual costs for monitoring and
 reporting may range from $10,000 to $25,000 per year.  Under the right
 conditions, some active form of remediation may have a lower life-cycle cost
 if active remediation can be completed within several months for less than
 the cost of one or two years of monitoring and reporting (Norris, Dey, and
 Shine 1993).
                            i
   At sites where intrinsic bioremediation cannot control the plume or where
 property boundary, receptor locations, or other considerations  dictate that the
 edge of the plume not be allowed to extend as far as it does under intrinsic
 bioremediation conditions, an engineered system is appropriate.
   Engineered biological migration barriers take many forms.  These include:
        •  air sparging wells;
        •  wells containing slow-release oxygen compounds;  and
        •  interceptor trenches with sparged air or some other method of
           adding an electron acceptor.
   All of these systems create a zone containing elevated levels of one or
 more electron acceptors (usually oxygen) through which the contaminated
 groundwater must flow. In addition, air sparging results in physical removal
 ofVOCs.
   Initially, it was  thought that the point of compliance should be immediately
 downgradient of the barrier. This led to the practice of installing  barriers at
 locations along transect A-A' where the concentration of the biodegradable
 compounds was fairly low. At this point, the electron acceptor flux provided by
 the barrier would be sufficient to reduce the constituents of concern to below
 regulatory levels within a short distance of the barrier. While this approach has
 merit, it does not give full credit to the natural assimilative capacity of aquifer
 systems and tends to be an overly conservative approach.
   If the barrier is located relatively close to the source area (e.g., along transect
B-B'), the contaminant flux may overwhelm the electron acceptor flux from the
barrier, and dissolved biodegradable compounds will emerge from the
downgradient edge of the barrier. However, the mass of biodegradable constitu-
ents will be diminished downgradient of the barrier compared to what it would
have otherwise been. As a result, the demand on the aquifer's natural assimila-
tive capacity will be lessened. This will result in compliance levels being '•
achieved at points closer to the source than would occur without the barrier.
   Placing the barrier further upgradient may actually improve the effi-
ciency of the barrier by introducing the electron acceptor flux where it
has the greatest opportunity to be consumed by beneficial reactions.

                                 5.169
-------
Groundwater Treatment Systems
Further, since the plume may expand in the downgradient direction, a
smaller barrier (shorter trench or fewer wells) might be required. In the
following sections, three types of biological migration barriers are dis-
cussed and corresponding results of laboratory and field tests and mod-
eling are provided where available.
 ' . •          ' ",!'    ,       ."   "! •    :> .    ' l.'l...' I ,..."        '   | ,";,	•

5.6.2 Air Sparging Barriers
   Biosparging is the basis for one approach to plume containment. A treat-
ment zone or curtain is created by introducing air into the solute plume
(Gudemann and Killer 1988). This provides VQC removal by mechanical
stripping and subsequent migration of the vapor phase into the vadose zone,
where attendant biological degradation of the contaminants in the presence
of introduced air and possibly nutrients can take place.  Degradation may
also occur in the sparged groundwater.
   Several configurations of the treatment curtain concept can be con-
structed.  Basically, all of the approaches involve the introduction of a gas,
usually air, into me saturated zone below the deepest level of the plume. The
subsequent upward migration of the air strips the volatiles from solution
converting them to the vapor phase. Once in the vapor phase, the contami-
nants migrate upward to be collected by vapor extraction pipes or wells or,
alternatively, treated in situ by the addition of amendments (primarily nutri-
ents) to enhance their in situ vapor-phase degradation (bioventing). One
fortuitous, but confounding,""factor in the engineering design of such installa-
tions is the potential bioenhancement effects of the air introduction itself.
Biodegradation and stripping will occur simultaneously but at different rates,
depending upon several geophysical and chemical factors of the aquifer
system. This situation represents a basic and fundamental trade-off between
the optimal rate of gas flux required in the saturated zone to reduce the sol-
ute concentrations to the design level and the proper residence time or reten-
tion time required in the vadose zone (and saturated zone) to achieve suffi-
cient in situ biodegradation of the stripped vapors so that sufficiently low
emission levels at the site are achieved.  This is a conflicting trade-off, since
a higher rate of gas flux is normally needed to strip the volatiles, while a
lower flux rate may be required to yield sufficient retention times to achieve
desired metabolic action by the indigenous soil organisms.

5.6.2.1  Background
   In situ stripping by gas (air) injection into the saturated zone, both alone
and combined with soil venting technology, has successfully removed dis-
solved hydrocarbons from groundwater. Applications usually require
                                         	i '    'I
                                 5.170
-------
                                                              Chapter 5
 extensive arrays of sparge and vent wells and secondary treatment of recov-
 ered soil vapors before discharge, as described in detail in Section 5,5 of this
 monograph. These systems are designed primarily for treatment of the
 source area. The use of an air-injection curtain, a continuous linear array of
 injection points transecting a contaminant plume, for interdiction of dis-
 solved hydrocarbon, could reduce construction and operating costs com-
 pared to either more extensive source area treatment or pump-and-treat sys-
 tems.  Focusing on bioremediation rather than physical removal of VOCs
 could further reduce the construction and operating costs.
   To successfully apply the air injection curtain technology for In situ
 plume interception and treatment, several design parameters need to be de-
 termined. They include the air flow rates required to remove the dissolved
 contaminants from the water; the location, number, and depth of the injec-
 tion points; the oxygen loading rates required for biological degradation of
 the contaminants; and the relationship of these parameters to in situ aquifer
 properties including water chemistry, groundwater velocity, and soil charac-
 teristics. Quantifying these items enables the proper design of the air injec-
 tion wells required for the curtain, and optimizes the field-scale application
 of this technology.

 5.6.2.2 Bench-Scale Tests
   In 1991, the U.S. Air Force funded a prototype demonstration project to
 explore the use of treatment curtains or zones to contain groundwater
 plumes. The demonstration project described here included design, con-
 struction, and operation of a prototype aquifer model to determine the effects
 of variations in design criteria on air injection and interdiction curtain perfor-
 mance. A typical sand aquifer with simulated groundwater velocities of 0.3
 and 1 m (1 and 3 ft) per day was simulated in a 1.7 m (5.5 ft) long by 1 m (3
 ft) wide by 1.3 m (4 ft) high Plexiglas® model (Figure 5.36). Several tests
 were performed using the prototype aquifer model.  Bromide tracer and
 toluene volatilization studies were used to hydraulically characterize the
 aquifer physical model. Following these characterization tests, air sparging
 studies were conducted with and without toluene addition and with sodium
 azide, a biological inhibitor.  Biodegradation was clearly present when the
biological inhibitor was not in the prototype aquifer. The concentrations and
distributions of toluene and DO were the main performance indicators used
to evaluate the performance of this bench-scale test unit.
   Approximate steady-state conditions for toluene were usually reached within
3 days under abiotic conditions (Figure 5.37). The baseline conditions were
satisfactorily reached between sparging at each air flow rate event.

                                 5.171
-------
"»' in1•"iiiiiiii!!1'i wiiii f ininnnii! ;<,	i< ini agspi i" j	\'.; i"' 
-------
                                                                            Chapter 5
                                      Figure 5.37
    Toluene Concentration Profile Sparging at an Air-to-Wgter Ratio of 1 1
                                     Presparge Profile
          10,000
        i!  9i00°
        ji  8,000
        g  7,000
        g  6,000
        |  5,000
        g  4,000
        "  3,000
        |  2,000
        [S  1,000
              0




v
Infl
—
uentS^















'




/ Sparge Wells Location
/


	

, . , ,
_Effluent.


0.0    0.5
r  i   i
   1.5
                                                2.5         3.5
                                     Distance Along ModesI (ft); x—*•

                                   After Sparging 1 Day
                                                   I   I   I
                                                   4.5   5.0
                   0.0   0.5
                 1.5          2.5         3.5
                  Distance Along Model (ft); x-

                After Sparging 2 Days
                                                                      4.5   5.0
                              i—i—i  "i  i—i—i—i—i—i—r~i—i—i   r—i   i
                                     Distance Along Model (ft); x-
di Row1
^ Row 2
•i Row3
C3 Row 4
— Approximate SS

Reproduced courtesy of Traverse Group (1993)
                                         5.173
-------
Groundwater Treatment Systems
                                 Figure 5.37 cont.
    Toluene Concentration Profile Sparging at an Air-to-Water Ratio of 11


                                  After Sparging 5 Days
y 6,000
•* 5,000
O 4,000
	 S
| 3,000
6 2>00°
g l.UUU


Fnfl




0

[font
\



0


1. 	 r



).5 1


	 	



.5 2
/ Sparge Wells Location
*

	 	 	

n. ~- f1
1 II 1 1 1 1 1 1 1
5 3.5 4.5 5.0
      -S' ' 6,000

      -.A  5,000

      : j  4,000

           3,000
                                  „ ,;  • ....... ;  ;;„ ;  ;  .... , ; ;;  ![ ;   ;-.
                                   Distance Along Model (ft); x
                                       ' \ '^\~.'  ,  , ':   _' ..... '
                                Postsparge (3 Days) Profile
                             ,  ".  »".! ...... . ../ItJI'1"".,,, I'lll III t '  ':, .  'T  I, I' , '"
                                    Distance Along Model (ft); x


                                Postsparge (8 Days) Profile
      i •          i	•:	
•*-!—T"T—i—i—r^T—i—i—r1-!—i—i—r*i—i—Mr
0.0   0.5         1.5         2.5         3.5
                                                                    4.5   5.0
                                   Distance Along Model (ft); x
c? Row1
^ Row 2
•• Row 3
G3 Row 4  .
— Approximate SS
Repraducad courtesy of Traverse Group (1993)
                                       5.174^      (
-------
                                                               Chapters
         1. A simulated groundwater velocity of 1 m (3 ft) per day, no tolu-
           ene addition, and sparging at various air flow rates to examine
           effects on DO concentrations;
         2. A simulated groundwater velocity of 0.3 m (1 ft) per day, no
           toluene addition, and sparging at various air flow rates to exam-
           ine the effects on DO concentrations;
         3. A simulated groundwater velocity of 1 m (3 ft) per day, toluene
           addition, no inhibition of biological activity, and no sparging to
           establish pre-sodium azide addition baseline conditions;
         4. A simulated groundwater velocity of 1 m (3 ft) per day, toluene
           addition at two concentrations, inhibited biological activity con-
           ditions, and no sparging to establish post-sodium azide addition,
           or presparge, baseline conditions; and
         5. A simulated groundwater velocity of 1 m (3 ft) per day, toluene
           addition, inhibited biological activity conditions, and sparging at
           various  air flow rates to examine effects on hydrocarbon and DO
           concentrations.
   Baseline measurements were obtained by water sampling at approximate
steady-state conditions before, during, and after operating the typical model
at each condition listed above. Influent and effluent aqueous toluene con-
centrations measured under several conditions were used to conduct a mass
balance on toluene.
   Early observations revealed that the toluene disappeared rapidly after injec-
tion into the model aquifer even before sparging, due either to significant bio-
logical activity or volatilization. To examine and isolate the physical effects of
air sparging on toluene removal, the aquifer model was treated with sodium
azide to discourage biological activity. The unsaturated zone was eliminated by
raising the water table to prevent the biodegradation of toluene in the vapor
phase.  The effects of the biological inhibitor on the model toluene concentra-
tions were quantified and recorded for comparison to non-inhibited conditions
where the biodegradation effects were clearly observed.
   Results of Air Sparging Tests. Results of the air sparging studies under
abiotic conditions verified that toluene removal was enhanced to 90% as it
passed the air sparging interdiction curtain at air-to-water ratios of 11 and
22. A minimum air-to-water ratio of approximately 10 was required to effect
the 90% removal of toluene as it moved through the interdiction curtain (see
Figures 5.37 to 5.39). Subsequent increases in the air-to-water ratio did not
yield increased toluene removal.                                   !     ..
                                 5.175
-------
 pv
 M«l
*-xi:
 O
p
vo
ved
uene Re
Fraction o
o
I—
                                                              Figure 5.38
                                    Effect of Air-Water Ratio (Volume/Volume) on Toluene Removal
                                        ir
                                                     10              15

                                                              Ak-to-Water Ratio (A/W)
20
25
30
      • Ms/Steady State M2
      OM3/M2
      At 3 days after sparging initiated


      Reproduced courtesy of Traverse Group (1993)
                                                                                                                                       0
                                                                                                                                       s
ent Systems
-------
                                                                   Figure 5.39
                          Effect of Air-Water Ratio (Volume/Volume) on Toluene Concentrations Along Flow Path
en

•vj
"•J
                    6.000
                    5.000
                    4,000
                     3,000
2,000
                     1,000
                                                                  2         2.5         3

                                                                 Distance Along Model (ft); x—!
      • 2.2 A/W
      • 6.6 A/W
      O 11 A/W
      D 22A/W      -      •       -
     	Approx. SS
     Average of Rows 1 and 3 each distance; after sparging 3 days

     Reproduced courtesy of Traverse Group (1993)
                                                                                                                               O
                                                                                                                               :r
                                                                                                                               Q
                                                                                                                                                    Ol
-------
Groundwater Treatment Systems
   Even though these air sparging studies were conducted under abiotic
conditions, toluene removal was significantly enhanced by relatively low
air-to-water ratios. There appears to be a limiting air-to-water ratio that can
produce a certain level of treatment under abiotic conditions. Removal by
biodegradation must be enhanced by a prescribed level of oxygen dissolved
in the water; the air flow required to deliver the optimum amount of oxygen
could prove to be greater than the minimum air-to-water ratio for enhanced
volatilization alone. However, the oxygen delivered at the minimum
air-to-water ratio to effect enhanced volatilization was theoretically suffi-
cient to degrade the remaining 10% toluene. If this holds true for field-scale
applications, retention times (controlled by sparge well depth and soil char-
acteristics) must be recognized as a critical parameter in specifying the de-
sign air-to-water ratio.
   Downgradient DO concentrations increased with each successive increase
in the air-to-water ratio. However, DO movement through the model ap-
peared retarded under biologically-inhibited conditions, leading to the
premise that "trapped gas pockets," possibly augmented by sodium azide,
could have continually leaked DO into the groundwater.
   Summary of Air Sparging Barrier Results. The feasibility of air injec-
tion curtains as an interdiction method was demonstrated in laboratory-scale
tests. Air sparging was effective as a migration Carrier even under abiotic
(worst-case) conditions. Approximate steady-state conditions with respect to
the toluene profiles were reasonably confirmed, which allowed a relative
mass balance on toluene to be performed.
   The minimum air-to-waterratio required to achieve an enhancement of
toluene loss of 90% through the interdiction well field in the aquifer model
was 10.  Subsequent increases in the air-to-water ratio did not increase tolu-
ene removal. The enhanced volatilization capability of the interdiction field
appeared limited by the depth of the air injection wells in the
laboratory-scale model.
   The air injection DO profiles showed that DO increased downgradient of
the injection wells in sandy soil, with a groundwater velocity of 1 m (3 ft)
per day under abiotic  conditions. Under these conditions, the size and loca-
tion of the biodegradation treatment zone depended on groundwater velocity
and the degree of potential biological activity inside the treatment zone,
which would have depleted the DO under biotic (field) conditions.
   Several criteria were identified for the design of air injection interdiction
curtains.  These include the minimum air-to-water ratio required to ad-
equately remove contaminants, the depth of the air injection points to ensure
                                 5.178
-------
                                                             Chapters


 optimum removal of volatile contaminants, amd the spatial location of the air
 injection points to maximize the biological treatment zone and prevent con-
 taminants from passing through the injection well field.
   The identified operational problems include increased surface monitoring
 as compared to soil venting systems; possible fouling of the injection wells
 (both inorganic and biofouling) that could alffect aquifer characteristics; and
 the occurrence of channeling, that  may be increased at higher air flow rates.

 5.6.2.3 Field-Scale Systems
   In the last few years, numerous air spargkig-based migration barriers have
 been employed. The following examples were implemented by Remediation
 Technologies Inc. (RETEC). One example is a site where the remediation of
 the release of 150,000 L (40,000 gal) of unleaded gasoline to the soil and
 groundwater at an operating marketing terminal is ongoing. The site is situ-
 ated on river alluvium composed of silt, sand, and gravel deposits. The un-
 confmed aquifer that was affected by the release has a water table that fluc-
 tuates between 7 and 10 m (20 and 30 ft) below the ground surface because
 of regional irrigation.
   A previously-installed groundwater containment system was not effective
 at recovering free-phase product. Therefore, soil venting was selected to
 remediate the subsurface contamination in the vadose zone above the water
 table and to remove the immiscible free-phase product on the water table.
 The soil venting system consists of three main components: (1) soil venting
 wells; (2) a vacuum blower and piping; and (3) a thermal oxidizer air treat-
 ment system. Permanent soil gas monitor probes were installed throughout
 the area to monitor organic vapors, vacuum, and oxygen levels.
   A microbial fence in situ bioremediation system was installed to contain
 the dissolved hydrocarbon plume by establishing a zone of enhanced biodeg-
 radation activity at the leading edge of the plume. A total of 14 air injection
 wells were installed to provide the oxygen necessary to  stimulate aerobic
 hydrocarbon biodegradation. An air compressor delivers air to the air injec-
 tion wells at a design flow rate of 5 m3/s (3  cfm) per well.
   Soil venting costs included $140,000 for construction, $15,000 for compli-
 ance and air permits, and $5,000 per quarter for air sampling and reporting.
 Construction and startup of the microbial fence system cost $90,000, which
 included operator training and preparation of a comprehensive O&M manual.
   In the first 19 months of operation, the soil venting system removed the
equivalent of 61,000 L (16,000 gal) of gasoline vapor. Concentrations of
BTEX in the groundwater downgradient of the microbial fence have

                                5.179
-------
Groundwater Treatment Systems
consistently been below detection limits, indicating that the microbial fence
is effectively containing the dissolved hydrocarbon plume. The remediation
strategy has achieved a faster rate of product recovery than could be obtained
through pumping, without excavation or disruption of existing facilities or
operations. The microbial fence system, in conjunction with source area
treatment, has successfully halted plume migration without the need for
groundwater extraction, expensive aboveground treatment, or the need for a
NPDES permit.
   A second example involves the design and construction of an in situ
bioremediation system to eliminate off-site migration of oil at a refinery.
The refinery, along with a neighboring facility, were required by court order
to cease off-site migration of oil within 1 year. Off-site migration included
contaminated groundwatef and oil seeps to a stream. To comply with the
order, the refineries cooperatively installed a product barrier wall and di-
verted the stream to a channel capable of handing a 10-year flood event.
Additional groundwater pumping wells were installed to locally reverse the
groundwater gradient.  A biosparging/soil venting system was designed to
remove hydrocarbon constituents from soils and groundwater and to prevent
the off-site migration of dissolved hydrocarbons along the former stream
channel downgradient of the product barrier wall. Approximately 0.4 km
(0.25 mi) of former stream channel was targeted for remediation and plume
containment.
                                             .I
   The system consists of 78 biosparging wells to inject air and provide
oxygen to the saturated zone. The injection air is produced by a 50-hp com-
pressor at a total rate of 5.8 mVmin (180). The soil venting portion of the
system includes over 310 m (1,000 ft) of horizontal extraction wells and 11
vertical vapor extraction wells and a combined flow of 25.6 mVmin (800
cfm) from a 50-hp positive displacement vacuum pump. The exhaust air is
treated with a catalytic oxidizer capable of achieving 99% removal efficiency
at a lower operating cost than normal thermal treatment.
   Concurrent with the design, a treatability study was conducted to optimize
the biodegradation rates at the site. This study determined that nitrogen avail-
ability would limit the rate of biodegradation under field conditions. Conse-
quently, a system was incorporated to periodically add low concentrations of
nutrients to existing wells within the area of the biosparging system. Data have
shown that the system reduced hydrocarbon constituents to below action levels
and has effectively halted the migration of dissolved contaminants.
    !'•	:	  ; •        	• •>	-• '•' •	 	 I	    •     '  -I	
   At a third site, various alternative technologies were evaluated for ground-
water extraction and treatment to control the migration of dissolved organic
                                 5.180
-------
                                                             Chapters
 constituents at a chemical plant.  Based on the contaminants of concern and
 the geological/hydrogeological conditions at the site, an in situ microbial
 barrier (microbial fence) was selected using biosparging as a potentially
 effective, low-cost approach to plume containment.
   The project was initiated by performing laboratory treatability studies to
 determine the oxygen and nutrient requirements of naturally-occurring mi-
 croorganisms in the saturated zone soils to maximize the biodegradation of
 phenolic compounds in the groundwater. The treatability study demon-
 strated that the constituents of concern could be effectively biodegraded by
 the microorganisms at the site and also demonstrated that the rate of con-
 taminant biodegradation was increased through the addition of supplemental
 inorganic nutrients.
   Based on the results of the treatability study, a detailed work plan for the
 pilot-scale in situ microbial barrier system was prepared.  A detailed engi-
 neering design for the system was completed,, specifying three groundwater
 aeration wells, an array of nested monitoring wells upgradient and
 downgradient of the line of aeration wells, and vapor probes to assess the
 effectiveness of the groundwater aeration system. The system is designed to
 deliver air to each aeration well using an oil-free air compressor at a rate of
 up to 0.1 m3/min (3 cfm) (actual air injection rates are less than 0.03 mVmin
 [1 cfm]). Inorganic nutrients, consisting of a blend of ammonium chloride
 and phosphate salts, are supplied through periodic addition to an upgradient
 injection well. A bromide tracer study was performed to verify the rate of
 groundwater flow through the aeration zone.
  During the first 2 months of system operation, concentrations of VOCs in
 the groundwater within the treatment area were reduced by an average of
 84%, while SVOCs, primarily phenolic compounds, were reduced by an
 average of over 85%.  The pilot study is expected to operate for 12 months.
 If the pilot system is effective at halting the downgradient spread of the dis-
 solved contaminant plume, a full-scale system will be designed for imple-
 mentation along the downgradient property boundary, eliminating the use of
 groundwater recovery for plume containment.  In situ bioremediation tech-
 niques may then be applied to remediate contaminant source areas within the
plant boundaries. The application of this innovative in situ bioremediation
process is expected to save the client several million dollars during the
course of the remediation program as compared to a conventional   ;
pump-and-treat approach.
                                5.181
-------
 	  	j;	    n|	j|M|	I	||	  	|	      , ,•	  ,   ,    ,  	;;	| n||	||	• in|  	
I'll	.'ii	It !,''H'    1	IH!1111    '   'i IHIIl'V "E.T'I -	, , 	 '   "I'll           "•	'   ; ,  "t  !'• .'\:	:•,",	I'll!	'.:[ V?*-, Wi""!", 1'"!.!	 "j11!1!:,,.'	.«	III1:!"1 "'"I!1!""!'
                                  Groundwater Treatment Systems
                                  5.6.3 Oxygen Release Compound Barriers
                                     Passive migration barriers can consist of one or more rows of wells con-
                                  taining an oxygen release compound (ORC) as shown in Figure 5.40. The
                                  wells are placed perpendicular to the direction of groundwater flow. The
                                  release of oxygen into the aquifer creates a zone of increased oxygen
                                  through which the contaminated groundwater must flow.
                                     The distribution of oxygen around the well is controlled by diffusion and
                                  dispersion. To serve as a barrier to cbritaminant migration, the wells must be
                                  closely spaced. Depending upon site conditions, the spacing might range
                                  from 1 to 2 m (3 to 6 ft).
                                     The rate at which oxygen must be supplied depends upon the contaminant
                                  mass throughputwhich'isthe product of the "average contaminant cohcehfria-
                                  tion arid the groundwater flow through the treatment zone. The migration
                                  barrier must release oxygen at a rate ffiat equals or exceeds the demand
                                  based on the reduction in contaminant mass that is required to meet the
                                  clean-up criteria.
                                     The rate of oxygen release depends upon the mass of the ORC and the
                                  oxygen release profile (e.g., the rate of conversion of a given mass of ORC
                                  to a given mass of oxygen).
                                     Various derivatives of hydrogen peroxide have been evaluated for use
                                  as migration barriers or for other biodegradation  processes (including
                                  soil cells) where a continuing source of oxygen is required for aerobic
                                  biodegradation processes. These peroxide "compounds include magne-
                                  sium.peroxide, calcium peroxide, sodium carbonate peroxide, and urea
                                  perbxide. Magnesium peroxide arid calcium peroxide have oxygen re-
                                  lease profiles appropriate for migration barriers.  Sodium carbonate per-
                                  oxide and urea peroxide release oxygen much too rapidly to be of practi-
                                  cal use. As a result, most bioremediation studies to date have been con-
                                  ducted with either magnesium perbxiSe or calcium peroxide.  Nearly all
                                  6f the studies involving migration barriers have been conducted with
                                  magnesium peroxide by or in conjunction with Regenesis™,,
                                     For oxygen release barriers to be effective, the constituents of interest
                                  must be aerobically biodegradable.  Secondly, it is necessary to determine
                                  whether the contaminant mass mroughput can be managed by a
                                  cost-effective amount of ORC. In general, this means intercepting the plume
                                  where its oxygen demand does not exceed 20 mg/L of dissolved hydrocar-
                                  bons. The actual maximum concentration of hydrocarbons that can be bio-
                                  degraded varies depending on grouriSwater flow rate, mass of ORC used,
                                  and oxygen release profile of the specific ORC product.

                                                                   5.182
-------
                                                      Figure 5.40
                                                  ORC Barrier .Concept
en

00
                                                                                                                       O

                                                                                                                       Q


                                                                                                                       f
                                                                                                                       cn
-------
                              Groundwater Treatment Systems
i	at14. i :.i:a „,!,„	,  i	IE., i: 11!.-%,a* ,•, M	JIB;' ^ii
                                 For a specific site, engineering design parameters for these barriers in-
                              clude location of the rows of wells within the .plume; well spacing and
                              screened interval; diameter of wells; and characteristics of the specific ORC
                              product. Location of the barrier relative to the contaminant concentration
                              gradient will determine the oxygen demand it must meet. The spacing,
                              screened interval, and diameter of wells determine the mass of ORC that can
                              be placed in the aquifer. Thus, the rate at which oxygen is  provided can be
                              engineered into the system.
                                 The advantage offered by passive migration barriers is that no mechanical
                              or electrical equipment is required, and no trenches are needed to conduct
                              water or air. As a result, these systems require no maintenance other than
                              periodic replacement of the ORC, which can be incorporated in a concrete
                              matrix contained in sleeves. Thus, one or more times a year, a technician
                              can remove the old sleeves and replace them with new sleeves. As with
                              other systems, periodic monitoring is required.
                                 To date, a number of field trials have been conducted to evaluate and
                              demonstrate the use of magnesium peroxide as a biological  migration bar-
                              rier. These include a controlled release test at the Bordon Landfill in Water-
                              loo, Ontario; gasoline release sites m North Carolina and New Mexico; and a
                              mixed fuels release site in Alaska. Each of these tests demonstrated that
                              qxygen could be provided at anci downgradient of the ORC-containing wells
                              and that monoaromatic hydrocarbons could be degraded downgradient of the
                              source area using oxygen released from these barriers.
                                 The controlled release experiments in Canada involved the release of
                              benzene and toluene through a row of wells located parallel to and
                              upgradient of a row of wells containing magnesium peroxide in a concrete
                              matrix and control wells (Bianch-Mosquera, Allen-King, and Mackay 1994).
                              Test data presented in Tables 5.34 and 5.35 demonstrate that increased DO
                              levels and degradation of toluene occurred downgradient of the
                              ORC-containing wells. The tables show oxygen and benzene concentrations
                              upgradient, at, and downgradient of the oxygen source  and control wells.
                                 The North Carolina study evaluated the performance of a biological mi-
                              gration barrier at an existing spill site (Kao and BoHon1994). A row of
                              wells spaced 1.6m (5 ft) apart intercepted the plume approximately 46 m
                              (150 ft) downgradient of the source area. Increased DO levels and decreased
                              BTEX levels were observed in monitoring wells 3 m (10 ft) downgradient
                              from the ORC-containing wells (Tables 5.36 and 5.37). Less clear results
                              were observed from monitoring wells located further downgradient. It has
                              been speculated that small amounts of free-phase gasoline may have existed
                              near some of the ORC-containing wells. If so, this would have resulted in a

                                                              5.184
-------
                                                                         Chapter 5
                                    Table 5.34
                 Oxygen Release Barrier Controlled Release
                        Test Results— Dissolved Oxygen
                                          Dissolved Oxygen Levels at Day 26
     Distance from Source
                               Control
            Concrete                       ORC/
           (2nd Control)   ORC/Concrete      Pencils
0.6 m"

1.0 mb(ORC Wells)

1.5m0
                                               «X5
<0-5

-------
                                                                                                 	I.	;	
                                   Groundwater Treatment Systems
ill'!;.'
i ;•
                                   larger oxygen requirement than initially anticipated. This test also produced
                                   some evidence of iron precipitation immediately downgradient of the
                                   ORC-containing wells.  Studies of methods for predicting conditions under
                                   which iron precipitation might occur, whether it is a significant potential prob-
                                   lem, and methods for mitigating the impact of iron precipitation are in progress.
                                                                       ' ....... ' ',!! .'I  , ill1' , ' I ,! " , ...... I i • ill1,1

                                                                             ..... il!    '"'
                                            Oxygen Release Barrier at Spill Site — Dissolved Oxygen
                                                                          I'm '" ''!""  IIF i i'li .] '
                                                                          Dissolved Oxygen Levels
                                                              DayO
                                       Day 23
                      Day 38
          Day 115

Background
Upgradient of ORC wells

4.4
0.7

3.4
0.3
f '
3.0
j
0.02

-
0.4
Downgrandicnt of ORC wells
 3m
 8m

 23 m
0.7
0.7

0.7
                                                                           0.0
0.2
1.0

b.b
2.8
1.0

0.3
                                   Adapted from Kao and Borden 1994
                                     Full details of the first oxygen barrier were reported by Marlow et al.
                                  (1995). This migration barrier was installed in Homer, Alaska, after a pilot
                                  study that compared the ORC method with air sparging. The shallow depth
                                  to groundwater, 1.5 m (5 ft), and dissolved iron concentrations in excess of
                                  100 ppm accounted for the limited effectiveness of air sparging at this site.
                                  Also, with air sparging, contaminant levels rebounded during treatment,
                                  indicating that channeling may have occurred and/or diffusive transport from
                                  low-permeability layers (clay lensesj etc.) was rate-limiting.
                                     The barrier was installed on a former utility site at which approximately
                                  2,150 m3 (2,500 yd3) of soil contained as much as 13,130 ppm of diesel- and
                                  gasoline-range organics (DR^/GRO) and: 32.8 ppm"of BTEX. The ground-
                                  water dissolved-phase concentrations were as high as 6.8 ppm for DRO/
                                  GRO and 3.1 ppm for BTEX.  After soil excavation, which removed most,
                                  but not all, of the free product, the barrier was installed.  Normally, all free
                                                                   5.186
-------
                                                            Chapter 5
                              Table 5.37
               Oxygen Release Barrier at Spill Site—BTEX
                                        Total BTEX mg/L

Background
Upgradient
Downgradient
3m
8m
23m
DayO
0
6

6
3
7
Day 23
0
11)

1
4
6
Day 38
0
11

5
6
12
Day 115
-
29

1
• 1
-
Adapted from Kao and Borden 1994
product should be removed; however, in this case, free product could not be
removed from areas under a warehouse. Nevertheless, regulators allowed the
migration barrier to be installed as the next most reasonable action to take in
order to pull the control point back to the source. The barrier consisted of a
series of short treatment barriers installed sequentially downgradient within a
trench running parallel to the direction of groundwater flow.
   Groundwater velocity at the site varied by two orders of magnitude, from
less than a 0.3 m (1 ft) per day to several meters per day, as affected by
freeze-thaw cycles. During the winter, when flow was minimal, excellent con-
trol was achieved, and DRO/GRO and BTEX were significantly attenuated. At
the furthest downgradient point, DRO/GRO was reduced from 7.4 ppm to be-
low detection limits, while BTEX was reduced from 1.36 ppm to 0.017 ppm.
   The barrier was recharged just after the spring thaw. As the water table
rose into treated pockets of sorbed material and as the groundwater velocity
increased dramatically, the barrier once again proved to be an effective con-
trol mechanism.  When measured in the summer, the control point outside
the barrier had a BTEX concentration over 50 times the previous measure-
ment at the time of recharge of the ORC wells. By contrast, groundwater
that passed through the barrier zones showed increases in BTEX concentra-
tions only 9 times and 1.5 times  higher than previous measurements at the
first two sequential points, respectively. At the final downgradient point
below the final barrier series, BTEX levels v/ere actually reduced by 40%.
Based on these results, the site owners chose; to again recharge the barrier.

                                5.187
-------
' fill iflllllllll!!!	il'linilll,, tijiili,;' Illlllll!!1!*', , HilF ' ' lilllllll J"I":" JIPPI	Ill 111 III I III 1111  I III Illlllllllllllllll IIII II III III I III III 11 I 11  	 "' :,i< " "I1"!"'-' II," '"   ' s I" ',,.:; :'(':'''!	"""IP1 III I*,	I Jil "'"' i'"'  !«'• • •'!' ili'iniini,!:1" ill", i»" <,•''.,: Hi1 "V <', fllln i, •:,. 	

,", .... |   . ; . | ..     :                         , ,       ;i .  • •:.   • .      ' .•.  .    :  :;

 ,:'!' ':" ,	':!" | ' '     if '   •      lilllllll   I I          .           '         ''•''",,  U

     i  I1       J1'!i! 'i  ,                   '                .         ' '       '•!,',
                                   Groundwater Treatment Systems
                                      The most extensive field demonstration to date was conducted by GRAM,
                                   Inc. in Belen, New Mexico^  The site involves an abandoned service station
                                   that required remediation due to the presence of gasoline-impacted soil and
                                   groundwater as well as small intermittent pockets of free-phase hydrocar-
                                   bons. The site is a state-lead site, and the study was conducted in close coor-
                                   dination with the New Mexico Environment Department (NMED).
                                      A BTEX plume extended across property boundaries for a distance of
                                   greater than 30 m (100 ft). The soils consisted of a heterogeneous mixture
                                   of clays, silts, and sands. The depth to groundwater ranges from  1.2 to 1.5 m
                                   (4 to 5 ft) below grade.  USTs at the site were removed in 1982 and the site
                                   and surrounding properties were the subject of separate investigations from
                                   1991 to 1994.  These events led to the use of the site for a pilot test followed
                                   by a full-scale oxygen barrier installation.
                                      The pilot study was carried out primarily to determine oxygen
                                   dispersivity from a single point source and to record, with lesser frequency,
                                   the status of dissolyed-phase hydrocarbons.  ORC was installed in a single
                                   15 cm (6 in.) well and monitored downgradient at 26 monitoring points and
                                   several existing monitoring wells. This well is designated ORC-1 in Figure
                                   5.41, and is referenced in the legend, of Figure 5.42.  DO and BTEX levels
                                   were monitored from October through December 1994. The results showed
                                   that oxygen from the ORC was able to disperse readily downgradient and
                                   remediate hydrocarbons. Based on these results, the NMED approved the
                                   installation of a full-scale oxygen barrier.
                                     The full-scale barrier consisted of 20 15-cm (6-in.) ORC wells. The
                                   downgradient sampling array was increased to 54 points as presented in
                                   Figure 5.42.  Vertical distribution of DO and BTEX were monitored with
                                   probes at 1, 3, and 6 m (3,10, and 17 ft) below the water table in the barrier
                                   zone and at various single and multiple depths at the downgradient points. A
                                   total of 342 ORC socks were installed on April 3,1995, and the system has
                                   been monitored extensively since.  Over a thousand oxygen and BTEX read-
                                   ings have been taken. Oxygen was measured using a modified Winkler titra-
                                   tion (Hach test kit) because the NMED considered chemical methods, such
                                   as this, to be superior to oxygen electrodes. Ohmicron BTEX immunoassays
                                   were made, and over 50 of these measurements were supported by conven-
                                   tional laboratory GC analysis to establish needed correlations between the
                                   methods; about a dozen immunoassays can be performed for the cost of one
                                   GC! analysis.'     •  •   •      •	

                                     Oxygen distribution was estimated by contouring the areal and  vertical
                                   distribution of initial  and subsequent oxygen concentrations at  10 sampling
                                   times over a 3-month period. BTEX levels were evaluated using the same

                                                                  5.188
-------
                                                              Chapter 5


methodology. Figures 5.43 and 5.44 are concentration contours generated from
these measurements for DO and BTEX, respectively.  Rapid increases in DO
can be observed following ORC barrier installation. It is clear that an oxygen
barrier forms across the wells and rapidly moves downgradient from them.
                               Figure 5.41
                      New Mexico ORC Barrier Site
                             (not to scale)
                                                MW-10
                                        GRAM-2 •

                                         • GRAM-3
 • Permanent Monitoring Well
 O Temporary Monitoring Well
                                  5.189
-------

I"! llfl
	Ill"
                                    Groundwater Treatment Systems
Sill!!!1: •!„,, •	  ii 	'
'l''i ,"'i'l|lll'!|!|ll rllll I i I  1 '   |i|i ,
i'lif	ir'-uL'.ti;
,it,, 'l|l|i''i, "'I* 1	!i	 "'
iS1"' I,!!:!!,,, SB1' ' , '!'„, !, !!	i1,!:  ' "
II"!! HP '•' ' I  li<  :•' ill  in"1

Eft'}	if'	1*1:	••
                :.1
                         \r. lO'-iJ
ill	J, ] ''" ' J " f ',!;,!	
•.'ills,', ?ijit;*;':.  '"i  is,;i; .•	*
      ,1: " '!!  >:l!l ,.  Illlllllilll:"!1' 'Ifi,
               w
                                                   '••  J|; " \  "   • :';:m Figure 5.42
                                                   Oxygen Source Wells anSTMotifrorihg Points
                                                                         Easting
                                           440   450   460  470   480   490   500   510  520   530   540   550   560
920 -
910 -
900 -
890 -
880 -
1 870 -
860 -
850 -
840 -
830 -
820 -
1 1 1 1 i 1 1 I 1 1 1
N
^ 0 ®
oJ^T
1 . •
' ' ' •I®®0®
• _ • *
* • • . ••*" . •
...
* *

ii
©ORC Source Well
O Pilot Study ORC Source Well
• Existing Upgradient Monitoring Wall
• Monitoring Points
                                      Oxygen levels remained fairly constant for the first 47 days following
                                    ORCI barrier installation while BTEX levels dropped significantiy. At Day
                                    47, a significant rise in BTEX was noted and oxygen began to be consumed
                                    more rapidly. Subsequent to this event, BTEX concentration levels contin-
                                    ued to decline with the corresponding use of DO through Day 9:5.
                                      The presence of the BTEX spike can be understood in relation to the im-
                                    pact of additional work at the site. The jsf^ED had directed that the con-
                                    taminated source area be excavated before the barrier was installed. The
                                                                     5.190
-------
                                                              Chapter 5
                               Figure 5.43
   Dissolved Oxygen Concentrations Before and After Barrier Installation

                              Distance (ft)
           460  470   480   490  500   510  520  530  540   550
        910
                                                            I—1 0
                                  Before

                              Distance (ft)

           460   470   480  490  500   510  521)  530   540   550
        910
     a
10

8

6


4
                                  After
excavated soils were located in the vicinity of well SH-4 (Figure 5.41), about
15 m (50 ft) northwest of the barrier. This physical disturbance, that
reached the surface of the water table, impacted the equilibrium of the soil/
groundwater system and is presumed to have increased the dissolved-phase
BTEX load moving toward the barrier. The riise of BTEX at Day 47 is
                                 5.191
-------
                                     Groundwater Treatment Systems
                                     consistent with the observed groundwater velocity of 0.3 m (1 ft) per day.
                                     Detailed chemical analysis of the specific hydrocarbon distribution in
                                     groundwater was also consistent with the introduction of fresh material.
                                     Pigure 5^44 presents me plan view of me study area with BTEX contours at
                                     the time of ORC installation and at Day 93.
                                                      ,'•'	!'.',,*!'• '!, V",J"*;i!!l":l"':  *'"}•
                                            	•  •••	;-  T":,, .•: , !:•:•	-;?:,/",FigureS.44
                                               BTEX Concentrations Before and After Barrier Installation
ill ill    ' ! <:|i
               Ills:
I- "I	 (
	I
                          ."Hif1:1!!11  i niillii
                                                                     Distance (ft)
                                                 460   470   480  490  500   510  520   5iO   540   550
                                              850
                                                  '»    ***£*  V
                                                  h   ,< yv*^L   s-.<-~Kfy-&r
                                                       *  X*   i   \ ->^%%'
                                                                        Before
                                             910
                                                                     Distance (ft)
                                                 460   470   480  490  500   510  520   530   540   550
                                                                 ,!„.„	I,,,,	I	I...
                                                                            10
                                                                            9
                                                                            8
                                                                            7
                                                                            6
                                                                            5
                                                                            4
                                                                            3
                                                                            2
                                                                            1
                                                                         After
                                                                        5.192
-------
                                                           Chapter 5
   At the end of the 3-month experimental period, the impact of added dis-
solved oxygen on BTEX levels was observed in well SH-6 located 37 m
(120 ft) downgradient of the barrier. BTEX levels decreased from several
hundred ppb to ND.  In essence, natural attenuation had previously resulted
in BTEX levels of less than 1 ppm at SH-6 compared to about 10 ppm in the
vicinity of the barrier.  Following ORC installation, BTEX levels further
decreased to ND despite the impact of the increased dissolved-phase BTEX
generated by source excavation. This particular result serves as a graphic
example of the principle of enhanced intrinsic bioremediation; the presence
of the oxygen barrier served to pull the control point back toward the source
with respect to well SH-6.
   The system continues to operate.  The available oxygen in the ORC filter
socks was about half consumed after 3 months, so replacement charges were
scheduled to be installed after 6 months of operation. Based on other stud-
ies, the life span of the ORC filter socks should increase as the BTEX mass
reaching the barrier decreases.

5.6.4  Interceptor Trenches
   Aeration trenches (curtains) represent another form of barrier for the con-
tainment of plumes (Gudemann and Killer 1988; Wilson et al. 1992). In this
technology, the water entering the trench is sparged with air, stripping VOCs
and providing oxygen for the degradation of biodegradable compounds. In
air stripping, it may be possible to use the overlying vadose zone as a
bioreactor for treatment of the VOC-laden air, or if some of the VOCs are not
easily biodegraded (e.g., chlorinated compounds), the offgas will need to be
captured by an SVE system for treatment.
                                        i
   Three aeration barrier trench configurations are presented in Figure 5.45.
All involve the injection of a gas (usually air) into the saturated zone below
the deepest level of the plume, followed by upward movement of the air,
during which volatile contaminants may be sitripped and oxygen transferred
to the groundwater to  support biodegradation. In addition, nutrients may be
added to the aqueous phase as it moves through the trench.  The extent to
which bioremediation and/or air stripping will take place is controlled by
several factors:
         •  biodegradability of the contaminant(s);
         •  volatility (Henry's constant and vapor pressure) of the
           contaminant(s);
         •  residence time of the water in the trench;
                                 5.193
-------
                                                                                                                                                                                                                                           I!!::

                                                                                          •3*
11   '
             i
. = > . - . - E . _ - . " °
Soil Surface
Vadose
Zone
y
1
2
3
GroundwaJer
Flow
Aquifer
n-2
n-1
n
,
•*-
r> —
v_/^
•< —



	 >• —
V-

t
O< -
*

Figure 5.45
Three Interceptor Trench Configurations
f Soil Surface / / Soil Surface / / ''
	 Soil
Vadose
	 SVEWell Zone
	 Gravel ,
JAX —
Groundwater
Flow


	 Sparging Well —
<-
C
*l
*l
r*i
I
r*
J
V>
~v
J*
< —

f
	 Soil
Vadose
	 Gravel T
-> 	
	 Barrier Aquifer
— »•
Aquitard
^— Sparging WeU
Aquitard 	 '
Standard Flow Counter-Current Flow

C
*l
-*l
r*1

^r^
-M—
-U
i -
o



t-
I

	 SVEWell
i f;
-— — Barrier


	 SpareineWell
Counter-Current Flow
Sunk into Aquitard
Source: Mulch, Norris. and Wilson 1 997. Used with permission. • >
;; ; J
Groundwater Treatment Systerr
55

:
-
-
-
.
-------
                                                            Chapter 5
        •  air flow rate;
        •  system design factors (configuration, bubble size, etc.); and
        •  temperature.
   Both biodegradation and air stripping are enhanced by longer residence
times, higher air flow rates, and higher temperatures within the sparging
trench.  Rapid biodegradation competes with iiir stripping; however, the
contribution of biodegradation is reduced if the constituents are highly vola-
tile.  Maximum biodegradation rates are typically achieved with relatively
modest air flow. Generally, therefore, the fraction of the VOCs that are bio-
degraded can be maximized by operating the system at the minimum air
flow rate at which the level of contaminant removal is satisfactory; much
less air is needed to strip most VOC constituents than to provide oxygen for
biodegradation. Maximizing the fraction of constituents biodegraded may
permit operation with no vapor recovery and treatment, greatly reducing
remediation costs. System configuration has relatively little effect on oxy-
gen transport to the aqueous phase; all three of the aeration barrier trench
configurations shown in Figure 5.45 provide adequate oxygen transport. For
biodegradation  alone, therefore, the extra expense involved in the
counter-current flow configurations (the second and third configurations in
Figure 5.45) would not be justified. However, if it is necessary to air strip
one or more biologically-refractory VOCs, the third configuration should be
implemented as it is the most efficient of the three for air stripping.
   Removal of TCE by the three configurations in Figure 5.45 was modeled
under standard conditions of trench depth, air and water flow rate, and VOC
mass transfer rate coefficients (Mutch, Norris and Wilson, 1997). The first
configuration (cross current) air stripped 88.4% of the TCE in the influent,
the second (cross current/counter current)  removed 96.4%, and the third
(cross  current/counter current with a purely counter-current section at the
bottom) removed 99.97% of the TCE.
   Air bubble size should be small to  increase; air-to-water surface area and
bubble residence time, thereby enhancing mass transport of oxygen and
VOCs at the air-to-water interface. Excessive: air flow rates will result in
large bubbles, undesirable turbulence in the trench, and even reduced hy-
draulic conductivity across the trench. This latter effect may lead to changes
in the plume flow pattern and may actually cause bypassing of the plume
around the ends of the trench.
   The air injection trench must be of sufficient length and depth to intercept
the entire plume with some modest safety margin. Its width must provide
adequate contact time under conditions of maximum  groundwater flow rate;

                                 5.195
-------
         Groundwater Treatment Systems
         the actual value will depend on both biodegradation and air stripping rates.
         If refractory VOCs are present, air must be provided at a rate to guarantee
         adequate stripping; this point can be readily explored by mathematical mod-
         eling of direct field testing. Provisions must also be made for offgas recov-
         ery and treatment if necessary.  If the intent is to minimize air stripping and
         maximize biodegradation, the air flow must be sufficient to provide a sto-
         ichiometric excess of oxygen for the oxidation of the constituents of con-
         cern, as well as for the oxidation of nontarget compounds. However, higher
         air flow rates may cause excessive air stripping of VOCs and the need for
         costly offgas collection and treatment.
 .!.'        : •    	;     ,:,    ,• •'    .  '  >:•;„?;:., • "V	i'l", '  •$ i  '•   '•  "•'	" '	'  " 	" •*•• ft	 "'"^
           Monitoring wells  should be located short distances upgradient and
         downgradient of the  migration barrier. These are used to demonstrate the
         barrier's effectiveness in removing/destroying contaminants  and in providing
         downgradient oxygen concentrations sufficient for degradation of residual
         organics. Residual organics should be degraded by intrinsic bioremediation
         processes in the aquifer between the barrier and the point of compliance.
         Monitoring wells should also be located at either  end of the trench to make
         sure that contaminants do not bypass the barrier.  Finally, monitoring wells
         should be placed farther downgradient from the barrier and somewhat
         upgradient from the point of compliance to ensure that the system is operat-
         ing satisfactorily.

;=  .•  .   5.6,5 Summary    '         	[  '
           Field and pilot tests have demonstrated the potential use of migration
         barriers based on air  sparging and ORCs.  Both methods provide a low-cost
         alternative to pump-and-treat methods and in some cases, intrinsic bioreme-
         diation. Site conditions, contaminant properties and concentrations, and
         regulatory considerations will impact the selection of methods to be imple-
         mented at a particular site.
           Biological migration barriers have a place in the continuum of bioreme-
        diation technologies that may be applied to contaminated sites. To properly
        apply this technology, it is necessary to understand what is required to meet
        site-specific objectives. In some cases, source  area remediation may be
        requireS during which the downgradient regions of the aquifer must be pro-
        tected; migration barriers are one approach to providing this protection.  In
        other cases, active source remediation may not be  required, intrinsic
       1 L , ..'" "i' ,,|,, '"in1, "   r 	 a:1""',,,  !r',, ,:i|	 "" ,„• "'-'I, • , mil" .I'linilih.iiHK'lii'lliiii!'' HWllli'i<',,iiilii!i< If 	y	 ,A  	7,	     		
        remediation may be inadequate, and some level of engineered bioremedia-
        tion may be appropriate to meeUhe overall site remed^
                                                                         • t'JI	,
                                        5.196
-------
                                                            Chapter 5
   One approach to selecting a remedy consisting of one or more bioreme-
diation technologies is to use modeling techniques to evaluate remedies
beginning with the least intrusive/costly and working successively toward
more intrusive/costly solutions until an acceptable remedy is identified. Re-
cently, Dupont et al. (1996) discussed the use of a model developed by
Demenico (1987) that provides a three-dimensional description of contami-
nant transport taking into account the dispersion, sorption, and first-order
decay of the constituents of interest. A methodology applying the Demenico
model at sites by first simulating intrinsic remediation and then conditions
where the original mass loading of contaminants has been reduced by vari-
ous remedial strategies  has been presented by Dupont, Noris, and Gorder
(1996).  Migration barriers, combined with inlrinsic remediation, were con-
sidered the least intrusive/costly technique incorporating proactive remedies
that can be evaluated at contaminated sites using this methodology.
                                5.197
-------
i'i	i1,'.	:• lilijl	llliWr  .
II	.ICII!' '   TIPil. I!  i;




     i, '  •&"'&'(
                                                                                                                                                                                                                    i'i	s	I
                                                                                                                                                                                                                                                                           •''Hi'I *',, "1  liifliiiii!  !
                                                                                                                                                                                                                                                          4  "Hill1            iiil;l'IPIiiii/i  !i   •  "' :  •  Jill,  ;;i'lili
                                                                                                                                                                                                                                                       •I     .(V
                                                                                                                                                                                                                                                              if  ;
                                                                                                                                                                                                                                                      ,M!i	i
                                                                                                                                                                                                                                                                                           •fill	!!E
                                                                                  .»fi I rllli•!;,",:!,".ill-ill	i

                                                                                  Ml I, :l liligil	''II Ih, HHllIlk:! 'A!1!
                                                                                UJ'i I1'1 .

                                                                                 .1': 11'"III II1'
                                                         i   ;<;   :'    •; '"H'ljiflfit	fl    	'i'i   ,Vsi	lij   ;li   '  '   i   ,'  '"  	,''     """-':.'.  <
-------
                                                         Chapter 6
      VAPOR TREATMENT SYSTEMS
6.1  Introduction

  Biofiltration is an ex-situ technology designed to degrade contaminant
vapors that are generated in manufacturing, waste treatment, or site
remediation activities. As such, the technology is not affected greatly by
site, soil, or waste characteristics except as they relate to: (1) the relative
volatility of the contaminant (the contaminant must move into the vapor
phase but not be so volatile that it cannot be sorbed within the biofilter reac-
tor), and (2) the air permeability and water content of the contaminated soil
(low moisture content is necessary to maximize vapor recovery and contami-
nant removal for treatment in the biofilter). The technology is unaffectpd by
conditions adverse to microbiai activity in the site itself (i.e., nutrient limita-
tions and non-ideal soil, water, and pH conditions) because these can be
easily modified in a controlled-reactor environment.  However, the technol-
ogy is affected by waste constituent characteristics that affect the toxicity,
biodegradability, and bioavailability of contaminant vapors. Toxicity and
biodegradability limitations can be controlled in a biofilter reactor through
dilution, addition of carbon sources to  stimulate! co-metabolic degradation,
etc. The biofilter relies on contaminant mobility in the vapor phase so it can
be collected for treatment, making a contaminant's volatility and solubility
(i.e., its Henry's constant, see Appendix A) important in assessing the poten-
tial effectiveness of this method to treat a specific contaminant vapor. Site,
soil, and waste constituent characteristics that are important in the evaluation
and design of the biofilter vapor treatment system technology are summa-
rized in Table 6.1.
                                6.1
-------
                                                                ,!	11,
                        Vapor Treatment Systems
iir;:i!  ,  -'in ... ,	  M	ill'..,!!
                                                                                        t i .1	.I „ | ,,n.
                             •  •  .'-'„"	'•,. • ;'.•;....'•,::: :;,„;	,:;i Table 6.1
                            Impact of Various Site, Soil, and Waste Constituent Characteristics
                                 on Biofilter Vapor Treatment Technology Performance*
	 : 	 ' 	
.'I'1 » ' '!"!i . : .' ''"I1,, 	 ' , '"'.....iiii'l .. .iiiiiiR. I,1 ..iiiiii: sii;"1, . "ii,, j.1!1!"" ',„ t ', . j1" .. „;;
Characteristic
11 • ' • ' „', " ! 	 i ' "1, 	 | 	 ,'i|i'« ',',,, • i,. 1 ,:l > ">, n ,| , , 	
Site Climatic conditions
Groundwater table fluctuations
Surface structures
Layered formation
	 	 	 ' 	 i 	
Product existence/distribution
; ' ii
Soil Fine grained
High-water content
Low-water content
Nutrients
	 ,. 	 ' „" PH ;" ;; 	 ; 	 ;•; _ ;; 	 ;J" "
1 .'i, '.•'.! 	 ' ,. ' " " : „ ' „ ,n " ' , '"..,"!!„ ,- ' ", "•' iiHiiii • "li'ftii;,!"!ii '"" I'll .'•!„ CMj, :!
-------
                                                           Chapter 6


   The work of Michelsen (1995), which appears in a comprehensive text
entitled Handbook of Air Pollution Control Engineering and Technology was
the primary basis for this section.  In addition to the previous monograph in
this series (Ward et al. 1995), other comprehensive references include Inter-
national Process System (IPS) (1990); VDI (1991); Leson and Winer (1991);
Bohn (1992); Michelsen (1992); Michelsen (1993); and Frechen (1993).
   Biological vapor treatment systems or biofilters use microorganisms im-
mobilized on wet organic packing material, through which the air stream is
passed, to remove pollutants from offgas streams.  Biodegradable organic
and inorganic compounds are adsorbed by the packing material and the
moisture coating the packing. Once adsorbed or absorbed, the odoriferous,
volatile organic or inorganic compounds are oxidized by microorganisms to
carbon dioxide, water, and/or inorganic salts.  The process consists of the
following steps:
        •  collection of raw gases;
        •  pretreatment to remove paniculate, adjust temperature, and in-
          crease relative humidity; and
        •  dispersion throughout the biofilter.
   The basic process requires that: (1) the compounds to be treated be
biodegradable under aerobic conditions; (2) the combination of adsorp-
tion and rate of biodegradation be adequate to permit efficient treatment;
(3) the packing material hasadequate adsorption characteristics; (4) there
is adequate oxygen, moisture, and nutrients; and (5) there is adequate
retention time.
   Suitable vapor streams must contain degradable compounds at concentra-
tions that are neither toxic nor too high for the capacity of the biofilter.
Some compounds, such as formaldehyde, are readily biodegradable at low
concentrations but toxic to microorganisms at Mgher concentrations.  Other
problems that may arise at higher concentrations include excessive biomass
accumulation and acidification.

6.2.2  Process Design Principles
   In general, the compounds most suitable for biofiltration are water soluble
and of low molecular weight. Water solubility contributes both to rapid
absorption by the moisture coating on the packing material and  to biodegrad-
ability. Volatile organic compounds (VOCs) with functional groups contain-
ing nitrogen, sulfate, and/or oxygen typically are water soluble and there-
fore, are rapidly degraded. Odorous inorganic compounds such as hydrogen
                                 6.3
-------
ill liillii    !  l)i
I!	1,1 '  , ' !!',',1
;Silill : '-I'1 i	ii
i'iii "it ill)!! i *" il
if ill:']'])' :  , •
I	till	 i   i 	
               I'iiii'ji't "i „
               ii'':
              iilllil,
              WL,"
                        ;>	It  B!!!
                          fir  Htii
                                       Vapor Treatment Systems
 sulfide, ammonia, and sulfur dioxide are also suitable (Eitner and Gethke
 1987; Don and Feenstra 1984; ftopkop arid Bonn 1985; and Ziminski and
"Ferrara" 1993)._ '"'   " ' '   " '	""' \	!' '"	"=	'"	'"""'"''"	'"" ''"!'"' "":'	"   ^'":'"  ^'	''''' "'  '"""
   Michelsen has developed the following general conclusions based on
 pilot-scale studies regarding the capability of biofiltration systems to treat
 various classes of VOCs (Yavorsky 1993):
        •  Aliphatic compounds with less than seven carbons, such as etha-
           npl, acetone, isopropyl alcohol, methyl formate, and methyl ethyl
           ketone (MEK), are very degradable! These compounds can be
           treated with high efficiency at concentrations of greater than
           1500 mg/m3! Control efficiency' of greater than 95% is possible
           with less than 1 minute retention time.
        •  Aliphatic compounds with more than six carbon atoms or aro-
           matic compounds, such as toluene, xylene, phthalates, and mix-
           tures of aromatic paint solvents are moderately degradable.
           These compounds are degraded at a slower rate man the VOCs
           listedabove/ A confM
           concentrations of 500 mg/m3 and moderate retention times.  Uti-
           lizing biofiltration to treat higher inlet concentrations or to
           achieve higher destruction efficiency is possible, however, a
           longer retention time is required.
        •  Some compounds, such as benzene, degrade poorly, and treat-
           ment by biofiltration is normally limited to an inlet concentration
           of 20 mg/m3. In some "cases",' wKere'there' is a mixture of organic
           compounds, biodegradation can be improved by co-metabolism.
        •  Chlorinated hydrocarbon compounds with one or two chlorine
           atoms, such as methyl chloride and 1,2-dichloromethane, can be
           treated by biofiltration. Inlet concentration is normally limited to
           a maximum of 20 mg/m3 and me'Mter material must cbntaiE a
           buffering agent to neutralize the hydrochloric acid formed by the
           biodegradation process.
   Table 6.2 summarizes the relative applicability of biofiltration technology
 -I i i ;i;	•	 - ••,'.,•.:.	; i. i -. i	',	ii n' JL *, ;;.""> i	: •  : ;	^' ":|;	;	»	.;;	-f fci ••. v.	•«>	wd..,	•,	*,.	,.	,,	i.-	«.  •	 •	;-,. = •
 to classes of compounds and industrial sources.
 ' ,|. ; IjilliilLli"11 j.' , i '.••T'lrtWlt!  I", fi'!11, )*Wi	i! "'""'i" 't'''Jf.(4i|l*lih)lll1'ilHr'l	i'liy IK I/I	I"!	ll!'!ijli*i"'">ill"l	niWHtWi J'il'. ""i ii'Mi'	KU. JiCflr", ill';-. I
   In addition to considering the biodegradability of individual compounds,
  "l|!J!lll,, , ,:;'!' < l!*i '" i ,1 .iilji,1" Jl, I1 JUnill'iii "I" ill I 111""	,J,i 'i11'1',, i,!,|ij," &,«, <	i,	,y ,;	u	ii 	ii,,,Zin	< i,	 ,1	1	lliilTrir	 ,	<	/	.< ,, HI, in, n „« if,,»	 * 1,	< <« i,
 it is necessary to evaluate the impact of the specific mixture of compounds
 present in the offgas stream!  The "complexity 'of'"this""tas"S:is iricfease3 for tfie
 treatment of sources, such as soil vapor extraction systems, because the con-
 centrations and relative composition of the various constituents changes over
                                                                         6.4
-------



Table 6.2

Proven Applications of Biofiltration
Industry
Used oil
Aroma extraction
Beer yeast drying
Fat processing
Gelatin production
Foundries
O
Oi Coffee roasting
Cocoa roasting
SftWflpft trftaftnftnt fmnnirinfll^
Sewage treatment (industrial)
Composting
Plastics processing
Adhesives
Polyester
Tobacco processing
Tank farms
Rendering

Odor
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X


Aliphatic Aromatic
Hydro- Hydro-
carbons carbons
X X




X X

V Y
X X
X
X X
X

X
X X

Organic Compounds
Containing
Oxygen Sulfur
X

X
X X
X X
X
X X
X X
V V
X X
X
X X
X
X
X
X X

i Inorganics

Halo- Aromatic
Nitrogen genated H2S NH3 Oils
X ? X
X
X X
X XX

X
X. X
X X
V Y Y V
X X X X X
X X X X
X
X
X
X
X X X X

Reprinted with permission from Handbook of Air Pollution Control Engineering and Technology, R.F. Michelsen, "Biofiltration," 1 995. Copyright CRC Press, Boca Raton, Florida.














O
Q
¥
O
-------
Vapor Treatment Systems
time, especially during the first several weeks of operation. For some mix-
tures; the consumption of readily-degradable compounds may increase the
biomass, improving treatment of marginally-degradable compounds. In
other cases, co-metabolism may aid the degradation of compounds that
might otherwise be poorly treated.
 |.i" ,,r'lf|r '" "	' '   ,, ':, I'"' i: , "!;"••  . Jri""!.!!,!'!	IN,," , „     I      11  I    III   *      IIII  "   ' »!', ' ', " '"', '. !j Vol ' '•
   For continuous influent flow velocities with relatively consistent compo-
sitions, microorganisms can acclimate to the feed and achieve relatively high
efficiency.  Where me flow is'Inte'frnittenf or variable in either rate or compo-
sition, the system."may never fully acclimate and might not achieve high
efficiency.  However, adequate treatment might still be achievable through
process design modifications to accommodate the specific gas stream. Ex-
amples of such system modifications include providing longer residence
time or providing intermittent makeup feed to the bioreactor.
   The design of biofiltration systems must consider that the process is a
combination  of absorption and biodegradation. As shown in Figure 6.1, a
biofilm surrounds the filter particles The biofilm must be moist so that'me
contaminant can be absorbed into the aqueous film surrounding the filter
particle as shown in Figure 6.'2. Once  in the aqueous phase, the microorgan-
isms can use  the contaminant as a carbon and/or energy source. As the con-
taminant is degraded, more molecules  can be absorbed.  Byproducts of me-
tabolism, such as carbon dioxide, increase in concentration in the biofilm
and then diffuse into the gas phase which escapes the biofilter.
                               Figure 6.1
   Schematic Representation of the Biofilm Surrounding the Filter Particle
Rerinted with permission from Handbook'of Alr'PoHuiion Control Engineering and Technology, R.F. Michelsen, "BibfitoratiorV
1995. Copyright CRC Press, Boca Raton, Florida.
                                  6.6
-------
                                                                Chapter 6
                                Figure 6.2
Biophysical Model for the Biolayer Cg is the Concentration in the Gas Phase
The two concentration profiles shown in the biolayer (C,) refer to (1) Elimination Reaction Rate-Limited and
(2) Diffusion-Limited
Rerinted with permission from Handbook of Air Pollution Control Engineering and Technology, R.F. Michelsen, "Biofiltration,"
1995. Copyright CRC Press, Boca Raton, Florida.
   The gas to be treated must reside within the biofilter for sufficient time to
achieve a high sorption efficiency. The system: must also be designed to
promote biodegradation, which requires moisture and temperature condi-
tions that are not optimal for absorption.  Thus, designs must account for this
contradiction by providing more retention time than is typically required for
absorption alone.  As a result, biofilters are relatively large and are best
suited for relatively low concentrations of VOCs. Residence time require-
ments are typically 15 to 60 seconds, but are frequently longer. The mini-
mum size of the biofilter system (volume of filter material) is governed by
the residence time required. This volume is equal to the volumetric flow rate
times the residence time.
   The required residence time depends upon the water solubility and biode-
gradability of the compounds being treated, their concentrations in the gas
stream, the filter material, the microbial community, and the required re-
moval efficiency.  The characteristics of the flow stream are generally deter-
mined by the application, although soil vapor extraction and bioventing sys-
tems can be operated in a fashion that provides more favorable offgas
streams for treatment.
                                    6.7
-------
flltt'.f	h	'i ,?	ps«l	;i'li;
j	.•i.vt!	^towfcp1;	ts I
                                      Vapor Treatment Systems
                          ,,'lilil  t;, (II •• ' ',   . ,  •  ; .  J   		   	  •.  ',. :',,',	 ,,T. •

                                         For given gas stream characteristics, biofilter design is first based on the
                                      filter material and me source of the microbial community. The material must
                                      provide a favorable environment for metabolic activity and must have ad-
                                      equate sorptive capacity. Materials that have been used include soils and
                                      compost produced.from 'leaves''bark, wood chips, activated sludge, paper,
                                      etc. Other filter materials used include peat, heather, and inert materials.
                                      Wood chips, plastic spheres, and ceramics, have been added to the primary
                                      material to provide greater structural integrity to the biofilter bed.  Some
                                      systems have incorporated activated carbon! For treatment of compounds
                                      with acidic degradation."products"(e7g.,sulfuric acid from treatment of H2S or
                                      SO2), lime, limestone, or dolomite can bei "added to the filter matrix.
                                         Necessary properties of the filter material include:
                                              •  high surface area for microbial attachment and transfer of the
                                                 gas-phase constituents to the aqueous phase;
                                              •  good water retention capacity;
                                              •  good water drainage;
                                              •  low-rate microorganism attrition;
                                                   1 ,,'•', J: • '!• 'i,,,,:, •,  i , '' , jir ,,, ,i  '",•:' I!!1 '  ii"'1™1 ".:•:  "'• i1! I]	[>'fli'l'" '"! M1, '! 	'':,,,• '• , ,1:',' ii"1' , i:      ' , ''I,, ',', 1'!"„» ..." ! !iS".'j,',
                                              •  a source of inorganic nutrients;
                                              •  a permeable structure that provides a low pressure drop during
                                                 filtration; and
                                        1 •..  " i ,  ,  . 0:  '.:•"•,.'	  '"'.',•' '.•" -i ', 'i"1"":"1 '""'  ',Vil; .  '"'"Mini"1  • ''lit''!:.:';,,!' l| :'k»i3:. I "•i.'rf "iii'1 j.1 '  MI: •'»', 'wVt.iSV'JBiv"'1'
                                              •  minimal compatibility to avoid cracks and minimize pressure
                                                 drops.   '    '   '	
                                         To achieve a low pressure differential across the filter material, the height
                                      of the filter bed is typically limited to ifp'to"'O' in (3 to 5ft) with amaxi:
                                      mum height of 2 m (6 ft) for materials 'm"aV'exni£it a low'pressure' drop! A
                                      minimal pressure drop across the filter is necessary to achieve uniform gas
                                      distribution; this requires a minimum height of about 0.5  m (1.5 ft). Where
                                      space is limited, separate biofilter segments can be stacked to reduce the
                                      footprint of the total system.
                                         Microorganisms suitable for biofiltration are typically present in soils and
                                      compost and are typically capable of degrading the classes of compounds
                                      shown in Table 6.2. After the gas stream has been introduced into the biofilier,
                                         I |II  |l I   I         I    |                                   , „ 'ii'||,,'!l,i'il,i"|i '' Jill UNI "ill	'! ".Mlllllllhllilt	1	 1,11 I
                                      the microorganisms will acclimate to the specific substrates and their relative
                                      concentrations.  If the gas stream remains constant in flow and composition, the
                                      microorganisms that can most efficiently use the substrates  as a carbon and
                                      energy source will predominate.  Where acclimation is not effective, it may be
                                      beneficial to add an exogenous source of microorganisms.
                                                                         6.8
-------
                                                            Chapter 6
   It is not possible to determine the optimum residence time, and thus
biofilter size, based solely on the composition and volumetric flow rate of
the gas stream.  For gas streams containing a single contaminant or similar
easily-degradable and readily-sorbed compounds, it is possible to determine
an appropriate residence time based on experience with similar gas streams.
For complex mixtures, the only practical approach to designing treatment
systems is to conduct pilot studies to evaluate: acclimation times as well as
residence times. Where variability in composition or volumetric flow rates
is expected, it may be advisable to incorporate activated carbon in the filter
to prevent immediate breakthrough  in response to increased flow rates or
concentrations.
   System designs include pretreatment for particulate removal and gas pre-
conditioning, primarily for moisture and temperature adjustment, as dis-
cussed in Section 6.2.5. The air distribution network, filter material, and
offgas discharge must also be included.
   The gas distribution system's primary function is to evenly distribute the
influent gas into the biofilter at rates appropriate for the loading of the filter
medium. In upflow systems, additional functions that may be performed by
the gas distribution system include drainage collection, leachate contain-
ment, and structural support. For soil beds, the gas distribution system con-
sists of a ductwork header that feeds waste gas through a horizontal network
of perforated pipes usually located 0.75 to 1 in (2 to 3 ft) below ground level.
Other systems have been constructed from slotted or vented plates consisting
of interlocking concrete or plastic blocks, concrete slabs, or metal grating.
The materials of construction must be compatible with the characteristics
(corrosiveness)  of the waste gas and byproducts.
   The gas distribution system must discharge gas  at relatively equal
flow rates across the entire system. Some systems include unaerated
spaces along the outer edges of the filter to prevent short circuiting of
gases along the side of the filter. The pressure drop across the network
must be minimized by using adequate diameter piping and by limiting
the length of perforated pipe sections.
   Pilot studies  should mimic the planned  final design as closely as practical
(Dragan 1993).  Because the total height of biofilters  typically ranges  from 1
to 2 m (3 to 6 ft), pilot systems do not have to be very large to adequately
mimic full-scale systems. However, the diameter should be large enough to
simulate heat loss/gain within the reactor.  In essence, the pilot system
should be a complete, small-scale system with inlet gas temperature and
humidity control. In addition, the closer the composition of the feed is to the
gas stream to be treated by the full-scale system, the better the design  data

                                 6.9
-------
                                     Vapor Treatment Systems
                           will be. Where theigas stream already exists, a side stream should be used
                           for the; pilot test. It is also important to use the same filter bed material and
                           source of microorganisms that will be used in the full-scale system for maxi-
                           mum representativeness of pilot study results.
                              The pilot study data should be used to develop full-scale system design
                           details after conducting an economic evaluation to ensure that a biofilter is
                           the most cost-effective gas treatment technology available. Provided a
                           biofilter remains the preferred apprbachT the degradation data should be fit to
                           a kinetic model that the design provides adequate capacity to accommodate
                           variations in feed gas composition, temperato                        The
                           design should incorporate volumetric flow rate," gas composition, degrada-
                           tion rates of all compounds present in the gas stream, variability of flow
                           rates and composition,  required destruction efficiency, biofilter capacity
                           based on the field test, and available space for the bioreactor.
     ill'  ,    • '  ",,'!lill 	 iiiilli! ,"', ,  " -:"'. I"11! ,'	i- ii ,   i I     I              I  II       I  II I    I      I    •      I,	!!> i •'
                           6.2.3  Process-Flow Diagram
                              	 ,                    n i    n n nn  mi    i   i i | n             .' ,'ii,   i  • "il jiir i
                              A general flow diagram for biofilter treatment systems is presented in
                           Figure 6.3. Specific systems vary based on the filter-bed design which may
                           include moist packing material in constructed cells, subsurface soils, or a
                           bed through which a continuous flow of water percolates. Typically, multi-
                           bed systems are used.
                              In"additionito the systems outlined in Section 6.2.2, biological vapor treat-
                           ment can be implemented in me form of:(1) fieldingfilters and (2) below
                           ground, native-soii-based filter networks!  Both are discussed below.

                           6.2.4  Process Modifications
                              Biotrickling filters contain conventional scrubber packing material instead
                           of compost, soil, peat, or wood chips as described in  Section 6.2.2 and oper-
                           ate with liquid flow over the packing to facilitate mass transfer (Yang anS
                           Alibeckoff 1995). Air flow can be in the same direction as water flow; how-
                           ever, performance is usually improved  with countercunrent flov^
                              Preconditioned contaminated air is introduced at the bottom of thepadc-
                           ing material and flows upward through the packing. Water inoculated with
                           microorganisms and amended with nutrients is introduced through sprayers
                           located at the top of the packing and percolates counter-current to the air
L     •!',     •   '       I  ' 1 ", .   •" '"'!!: A iiP'I'Jiiii!11""!' '"• !'„' ''ill l! ''ill!'! 'ir'nl'1'' ,,i 'iiiiil1*,;. V i,n"" I , ••' '"K I" ••	 "I	g..:.. •: i " ', ...fi'i'iu:! •. 1,,:1,ii,:i|iiil!»i,1 	II tin1", „' II frl't!':,',,	,	p. ,„  iTum'.1* • .1  , « 	 ^«	••	i]«r	,n ."•
                           flow. After a biofilm is established on the packing material, further inocula-
                           tion is no longer required.  Drainage water is recirculated, reducing the re-
                           quirement for freshi water to replace evaporation losses.  Drainage water
                           recirculation also returns microorganisms to the top of the packing.
     f|i ,,'!', Hi i '!!                   .  1 "  |	Illil"   i      : '	 - nil1   |	 i,,* „' I i!',,,	..-.I  n >,„	|ji	i	"|| i .ii'njMi,	i? i» i , ,'H,I ,„!!„. .1   	n

                                                            6.10
w f!i	i; ,i! ...i,,;,,"
  in n
-------
                                                             Chapter 6
                               Figure 6.3
                General Flow Diagram for Biofilter Systems
  Contaminated Air
    Potable Water
                     Nutrient      pH Adjust
   The counter-current flow provides efficient transfer of compounds from
the gas phase to the aqueous phase and permits easy addition of nutrients
and control of pH. Control of pH is particularly important when the gas to
be treated consists of compounds containing chlorine, sulfur, or nitrogen that
bio-oxidize to yield hydrochloric, sulfuric, or nitric acid, respectively. A
potential advantage of biotrickling filters is mat they generally have a
smaller footprint than conventional biofilters.
   Trickling biofilters have been used to treat the classes of compounds
listed in Table 6.3. Recently, Envirogen introduced a modification of this
system that is being developed to treat chlorinated solvents.  The primary
new feature is the introduction of strains of bacteria that can be "turned on"
by toluene to degrade chlorinated ethenes such as trichloroethylene.
   Another type of biofilter was developed by S.C. Johnson, Inc. for use at a
manufacturing facility in Racine, Wisconsin (Kampbell et al. 1987). Vapors
containing aromatic hydrocarbons from an indoor processing area are fed to
a network of perforated pipes located several feet below the ground surface.
The hydrocarbon-laden vapors pass through native soils which have been
fertilized using  standard landscaping fertilizer mixes.  Other than
                                 6.11
-------
,,!,!i' I ill. I 1, ' ' •: ! 1 '8J"1'
illlir .....
 I'll''!'?*''!" ii1!'!'! <    11  II!	'
 III TEW;,,,!„,,   ; P.; Ill '\;,[f  ,  Jill,inn
            "I  K,;    '.,.."! " "Mir 	'Jli
                                     Vapor Treatment Systems
monitoring, me major operating activity is somewhat more frequent mowing
of the grass located above the filter.
                                                       	        6,3
                                                    Relative Bjodegradation Rate of Individual
                                                       and Classes of Volatile Compounds

Inorganic
Rapid Rapid
Hydrogen Sulfide Alcohols
Ammonia Methanol
Sulfur Dioxide Butanol
Aldehydes
, Formaldehyde
Acetaldehyde
Amines
	 •>• : " .".'•• Organic Acids
Butric Acid

•• , .;";• 	 •: ., ., ', 	 	 ,•,.;;•;.•. ;;
Organic
Moderate Slow
Esters Alip
Ethylacetate Hyd
i! ii
Ketonos Mi
Acetone Pe


Very Slow
hatic Many Halogenated
ocarbons Hydrocarbons
ithane 1,1,1,
Trichloroethane
itane Polyaromatic
Phenols Cyclohexane Hydrocarbons
Aromaticii
Benzene
.11 ;i " 	 lull1 j, , ,,111!..;.; .|,; ;;; ,, ,{ , ,<
"|l" Styrene 	 '
Mercaptans
M; 	 ""i ' ViL: i, ii .f,,,,, 	 M |,
Methyl
Mercaptan
"., 	 1!, .. i,,,,',K,i,.'l!i ' 	 ill!";'.!,,1 ,."'("::'<
,/;•' :;-;,; .; 	 „ ;• 	 ;;, •• • <^ 	
i: 	 | , „ 	 	 M|||| 	

,! .i1"1! """"i; 'ii"j ..v,!*,,,"1!1"1!11 '.• Kill!' , \ i 'I .',:• i| ,,ii»r ' .i!'1"1"
                                     6.2.5  Pretreatment Processes
                                         	      	   •     	  ':"' ,	 i                ,
                                       Pretreatment of the raw gas is essential and fairly straightforward. Pre-
                                     treatment commonly includes particulate removai, temperature adjustment,
                                     humidification, and, if necessary, toxin removal.
                                     • •"•"'	.-	::	;	:•	v	 :.  .. ;	|	:	>	'	•	'!.-	".
                                       Particulates such as dust, oils, or other aerosols can add to biofilter bio-
                                     degradation requirements as well as obstruct or cfog pore spaces of the filter
                                     media or the air distribution network. Filter systems such as fabric filters,
                                     yenturi scrubbers? or electrostatic precipitators, alone or in combination, can
                                     be used to remove particulate matter before the gas is passed through the
                                     filter media. Filter systems are described in detail in Chapters 4 to 8 of a
                                     recent book by Mycock, McKenna, and Theodore (1995).
                                                                      ,/'i'iiii'	, ii!1;	ii,,111'1,;!	' ii': i ,ff",n,ii
                   Iliiffi SI .JfWiiklJll

                                                                                                      >	I.	
-------
                                                            Chapter 6
   The temperature of the influent gas largely controls the temperature of the
 biofilter. The optimum temperature of the biofilter is a trade-off between
 maximizing biodegradation rates and reducing contaminant water solubility.
 Typically, biofilters use mesophilic bacteria whose optimum temperature is
 approximately 35°C (95°F). At lower temperatures, the rate of biodegrada-
 tion decreases by about a factor of one half for each 10°C (SOT). Above
 40°C (104°F), mesophilic bacteria are inhibited and eventually die. Fungi
 tolerate a somewhat broader temperature range than do mesophilic bacteria.
 Bacteria found in cold climates can achieve reasonable biodegradation rates
 at low temperatures, while thermophilic bacteria are active at temperatures
 of 40 to 65°C (104 to 149°F).
   Treatment of relatively water-soluble compounds such as oxygen, nitro-
 gen, and sulfur-containing compounds is rapid because the compounds are
 efficiently absorbed and rapidly biodegrade. Temperature control is more
 important for alkanes, alkenes, and aromatic hydrocarbons. Also, relatively
 low water solubilities limit their rate of transfer into the aqueous phase mak-
 ing their biodegradation rates somewhat slower than those of more polar
 compounds.
   Depending upon the microorganisms used in the filter and the tempera-
 ture of the raw gas, either heating or cooling may be necessary. Cooling
 increases the relative humidity of the gas stream passing through thfc filter
 which may lead to condensation ahead of the biofilter and create the poten-
 tial for drying of the filter material. If the filter becomes too dry, transfer of
 compounds to the aqueous phase is reduced,, the filter material can shrink,
 and cracks and channels can form. Furthermore, filter dryness is undesirable
 as the microorganisms require moisture at levels of 50% to 75% of field
 capacity for survival and efficient metabolism.
   Humidification of the gas stream is generally achieved by use of a water
 spray humidifier which can also serve to remove particulate matter and ad-
just temperature (Allen and Yang  1992). Moisture and nutrients can also be
 provided by spraying water on the surface oif the filter material. As degrada-
 tion proceeds, exothermic reactions increase temperatures across the bed and
 drying of the filter material. This moisture can be replaced through the use
 of automatic sprinkler systems. However, use of this method  alone can
 cause localized drying and thus cracking and channeling. For gas streams
 that contain compounds at levels that are toxic to the microorganisms either
 pretreatment or dilution may be required, at least until the microbial commu-
 nity can acclimate or until the composition of the influent changes.
                                 6.13
-------
   nil 1  inn 11 ill n i    i iiiiiiiinii ill inn ill ill inn ill  inn ill
             i iiiinii ill inn ill i ill inn ill  n n ill  i    n                    i              i  iiiiiii i  i  i  i i  i in i i inn i  11      i  n in n I linn
                                    Vapor Treatment Systems
                                    6.2.6 Pbsttreafmehf Processes
                                                   	;	:  	I	; 	 | •••
                                      Posttreatment processes address water drainage from the biofilter bed,
                                    disposal of the spent filter material, and in a few cases, treatment of residual
                                    vbCs'not treated in the bidfilter	 -"•	
                                      While most systems recycle water that drains from the filter bed to the
                                    humidifier, periodic discharge of the leachate is necessary to remove
                                    nonbiodegradable contaminants. The discharge can be treated by activated
                                    carbon, chemical treatment, etc.
                                      The filter material wili degrade over time and typically will need to be
                                    replaced after a maximum of 5 years of operation (Lesoh and winer 1991).
                                    Normally, systems are constructed of several cells so that individual cells can
                                    •^;s^^||^^'^^^^1^j"n^j1g^1pf lime, 6f addition or replacement of
                                    the filter material. If the filter material does not contain hazardous or listed
                                    compounds, it can be used beneficially. If hazardous compounds are
                                    present, the filter material must be properly disposed.
                                    , i 	   	  /    .    h  ,,„..., in  "  	  	#	„  	 i 	  ir   •	      '   i ,Hi	'MI
                                    6.2.7 Process Instrumentation and Control

                                      Process instrumentation includes the following:
                                            • monitoring equipment for gas velocity on the inlet and effluent
                                              sides of the biofilter;
                                                              .,  '  '  • "  f     "  . I!  '                    '
                                            • temperature and.'humiditysensors and controllers in thepretreat-
                                              ment subsystem;
                                            • pressure gauges on the influent and effluent side of each cell;
                                            • thermocouples within each cell; and
                                                     i||, ,; 1'.  	|i";,|i< i",,i ''I'LI'ii'l! .'i".,,1'1' IK.IAI "ILI','1 n flill,1 III',!'! , i1 "PUPPl'l'.'''!!.!^!.'!!!!! IJJli!	J'P fl!". l.i,,!"!'.'' I ,,'ifl' if I, l|i: il.'ii.l.'K ' ' :' tf 11 J, f f '11 I i "'I1'1 ,,!: "'' f  '" "LI	,',!!» h, ill .'Sill!1 "|. • Tllll 11
                                            • meters for total VOC, oxygen, and carbon dioxide concentrations
                                              in the effluent and VOC concentrations in the influent.
	'	 "   i r  .                                          	.',".'	'	••  '•*  '	'•	 	•	i.	•	.Hi!'	"-.I!	I!	p.	1«	 I" Tni*	.i-r	i,,i.	 n, ••
!<•!;:;. Hi;:
lull'1! #!l|ll|r»
                                       Continuously-recording monitoring equipment is preferred. Alarm sys-
                                    tems are advisable to notify operators if the temperature, humidity, or offgas
                                    quality exceed prescribed limits.
                                    6.2.8  Process and Instrumentation Diagrams
 1	ii!   •lir';'"':  *'" '!l   '::' r11*  ":,:^  i"'':,,v:	   Process and mstramentation o^
/Is I!!!1"'!',, i !'  " ' I , "•"
                                    bed,pfetreatment system,''electricai systemTam!	controls, and shouldillu's- .
                                    trate the following:
                                            • dimensions of filter including height, length, and width;
                                                                                 1
                                                                     6.14
-------
                                                            Chapter 6
        •  electrical system and controls highlighting flow and pressure  *
           measurement upstream and downstream of the filter; temperature
           within the filter; and temperature, flow, pH, and humidity in the
           pretreatment system; and
        •  monitoring system highlighting contaminant measurement up-
           stream and downstream of the filter and in the filter drain water,
           and bioproeess monitoring (oxygen and carbon dioxide) in the
           filter effluent stream.
   Details should include the location and type of pressure gauges, flow
meters, humidity sensors, thermocouples, valves, sampling ports, meters,
and controller sensors.

6.2.9  Sample Calculations
   The capability and efficiency of an operating biofilter in eliminating pol-
lutants in a waste gas is a result of both physical and biological factors. The
physical factors include mass transfer processes, flow behavior of the vapor
phase, and residence time. The microbiological phenomena involve the rate
of pollutant elimination by microbial oxidation.
   Because interaction between the physical  and microbiological phenomena
is complex, simplifying assumptions are necessary to model the system.
Published research describes a theoretical model for the degradation of or-
ganic compounds in a biofilter (Ottengraf and Van der Oever 1983; Ottengraf
1986; Dragt & Ottengraf 1987).  The biophysical model of the phases in a
biofilter are schematically represented in Figure 6.2. As the gas flows
through the biofilter, soluble compounds partition from the gas to the
biofilm. The mass balance for a compound in the liquid biolayer is de-
scribed by the Equation 6.1:

                D(d2C,/dx2)-R = 0                              (6.1)

where:
        D   =   the diffusion coefficient (L2T-J);
        C,   =   concentration in the liquid phase (ML'3);
         x   =   distance (through biolayer) (L); and
         R   =   the substrate utilization rate (M^T1) (biodegradation
                reaction rate).
   Ottengraf (1986) described the reaction rate due to microbial degradation
using the Monod kinetic model, which is widely used for biological pro-
cesses. Ottengraf's model assumes:
                                 6.15
-------
                                        Vapor Treatment Systems
                                                                                                              ,:!,;	il»	 i! i I'HIL 'in	

                                                 •  biodegradation occurs in the liquid phase of the biofilm and is the
                                                    only substrate elimination reaction;
                                                 •  the biofilm thickness is small compared to Ihe diameter of the
                                                    coated filter particle  (This assumption" treats the biofilm geom-
                                                    etry as flat).;
                                                 •  the flow of gas through the packed filter bed is plug flow;
                                                 >  the concentrations of each component in the gas and liquid
                                                    phases are described by Henry's law and are at equilibrium at the
                                                    phase boundary; and
                                        '  i ,„ ' ','!'!". "" iii,! '  	, i!" "  .i'i I1 „ I'11 fil,1'! .'  ', ,!•',:'• '. iii ' 'il'iW! ''il'i'1 iiT'i'i, ill1 I'ji,"!	!lil||llill!lS1;1., / •] ,ir '•,;' 1 '!!• .i J i' i  „ '.',.i'i'"i!  'i/'Kiiii''"""'i iii'Vl1" , "'ii	Ii:1,1"1 lilHI I* j .liilW; •... ivi*
                                                 •  the modeled compound is the; only rate-limited substrate (no
                                                    interaction between compounds in a gas mixture).
                                           •'The reaction rate, (R), is described by the Monod (or Michaelis-Menteri)
                                        substrate utilization relationship:
                                       .                               .    ......       ....... ..............  .......... .....                 (6.2)
                                        ; ' '  ' : , »i ;„„ ",'n  ..... ,'" ' , • 'ini,i:,i! ..... • ,,,!'i!, ,:'  '„ :'/!,; „• ,•,'„ "i '» • :''';„ 111! 'J 'ilii'i'1 ,,,„! .J'l'l'w ,''•"! 'iijjjiijii aJrnhitiv, '" ; ,i;'"' i  , ']hilj'' „ • i ;,,  I              i'i I I  II I  I II
                                        where:
                                                 K    =    the Monod (Michaelis-Menten) constant (MI^3);
                                                       =    tne maximum substrate utilization rate (ML-3Tl).
                                                     " I* ' i ..... lijf »,„•>," WMIi, ..... J ..... WH ...... M»''i ..... !H^^^^^^^^^^^^     ..... .KBiTJfBKIilVf'^1''^ "•'*": ..... -»'•'. ....... ..... '•• .......... •*' .............. %
                                           R   is a function of the concentrations of active microorganisms in the
                                             max                                                         •
                                                  _
                                        biomass layer:
                                        where:      _      [[[ ' 't ................................................. [
                                                  X   =   the cell concentratipn of the active micrporganisms
                                                       '                                 ' ...... [[[ .......... '
                                                 |j,ra   =   the maximum grov/th rate (T1); and
                                                 Y:   =   the cell yield coefficient, (Mcells/Msubstrate).
                                           Ottengraf and Van der Oever (1983) conducted intensive laboratory and
                                        pilot-scale experiments with several common VOCs to determine the overall
                                        kinetics of the biofiltration process. Two kinetic cases were distinguished in
                                       .1	!,.	-'"'I	i" liiiiiii'ij "ji	iivu'.***,'	-t	,,	, i	irii;,	j,. ii,	•H-™	j	-	t	  ~i	JBJ	I	
                                                    active at high pollutant concentrations. In this situation, C: is
                                                    much greater than K^.  Since Km is insignificant, the reaction
                                                    rate, (M5, Is maximized arid there is a linear relationship
    I'i ........ ;i;; ..... :'•"'"& ..... <
          "
   i III	llljiiii i:|.i II il" in    '. " Bill	' '.	  ;,   :	dm, " I'1"1!, i,,;,
    I!1!"	ill!;!	[ "Mis11	:,  5	-I!!' -.(i..1}  •  .  " I'lf! ''FKu  ' ,-, J	!,.	-''"'I	I!" BiiiiM'1 "Ji	HVtj'.iO't ,«!•'' !"•"!	t,i", .' T i i'-JClL".".!)'*" l4IHw£mi	IIII.T"11. Ju1'".: 'IH'V I'-ul'l	il.ii.'afli	,'. l"Jlt	'"" '  '•!	i'S",i	, l.iAlWW	IH
                                   .
i|i> R ; . ....... •.  i iji"!!,  • ' i   ...... i ,(. •  : ..... -   , : » ! :  :i!: j ,f , "' ..... ,i, ..... •
  I iiinilii'	I, r  i| in ill   ,„,  'i!ii!!"",,| * ,	If .'"  ,,, llliillL inj \\v	IBi
-------
                                                             Chapter 6


           (zero-order) between the reduction in pollutant concentration and
           bed height; and
        •  Diffusion Rate Limited (First-Order Kinetics):  at lower pollutant
           concentrations, the biofilm is not saturated or fully active. In this
           situation, Km is greater than Cr  The conversion rate is controlled
           by the rate of diffusion in the biofilm.  The rate of reduction in
           pollutant concentration decreases with height in the filter bed.
   Ottengraf and Dever's experiments with various VOCs (including toluene,
ethyl acetate, butyl acetate, and butanol) determined that the concentration
where the biofiltration kinetics shifts from zero- to first-order is specific to
an individual compound and varies widely among chemicals.
   Ottengraf's kinetic model can provide a reliable basis for sizing a biofilter
in the case of a single-component offgas. However, this model has limited
utility for offgases with a mixture of pollutants. Complications due to inter-
actions among compounds can have either a. positive or negative impact on
the biodegradation rate. These interactions Include co-metabolism, which
can increase the degradation rate of recalcitrant compounds (Kampbell et
al. 1987); cross-inhibition, which can diminish degradation rates; and verti-
cal stratification, where the most readily degradable compounds are metabo-
lized at the inlet portion of the biofilter while less degradable compounds
pass through to be metabolized at upper levels of the bed. As mentioned in
Section 6.2.2, pilot-scale testing is generally required to correctly size a
biofilter treating a multicomponent waste gais stream.

6.2.10 Safety Requirements
   Since biofilters are used to treat relatively low concentrations of volatiles,
explosion hazards are minimal.  However, it is still advisable in any setting
to incorporate only intrinsically safe blowers and other equipment into
full-scale system design. This is particularly true as interest grows in devel-
oping biofilters to treat low flow, high concentration gas streams. If the
volatile compounds in the influent gas  are toxic, it is important to: (1) pro-
vide detectors at locations where leaks might occur; (2) locate the intake side
of blowers as close to the filter as practical so that transfer lines operate
under a partial vacuum; and (3) to schedule regular inspections of equipment
and transfer lines.

6.2.11  Specification Development
  System designs must specify the filter material, microbe source, pretreat-
ment system, gas flow network, and residence time. The gas volumetric

                                 6.17
-------
ill 111 I
III III II
PI1 Al .11 I.   I1
                                  Vppor Treatment Systems
 flow rate and composition and the required destruction efficiency are the
 major inputs to system specifications. Filter materials, microBe sources,
 pretreatment systems, and gas flow networks are discussed in earlier sec-
 tions; this section focuses on residence time.
    The residence time is dependent upon biodegradation capacitjr, which is
 determined by kinetic modeling arid/or pilot testing. Pilot testing, which is
 almost always required.! determines degradation fates expressed as grams of
 compound degraded per volume (cubic yards or cubic meters) of filter mate-
 rial per hour. Typical rates are between 10 and 100 g/m3-hr (0.02 and 0.2 lb/
 yd3-hr) for the most common air pollutants —  approximately five times the
 rates commonly achieved in bioventing systems.
    Residence times can be calculated from the biodegradation rate.  From
 this value, the size of the biofilter can be calculated; using an appropriate
 height arid the composition of the influent gas. The resultant area is depen-
 dent upon the filter material. The pressure drop varies with the filter mate-
 rial, anH'tnus,' different heights can be achieved with various filter packing.
 Soil beds are limited to heights of approximately 0.6 m (2 ft), while beds
 made from higher permeabili ty materials, such as compost, wood chips, or
 bark chips can be 1 to 2 m (3 to 6 ft) high depending upon the gas flow rate.
      ^ _   ....................  .      ................. . ....... '_,_, ...... *; ........... ' ........... "'V:; ...... | ....... ; ....... ,,, .......... |      .      .'•
..... ' • ": The" size of a biofilter can be expressed ..... as"'me' surface loading ratio, in
 which the influent gas volumetric flow rate (mrVhr or cfm) is divided by the
 biofilter surface area (m2 or ft2). Surface area loads typically range from
 about 30 to 300 m3/hr/m2 ( 1 .6 to 1 6 cfro/ft2) depending on the characteristics
 of the influent gas and removal efficiency required. Maximum achievable
 surface loading rates can be as ranges to 500 m3/hr/m2 (27 cfm/ft2) with opti-
 mized, low-pressure drop material. Modest fluctuations in volumetric flow
 rates or gas composition can be tolerated because of the sorptive capacity of
 the filter material.
    Residence times can be as short as 15 seconds for low concentrations of
 easily sorbed and readily biodegradable compounds, such as those usually
 found fin offgases from food processing, and as te
 cohceiitrations of less soluble an2 less readily ^egraSable compounds.
                                    I   I i     I |
 6.2.12  Cost Data
    Costs vary widely among systems based on  design,  contaminant treatabil-
 ity, and gas concentration. Costs are frequently estimated based on capital
 arid operation and maintenance costs per volume of gas to be treated. As a
 result, costs are approximately direc&y proportional to the required system
 retention time. Michelsen (1995) surveyed published data on biofilter costs
              IPI1 <|i     i   i il ........ Ill             i         i            I    In I
                                                                   6.18
-------
                                                                 Chapter 6
and arrived at a range of capital costs of $3 to $30 per mVsec ($5 to $50 per
cfm). The low range is for low concentrations of readily-degradable com-
pounds that are easily sorbed. Typical annual operating costs are on the
order or $4 per mVsec ($7 per cfm).
   Table 6.4 provides a summary of typical costs for aromatic hydrocarbon-
laden air based on several years of experience of one of the few U.S.-based
firms that specializes in biofilters. The total cost of $20,400 includes routine
laboratory tests, design, construction, and one year of operation.  Permitting
costs are not included and vary widely depending upon local and state regu-
latory requirements.  These costs also do not include preparation of the treat-
ment area prior to construction of the biofilter and are based on the assump-
tion that electrical service is available.
                                 Table 6.4
                    Typical Soil Biofilter Process Costs0
Cost Element
ENGINEERING
Prepare Work Plan
Laboratory Tests
Design
Subtotal, Engineering
TREATMENT COSTS
Materials
Construction
O&Mb
Analytical
Management
Subtotal, Treatment Costs
GRAND TOTAL
Cost($)

2,000
1,000
5,000
8,000

170m3@$20/m3 3,400
5,000
2,000
1,000
1,000
12,400
$20,400
Treatment of 250 m3 /hr (140 cfm) of air containing between 0 and 2,000 ppm (v/v) aromatic hydrocarbons with a
removal efficiency of 90%. The biofilter contains approximately 170 m3 (210 yd3) of soil.
bO & M includes electrical power and 30 minutes of maintenance p»r week by the owners' employees.

Courtesy of Bohn, Biofilter Corp.
                                    6.19
-------
JIB iilllH     , linn'1 r nip i i,j,r;|"i ' i, • '"•',  l "   "i11'     ' • '  '",*' h' i '' • •  ,'V !«• '  ''"', ',,'": ' 'H?, !" '•':  ' '/!' "''S;1' ^ .''••« • '! ll«!,1 '!i,,;i''"',!'' "'''i"." '• ' ', '•• 'ifrlii'iij. i'1"1;   ''"W "" V71*!'1'1!1!*''1!!!' "'jiVi!'11' I'.'.iS' „l ''"vi?1'!


                 Vapor Treatment Systems
                  „,,	 ....  .  ;	,.,.;,, r  ,	,.  . ,                    j                     —-

                 6.2.13  Design Validation
  • - ••  "";  .•••:   •   ••  •"  '"  ":  ••  '    ' :  - •"	':  :	';:;':	 •  	  "   •• '-^'	: ;-. • ; • | ••":;••; ;;;';	':	' •  •	•••	-;•••• 1	~- '''z:
                    Pilot test results and experience with similar systems, if available, should be
                 reviewed and compared to the anticipated flow rates, retention times, influent
                 flow velocities, and influent gas composition to ensure that the biofilter has
                 sufficient volume to torattte
                    All control an4 monitoring  equipment, piping, blowers, humidifier, heat-
                 ing coils, etc., should be reviewed to corifmn that they have been sized cor-
                 rectly and arei craistractedl""of materials" "compatible" with the influent raw gas
                 and expected effluent products.

                 6.2.14  Permitting Requirements
      11     i           MI  liii1":;!1;	;	rMlL'iJ. ii	i.'ii::"!li;,	l;:  !*iiii •. '•;;'	  +'-Si:i; i iiiiiiiiiJiiJ ,'-'i	Ht''it4:kiMV' "' - '.liMiiii	,:i:l||!>  :':"u$ '	'viw-piliiiS1 < Ji.it''!
                    As with any gas treatment system, air emissions are subject to air permit-
                 ting requirements according to state and, possibly, local regulations. Unless
                 federally-enforced regulations  apply, individual state regulations will dictate
                 permitting requirements.  Offgas treatment requirements for biofilters are no
                 different man for any other air emission  control system. Typically, the com-
                 pliance process includes obtaining permits to conduct a pilot study, construct
                 the system, and operate the system.  The discharge can be regulated based
                 on:  (1) percent reduction in emissions,  (2) specific concentrations in the
                 discharge or at the property line in which case the height of the discharge
                 point and the distance to me property line may have to be used in the calcu-
                 lation of allowable discharge concentrations, or (3) a specific rate (e.g.,
                 pounds per day) for each compound^  If the discharge exceeds the discharge
                 limit for short periods of time,  the regulating agency must be informed, but
                 generally, operation can continue.
    I   III II I I II I           I    I  ,!,',!"  , '...i'.!"1!1 'i1',»', "',•    .   '! !• "I!1" i1 il< ',"., •!,, '"!'!! ,i	l!,,il!i ii, •:' •!' „ I,1:*,!!1	,': , , "I M  „• Jlhl '!•  .,, "i, ,,il ' "il1 i, ' ,ill!i««i||''i.||

                 6.2.15 Design Checklist
                    Checklists will vary somewhat with design, but must include those items
                 listed in Table 6.5.

 	•;;;;     6.2.16 implementation	'
                    The contaminated gas collection system, if not already existing, must be
                 constructed, and piping must be installed to transfer the gas to thepretreat-
                 ment system. The pretreatment system should be located close to the filter
                 bed inlet to minimize heat and  moisture loss due to condensation.  The filter
                 bed may require a concrete pad for support. Pad construction must allow  for
                 adequate sampling access and ease of service to re-fluff or replace the filter
   .',  ":,:,: .:":  ' /'"material,	        	
                                                                      6.20
                                     	i	
                                                                                                       	l
-------
                                                                            Chapter 6
                                         Table  6.5
                   Biological Vapor Treatment Design Checklist
Feed (Influent Gas) Characterization
Contaminant identification
Range of concentrations
Variability in composition
Temperature range/patterns
Humidity rang;e/pattems
Site Features
Operation schedules/variability of influent source
Location of source
Location of treatment system
Path from source to treatment system
Normal "planl:" operations/activities
Pilot Test Results
Flow rates/bed size
Vapor composition
Degradation nates
Percent contaminant reduction
Humidity, pH, temperature ranges
Design
Retention time
Filter bed size
Blower(s) size
Piping heating, if needed
Pretreatment, filter, humidify/dry heat/cool
Filter bed moisture, pH (buffer)
Nutrient source/levels
Health and Safety
HASP complete/kept on site
Groundwater protection
Electrical codes
Emergency shutdown procedures
Emission excursion alarm
Miscellaneous
Permits
O&M manual
Monitoring plan/schedule
                                        6.21
-------
                                    Vapor Treatment Systems
fl II	,  :  I'., I,
             A monitoring and control panel should be centrally located.  Instrumenta-
          tion should include components to measure: (1) temperature and humidity
          of influent gas before and after pretreatment and biofilter effluent; (2) pres-
          su*e"|rqp through'the" filter;	(3^	gas'flow	tnrougffeac'fi	"^» (4)"'temperature	
          within each cell; and (5) influent and effluent composition including oxygen,
          carbon dioxide,  and VOC concentrations.

          6.2.17 Start-Up Procedures
in	i   i         ii .n  .    _ .     j     in  . i  mi  	ii    J  i L       -I-    j/j1	''^ukit,  \.
             Acclimation of the microorganisms and establishment of conditions that
          can be maintained.over long periods is important for both offgas and recov-
          ered water treatment because treatment must occur in a relatively short pe-
          • i,  n!f  "fl " • "" "•	 ,„" ''Mi1'., ' "''...iiiif1 'ihi;;' ' ii;,"'.ill1 Hill!!. ,",'!i,'I , '''.iJlil'I'lllWI'''''!..!!!!'!!!'.:'''''!'!!!^!!!!..!'!!''!''!!'!'''!!!!1!	IIHII, i« ;" "111 i 1||P,, in!	! l|',niiii	• 11,. •.,„': i ,|v n: -i	• „ •:    :,,!,„• :i"'n	T,II. ;|,	 4,, •••
          ripd compared to in situ systems. In almost all cases, destruction efficiencies
          will be lower immediately after startup than following a period of acclima-
          tion.  Microbial populations will adapt to the composition and concentrations
          of biodegradable materials in the feed gas. Acclimation times can vary from
          a few days to a few months (Dragan 1993).
            Microbial populations will adapt most rapidly to readily biodegradable
          compounds. JE*rbvi3ec![ the filter material contains a source of an active indig-
          enous culture, acciiniation should occur within a few days, and seeding with
          exogenous cultures will not be necessary.  For other moderately-to-easily
          degraded compounds, acclimation may require 10 to 15 days, depending
          upon the source and condition of the indigenous culture (Ottengraf 1986).
            Where the compounds in the gas feed are less easily biodegraded or the
        • I", ,., I"'. :i'l,i"5"!ilii ,,li!"'!IT i'T'll'' J(il „ » v jiii J1!1"1 J"'j'JIf:1'1! I1 "i:.,,|il, i""'"1'!,: ,,il'i	LIII,IH|||'||	 H	Iliiliiililil'ni/ir'Ji'il.'Tiliiiif'lP'l	HlllllUi. Jl.iHIilli .,!•"!•. ,'"' '+• Jti •»' II1": 'I	MI!."' ''n'l'1' *  . "nuil'li'1	'.HI " •" Mill',	I	P'UIIMiP11 '	II	inUHi	
          filter material lacks the appropriate cultures, seed from cultures obtained
         n '!' ' c ii'1', 5','i" S::;1 .si* Jit 111,'"1 j1""1 ', \ • \, ',/.' "||il|i"lii i" 'iii'Li '."iTifiiH,,,:"; fWf a, jioi!iiHi!!!iiiiii!:'i'jn;iw:i:!ii!™^^^^    	.iisrivmiii.,1 fKffn \\\ \\vr ii.. iiiiir'ikiiv,, "liiFiHiiiN".,, :.'• »• •IPO® . ,'i,	•" IPW,n i • »::."«
          from wastewater treatment sludge, soils that have been impacted, for an ex-
          tended time by the compounds to be tfeatgd, or a commercial source will
          have to be blended into the filter material. Documentation of the ability of
          tfiese cultures to degrade the compounds in the gas feed is necessary, and
          such cultures should be tested in the pilot phase  Seeding of the filter mate-
       I   '     	 '' •!•"   "J " 	I	I'll'1	 	 I"	 "" •  	I1	 JT    IT       ^     O         	
          rial can also be used to shorten the acclimation time.
            During the start-up phase, it is essential to provide a backup treatment
          system for the biofilter to ensure that discharge limits are  not exceeded.
          Also during this period, the temperature and moisture content of the air en-
          tering and discharging from the reactor must be carefully monitored, and
          appropriate changes should be made to the operating conditions as tempera-
          ture and moisture balance can be expected to change throughout the acclima-
                                       11   '               ii                     i  i. .
          tion period.
sill;,!, iii, if
31 Mi1.   ; i!  i '•     ,:! .11  ,' 	 •     II
'nil ilil1'1!1 I I n. : I, ;    '   ,   .   ,     I
               1       •
                                                                      6.22
-------
                                                            Chapter 6
6.2.18 Performance Evaluation
   Operating and monitoring procedures are relatively minimal compared to
many other treatment technologies; however, they are critical to maintaining
performance. These procedures relate to maintaining optimum conditions
for the sorption of the volatile compounds and for microbial activity.

6.2.18.1  Operation Practices
   Moisture must be maintained in the proper range (Eitner 1990-1991). If
the moisture level is too low, microbial activity will decrease.  However,
excessive moisture content will hinder air flow, increasing the pressure drop
across at least some portions of the filter and forming pockets which the
volatile compounds cannot reach and where anaerobic conditions may de-
velop. Typically, moisture should be maintained between 30% and 60 % by
weight (50% to 75% field capacity).  Optimum conditions may vary for
individual filters and should be determined as part of the pilot study or dur-
ing operation refinement.
   Moisture can be adjusted by either increasing or decreasing the amount of
water introduced during humidification of the feed gas.  If moisture levels
cannot be maintained by humidification of the inlet gas, the filter will have
to be irrigated.
   As with all biodegradation processes, nutrients must be sufficient to sus-
tain the formation of new cell material; typically, a C:N:P ratio of 350:10:1
is adequate. A portion of the nutrients required may be present in the filter
material. Nutrients can be added as necessary during preparation of the filter
bed, intermittently during operations through irrigation, or during periodic
servicing of the filter bed.
   Control of filter bed pH is important to  maintain microbial activity. Deg-
radation of hydrocarbons generates carbon dioxide, which may lower pH.
Degradation of chlorine-, sulfur-, and nitrogen-containing compounds gener-
ates strong acids which require greater buffering capacity. In some cases, it
may be necessary to add lime or limestone during periodic  servicing of the
filter bed to  maintain desirable pH levels in the biofilter.
   Unique to biofilters is the need to maintain gas flow resistance in the
proper range. Conditions that impede flow or create channeling will reduce
treatment efficiency.  Use of the filter material by the bacteria as either a
food source or nutrient source will result in decomposition of the filter mate-
rial.  Fines formed by this process will fill the void spaces between larger
particles, resulting in blockage of the gas flow in some regions and channel-
ing in others. If this occurs, treatment of volatile compounds will be greatly

                                 6.23
-------
                                    Vgpor Treatment Systems
                         MI,: •	sail
reduced in both regions.  Increased gas flow caused by channeling can be
detected by a reduction in pressure drop across the system.
   Periodic servicing is required to re-fluff the filter material, or after several
yearsTthe filter material may need to be replaced. Systems should be de-
signed to include several treatment cells so that the system can continue to
operate while one cell is being serviced. During servicing, nutrients and/or
buffering agents can be added if needed.
     :	." -,.'	'',   •',	!	"i"  ';".'.. ',.'i,°r,'".":	•	™	         i
   Cessation of operation for more than a few days may result in detrimental
changes to the microbial populations and may require another period of
acclimation before optimum performance is achieved (Eitner 1990-1991).
At a minimum, a low flow of moist air should be maintained through the
filter to prevent anaerobic conditions. SucE a flow can be accomplishes fey
mtermittent operation of the inlet fan to provide fresh air. To supply carbon
for maintenance oYtneHb'rilter p'bpulaUonC oni'e'possible apjproacli"lisl to intro-
     •livr, ('"'I!;,;1!,! ,  	L'M „ : :•: .Ollll!	! IMHI','!1'MH IIIUIK i RJIII	A	S	, ,1	M	•	  	:i,	jr	,	,:,	 .		i, tin ,	r IIMM.;,	,	,	,
duce a source of gas containing a low level of the same volatile compounds
found in the contaminated gas source during intermittent operations.

                                    6.2.18.2 Operation Monitoring
                                        ;. . ;.;• "  ........ ;,  ;;; ............ ;; ......... , •;;;.; ....... ;;; ..... :;:;,; ..................................... : ........................  ! ....... : ..... '" ....... ',:: ...... :::"..;. ...... , ....................... r::rn
                                      Operation monitoring is performed to ensure that permit requirements are
                                    being" met and to 'idfentify'cnanges'in ^ei^mance lliarmay warn of a de-
                                    crease in treatment efficiency. This is accomplished by sampling and ana-
                                    lyzing influent and effluent gases on a routine basis and periodically sam-
                                    pling the liquid effluent from individual cells. Additionally, oxygen, carbon
                                    dioxide and volatile byproduct concentrations, moisture levels, and tempera-
  111    II        111           III  I III       I   I       ,'• 1,1 "1 ..... 'I'll ..... IVIMr, . J.f [[[ ............................ :. ..... , ...... | ....................................  | .................... .". ...... .................... K, ...... ,
                                    ture of the influent and effluent gases  should be determined. VOCs, oxygen,
                                    and carbon  dioxide (and other volatile byproducts) should be plotted and
............... '' ..........            ...........    trends evaluated over time.
                                                                                  I        '                i
                                      Periodically, the filter material should be sampled and analyzed for mois-
                                    ture content, pH, nutrient availability,  microbial populations (plate counts),
                                    total organic carbon content, and filter particle density/porosity. These data
                                    should also  be plotted land [trends compared ito trends in performance to
                                    evaluate long-term changes in bTbfilteFniedia. Md'ttielmpact on
                                    contaminant removal efficiency.
                                                                         t ..... P'l;,,*!!!
                                   6.2. 1 8.3  Quality Assurance/Quality Control
                                      Quality assurance and quality control (QA/QC) practices include those
                                   common to other remediation technologies, including the use of blanks,
                                              I1"'' 41' ill11	Ini 'I 1%'' III nl < 'I	Ill llh"l' II"" hi ,'•. ' II' "HII'l. ill>' .HI.1 'illKiillllllll	HI' ml	i;• I. ilplllllliU1	'	p	* . n	 ,	,i»,	'.Ulpi'.'' . »f	p	p ,HI|IH	»	' n, ,	 	,p	'' nr 11	|||i,i	|	pip	|	pin
                                   blind samples, and spiked samples for monitoring of influent and effluent
                                   air, and review of data for consistency to identify potential labeling
i in i n    nil i      i           i   n i                  • 'i •• , • '"id : ..... - 1,1;,', ...... "i, ••». i-'tiii «),,j ..... [if'iitKMi1'?! l;»:'l;ii:;l ...... mrlititw.i.1 ..... •|t';s .'tifcTis
-------
                                                             Chapter 6
problems, etc. Field monitoring equipment should be routinely tested, cali-
brated, and serviced. In general,  quality control practices resemble those
appropriate for bioventing as discussed in Section 4,2.18.3.
                                  6.25
-------
-------
                                                        Chapter 7
      INTEGRATED  TECHNOLOGIES
7. J  Introduction

  On many sites, both soil and groundwater have been contaminated by past
chemical releases, so an integration of soil and groundwater treatment meth-
ods is often required for effective site remediation. Proper integration of
treatment systems will reduce overall treatment costs and required treatment
times. For example, there is little value using an in situ groundwater treat-
ment system when the soils overlying the aquifer are contaminated with
concentrated, residual-phase contamination which continues to leach soluble
contaminants into the underlying groundwater.  In this case, source
remediation must be implemented before or during groundwater remediation
if the rate of site remediation is to be accelerated above that which occurs
naturally due to source area weathering. On lihe other hand, there may be
little value in treating low levels of soil contamination when the primary risk
at a site is from groundwater which is migrating toward an off-site receptor.
Knowing when and how to integrate soil and groundwater treatment systems
must be based on the distribution of contamination and the overall clean-up
objectives for the site. On most sites, the objective will be to reduce the risk
of contamination at the lowest possible cost within an acceptable timeframe,
i.e., to accelerate the rate of contaminant removal to reach specific target
levels in each medium in the shortest time that is economically acceptable.
   Sites which contain concentrated amounts of free-phase product or re-
sidual saturation are often remediated in sequential phases, with each phase
focused on a particular contaminated media. For example, free product re-
moval normally precedes the installation of a groundwater biological treat-
ment system. Optimized remediation systems will focus on removing the
most accessible mass of risk-related contaminants first, and then shift to
more dilute contaminants during a later phase. In this way, the greatest po-
tential risk is removed for the least cost while each additional increment of
                                7.1
-------
                              Integrated Technologies
1 , 	 ,;
                  Jilt"
 risk reduction is achieved at a higher cost. With the development of new
 remediation technologies and the growing acceptance of intrinsic biological
 remediation (Section 5.3) as a "polishing" step for impacted groundwater,
 more sites are being remediated using either a parallel or sequential treat-
 ment train approach.
    The parallel treatment of impacted soil and groundwater media has his-
 torically been carried out using separate technologies (Brown and Sullivan
 1991). Contaminated soils are often treated in place using bioventing (Sec-
 tion 4.2) at the same time that in situ groundwater treatment is enhanced
 using the Raymond process (Section 5.2). A treatment train that is becoming
 standard practice is the use of bioventing to reduce the long-term leaching of
 degradable contaminants in the vadose zone, while intrinsic biological
 remediation (Section 5.3) is used to complete risk reduction in a stabilized
 groundwater plume. For sites with small quantities of contaminated soils,
 excavation and aboveground biological treatment of the source (Sections 4.3
 and 4.4) are also effective methods of reducing contaminant loading to the
 groundwater and accelerating natural biodegradation of the dissolved plume.
    A summary of possible technology combinations for the biological treat-
 ment of soil and groundwater contamination is provided in Table 7.1 as a
 function of the nature and distribution of contamination existing at a site.
 The specific selection of a given component within the treatment train de-
 pends upon the magnitude of the impact on a given media, i.e., concentra-
 tions above MCL or risk-based action level, and whether a receptor is or will
 be impacted, i.e., concentrations above MCL or risk-based action level at a
 downgradient well, within a basement or utility corridor, etc.  A more ag-
 gressive treatment component is generally selected if recoverable, free-phase
 product exists at a site or if concentrations of contaminants of concern are
 high in one or more media, and if an existing or imminent threat to a sensi-
 tive receptor can be demonstrated.  Table 7.2 summarizes a number of stud-
 ies that have recently been presented in the literature that demonstrate the
 integration of product recovery, soil, and groundwater remediation technolo-
 gies for accelerated and enhanced site remediation.
   Several technologies have been developed during the past decade to treat
more than one media with concurrent processes. Bioslurping technology
described in Section 7.3 is a good example of concurrent treatment designed
into a single treatment technology. Using the bipslurping technology, con-
taminated groundwater and free product can be removed while air is drawn
through contaminated unsaturated soils to stimulate in situ bioventing.
        t , ' „    ,„ ,,' ,|«    •, • -     , ',,1   •' ,• rt ,  '"Hi, •!	,' ,   I 'I , II  I  "'„    ",', '!",   •    1° 	
Biosparging (Section 5.5) is another example of concurrent treatment using a
single technology. Air introduced below the water table provides some
                                                              7.2
-------
                                                                    Chapter 7
                                  Table 7.1
             Technology Combinations for Biological Treatment
                  of Soil and Groundwater Contamination
 Nature and Distribution
   of Contaminants
 Free Product Recovery
                       Soils Treatment
                                       Groundwater Treatment
 Recoverable free-phase
 product and soil and
 groundwater
 contamination
 Little free-phase
 product, residual-phase
 soil contamination, and
 groundwater
 contamination

 No free-phase product,
 residual-phase soil
 contamination, and
 groundwater
 contamination

 No soil contamination,
 moderate groundwater
 contamination
Dual-phase extraction
wells; OR Bioslurping;
AND
Product skimmers; OR
Passive wicking
Not needed
Not needed
                   Bioventing; OR Partial
                   excavation and
                   landfarming/biopiles;
                   AND
                   Bioventing; OR Partial
                   excavation and
                   landfarming/biopiles;
                   AND
                   Bioventing; OR Partial
                   excavation and
                   landfarming/biopiles;
                   AND
                   Not needled
Raymond Process; OR
Biosparging; OR
Pumping and
aboveground air
stripping or biological
treatment; AND/OR
Natural attenuation
Raymond Process; OR
Biosparging; OR Oxygen
release compounds;
AND/OR Natural
attenuation

Raymond Process; OR
Biosparging; OR Oxygen
release compounds;
AND/OR Natural
attenuation

Biosparging; OR Oxygen
release compounds;
AND/OR Natural
attenuation
oxygen to enhance the aerobic biodegradation of dissolved hydrocarbons.  In
addition, contaminants volatilized with the injected air enter the unsaturated
zone where they can be degraded by aerobic organisms which use the oxy-
gen not transferred to the sparged groundwater in a bioventing mode.
                                             i
   Table 7.3 provides a matrix summarizing site, soil, and waste constituent
characteristics that are considered important In the evaluation and design of a
number of selected integrated technology treatment trains and the
bioslurping technology listed in Table 7.2.  For the treatment trains, Table
7.3 was generated using the site, soil, and waste limitations identified for
each individual technology in the treatment train from the corresponding soil
and groundwater treatment technology tables in previous sections of this
monograph, Tables 4.1 and 5.1, respectively.

   The bioslurping technology is an in situ one, and as such, the technology
is affected greatly by site conditions that negatively impact the uniform dis-
tribution and transport of product being recovered and reactants (air) being
distributed throughout the contaminated site 
-------
Integrated Technologies
                            •'   Table 7.2     '
             Example Case Studies Demonstrating the Use of
             Integrated Technologies for Biological Treatment
                 of Soil and Groundwater Contamination
Technology Combinations
SVE/Bioventing
Raymond Process/Liquid
Biorcactor/Intrinsic Remediation
SVE/Free Product Recovery
Raymond Process/Soil Cell
Air Sparging/Bioventing
SVE/Raymond Process
Biopiles/Raymond Process
Biopiles/Liquid
Bipreactors/Raymond Process
Vapor Biorcac tor/Liquid
Bioreactor
Raymond Process/Biosparging
SVE/Bioventing/Biosparging
Raymond Process/Soil Cells
Bioslurping Technology
Applications
.1 ' • . :' , • 1! ' . . ,' "
Contaminant Type
JP-4
BTEX
Chlorinated VOCs
TOC
11 ' ''!;.'
Gasoline
Gasoline
Mixed Fuels
Toluene
Diesel and Gasoline
Mixed Fuels
Gasoline
Diesel
Gasoline
Oil Refining
PCP/PAHs
BTEX
Natural Gas Site
Gasoline, Varsol
No. 2 Fuel Oil
Jet Fuel
Gasoline and Diesel
Jet Fuel
Jet Fuel
„ ' :
1111 '"
References
Dupont et al. 1990
Thompson et al. 1995
Norris et al. 1990
Dey et al. 1990
Lord et al. 1995
Brown et al. 1995
Ratz et al. 1993
Jarvanmardian et al. 1995
Martinson et al. 1993
Rhodes et al. 1995
Norris etal. 1993
Ellis 1994
Piotrowski et al. 1994
Marsman et al. 1994
Raetz and Scharff, 1995
Dey et al. 1996
Leeson et al. 1995
Connolly, et al. 1995
Hoeppel et al. 1995
Kirshner et al. 1996
Operating
Mode: Parallel
or Series
Series
Series +
Parallel
i| y,1, v
Parallel
Parallel
Parallel
Parallel
Parallel
Parallel
Parallel
Parallel
Parallel
Parallel
Parallel
Parallel
:. i 	 	 	
• '-.; 	 .' ,;,.'" I ' ; ' - "' ' • " " !• ;.'• 	
groundwater recovery, while also moving air through the vadose zone.  It
incorporates bioventing into its operation, and as with bioventing, air perme-
ability and high water content in the vadose zone are both important soil
characteristics affecting system performance. Finally, performance of the
technology will be affected by additional soil conditions impacting microbial
activity (i.e., non-ideal water content and soil pH limitations), and by waste
constituent characteristics that affect a contaminant's toxicity, biodegradabil-
ity, and bioavailability.
   The following section provides additional discussion of applications and
limitations of bioslurping technology identified in Table 7.1 for the inte-
grated treatment of a site containing free product, residual-phase material,
and contaminated soil and groundwater.

                    •  -'  •   '  •   7.4         '                    ;
-------
                       Chapter 7

Table 7.3
Impact of Various Site, Soil ancl Waste Constituent
Characteristics on the Performance of Various
Treatment Trains and Bioslurping Technology Performance*

Site Climatic Conditions
GW Table Fluctuations
Surface Structures
Layered Formation
Product Existence/
Distrubtion
• Soil Fine Grained
High Soil Water Content
• Low Soil Water Content
Nutrient Limitation
Adverse pH Conditions
Waste
Constitutent Volatility
Biodegradability
Bioavailability
Water Solubility
Toxicity
*SVE = soil vapor extraction
BV = bioventing
AS = air sparging
RP = Raymond process
SC = soil cells
SVE/BV
N
I-
1+
V-

v+
I
V-
I-
N
I-

1+
I
I
I-
I





ASfflV
N
I
I
V-

v+
V-
N
N
I
I

V
V
V
I
V-





SVE/RP
N
I-
1+
V-

v+
V-
V-
I-
I-
p

1+
V
V
I-
V-





SC/RP ,
I
I
N
V-

V+
V-
I
I-
I-
I-

I ;
V
V
I-
V-





BSI
N
I
I
V-

v+
I-
V-
I
N
I

v+
V
V
I
V





BSI = bioslurping .
N = not important related to the performance of the technology
1 = important related to the performance of the technology
V = very important related to the performance of the technology
+. = characteristic positively impacts the performance or selection of the technology
- - characteristic negatively impacts the performance or selection olf the technology
7.5
-------
                                  Integrated Technologies
it '„,'   '          ',         ' '	  ,.!»                     ' 	  -i
                                  7.2  Bioslurping

if  •  '.  '   '  ."      •" -  !-  '	^      •   ,':,      .        ,       r  '       \    ';..!.. I.  ^  " '.'."'    •    •' |f '
                                  7.2.1  Principles of Operation
                                    Bioslurping is an integrated in situ process that combines vacuum-en-
                                  hanced dewatering and free product recovery techniques with bioventing for
                                  the simultaneous recovery of LNAPL, contaminated groundwater, and con-
                                  taminated soil vapor; and the transfer of oxygen to stimulate the aerobic
                                  degradation of contaminants within the dewatered capillary fringe and unsat-
                                  uratedzpne.  A typical configuration of a bioslurping system is shown in
                                  Figure 7.1.
                                    As indicated in Figure 7.1, a typical bioslurping system consists of the
                                  following components:
                                         •  a small diameter (typically 5-cm (2-in.)) PVC bioventing well
                                            screened across the yadose zone, capillary fringe and groundwa-
                                            ter table; and sealed at the surface for maintenance of a vacuum
                                            within the well;
                                         •  a smaller diameter (typically 2.5-cm(l-in.)) PVC suction tube
                                            located within the bioventing well, and placed for extraction at
I!  ih     '       	          'i    '                    , :L ,„!".•' l»  ,!"•",   ',i ''   •  '' 7^ „, ,,     i  I,A,,,    , ,,  •  i   	,,  ,	
                                            the LNAPL/groundwater interface;
                                                                              j
                                         •  a high vacuum liquid ring pump for extraction of liquids and
'	';••:      '     "••••"  .       "''   ,„'•!               vapors;
                                         •  an oil/water separator (OWS) for the gravity separation of recov-
                                            ered LNAPL and groundwater;
              1	          ''"   	                                 ' '    	    ' '  'I '   ' ' 	 '        'J	  '
                                         •  OSW effluent holding tanks for recovered product and separated
                                           groundwater; and
              I.1'         "'!	   •'  .                             '    .. -  ' .  "i ' " :,' !•!•  'j     "•   .'..:..   •  ; ..' ::  „:
                                         • groundwater and vapor treatment systems as required.

                                 7.2.1.1 Vacuum-Enhanced Free Product Recovery
                                    Vacuum-enhanced free product recovery (FPR) uses a vacuum on a well
                               •  point to increase the hydraulic gradient and improve the transmissivity of the
                                 aquifer in me capillary fringe to increase rate of flow of LNAPL and soil
                                 vapor into the well(AFCEE 1995). Vacuum-enhanced FPR is an improve-
                                 ment overconventional FPR.systems usingsingle or dual pumps, as conven-
                                 tional systems increase hydraulic gradients by creating a cone of depression
                                 around the extraction well, causing smearing of product and reducing the
                                 saturated thickness of the formation. With vacuum-enhanced FPR, minimal
                                 liquid drawdown occurs due to the production of reduced pressures around
                                                                      ••  •      j    •    •            '	
        	              "                                •    7.6
IIJL ' nij,	!'

I n
	 iil!
Vi i	
-------
                                                              Chapter 7
the extraction well, resulting in horizontal rather than diagonal flow through
more permeable horizontal flow paths. Increased hydraulic gradients with-
out a reduction in saturated thickness using vacuum-enhanced FPR results in
improved liquid flow rates and improved LN^VPL recovery.
                                Figure 7.1
                Schematic of a Typical Bioslurping System
             Compression Screws
              Metal Plates


r
i
Tee
i r



Valve
T

                                                            - 6 in. Header
            1 in. Suction Tube
       Free Phase Product
                                          — Water Table
Source: AFCEE1995
                                   7.7
-------
 Integrated Technologies
                                 Figure 7.1 cont.
                   Schematic of a Typical Bioslurping System
                                                      Vapor Treatment
                                                       or Discharge
   Holding Tank •«
                •Totalizer
                                 Hand Pump
                           Holding Tank

r


Bioslurper


Liquid
Ring
Seal
Pump
SJ-
                                                                          From
                                                                         • Bioslurper
                                                                          Well
       Discharge
                           •• Sampling Port
       Activated Carbon Vapor Treatment System
                                          Oil/Water
                                          Treatment
                                          System
6                                                       Pressure Drop
                                                       Measurement


                                                     T
                                                    Vapor
                                                     I
n
                                                             Liquid
                                                             Ring
                                                             Seal
                                                             Pump
Source: AFCEE1995
7.2.1.2  Bioventing
                                      :•::  • ,;	: "       ij       •   • •     • ;,       *./:	
   The principles of operation and design of bioventing systems have been
detailed in Section 4.2. As in a strict bioventing system, bioventing pro-
cesses integrated into a bioslurper system use a vapor extraction system to
                                       7.8
-------
                                                           Chapter 7
move oxygen through the unsaturated zone and enhance the aerobic
degradation of contaminants within the unsaturated zone and throughout
the capillary fringe. The use of conventional air extraction bioventing
results in upwelling of the groundwater table and a reduction in air flow
through the capillary fringe, the area typically containing high residual
masses of contamination. Groundwater upwelling is eliminated with the
use of the center suction or "slurp" tube incorporated into the bioslurper
system design. This design addition allows the bioventing of the capil-
lary fringe in extraction-mode bioventing systems, and significantly
improves bioventing system performance when air injection bioventing
systems cannot be utilized at a site.

7.2.1.3 Integrated Bioslurper Technology
   As indicated above, bioslurper technology combines both vacuum-as-
sisted FPR and bioventing technology concepts to provide an integrated
approach for the removal of free product, contaminated groundwater, and
contaminated soil vapors — while enhancing the recovery of mobile and
residual-phase product and the in situ, aerobic; treatment of unsaturated zone
and capillary fringe contamination through the application of a high vacuum
and use of a slurper tube in a small diameter extraction well.
   The bioslurping system is designed to minimize the extraction of ground-
water and soil gas by operating at low extraction rates with minimal product
and groundwater drawdown using the slurper tube to extract mixed ground-
water and soil gas vapors at the groundwater interface in the characteristic
"slurping" mode. A comparison of a conventional dual pump LNAPL recov-
ery system to that of a bioslurping well is presiented in Figure 7.2. The
slurping action describes the movement of slugs of air/groundwater/product
mixtures up the center slurper tube as high vacuum (up to 51 cm  (20 in.) of
mercury) is applied to the extraction well. This slurping action also in-
creases the theoretical maximum suction lift ~ 8m (« 25 ft) of liquid from
the groundwater table since the extracted fluid is a mixture of groundwater,
soil gas, and free product, resulting in liquid fluid entrainment.

7.2.1.4 Technology Applications and Limitations
   Keet (1995) summarized the applicability of the bioshnping technology
as follows:
       •  recovery of free product from the groundwater table in fine- to
          medium-fine sediments and fractured rock;
                                 7.9
-------
Integrated Technologies
            enhanced recovery of free product in formations where NAPL
            mobility is limited; and

            enhanced removal of aerobically degradable contaminants in the
            unsaturated zone and capillary fringe overlying free product lay-
            ers via bioventing.
                                  Figure 7.2
       Comparison of Conventional LNAPL Recovery and Bioslurping
              Conventional
        Water Treatment/Discharge •*-

               Oil/Water Separator"
             No Airflow
   Oil Smear Zone in Cone of Depression -
 Flow Due to
 Pressure-Induced
 Gradient
  Groundwater Depression Pump
                         Oil
                         Skimmer —'
                         Pump
  Bloslurper
                                               Air Treatment
                                               or Discharge
       . Oil/Water
       Separator
                                                              How Due to
                                                              Pressure-Induced
                                                              Gradient
                                                      Horizontal Flow Lines •<— Oil
      Airflow in
     Vadose Zone
Groundwater
Source: AFCEE1995
                                     7.10
-------
                                                            Chapter 7
   The bioslurping technology has the following advantages over conven-
 tional free product recovery and remediation techniques that are characteris-
 tic of innovative technologies designed with technology integration in mind
 (Keetl995):
        •  free product recovery is enhanced in all medium-grained materi-
           als, particularly medium- to fine- grained materials, due to high
           vacuum operations;
        •  groundwater depression is practically eliminated due to the use of
           the slurper tube, resulting in little or no change in the relative
           permeability of the capillary fringe to residual product, produc-
           ing enhanced product recovery rates compared to conventional
           FPR systems;                 !
        •  operation under a vacuum results in recovery of residual product
           not mobile under atmospheric pressure conditions; and
        •  operation under a vacuum results in the movement of soil gas
           through the vadose zone and capillary fringe, stimulating volatil-
           ization of contaminants and oxyg;en transfer for the stimulation
           of in situ aerobic degradation of contaminants in a bioventing
           mode. Significant improvements in system performance can be
          expected in a bioslurper system as compared to conventional air
          extraction bioventing despite operating at high vacuums, particu-
          larly in the capillary fringe, due to the control of groundwater
          levels in a bioslurper system with the use of the slurper tube.
   Two disadvantages of the bioslurping system have been identified (Keet
 1995) as follows:
        •  high velocity extraction of groundwater/vapor/NAPL mixtures
          can form emulsions that are difficult to separate in post-extrac-
          tion OWS units; and
        •  bioslurping systems generate not only a groundwater and free
          product stream, but also a contaminant-laden vapor stream that
          must be treated prior to discharge at many sites (see Figure 7.2),
          adding to the complexity of a frees product recovery  system.

7.2.2 Process Design Principles
   Most of the information regarding bioslurper design comes from the
AFCEE Test Plan and Technical Protocol document (AFCEE 1995) describ-
ing field treatability testing of the technology.  The reader is referred to this
document for complete details regarding recommended procedures for

                                 7.11
-------
Integrated Technologies
   ,:.      .            .    ,i  <'     . .1   ,n ' '  :• i i,  '  •, • .I       '         I ,| n.,   '"'I ',
conducting a field treatability test for bioslurping prior to final field-scale
design.  The following information is a summary of recommendations from
the AFCEE protocol document, augmented with additional reference mate-
rial as noted.
  •   ''  ;'    '    • '    '   !•   '.  . • ""»;!'  '• j  '• .. • I1',  j '••'•  ••'   '  '        '  i.:i  '
7.2.2.1   Extraction Well and Slurper Tube
   A bioslurper extraction well consists of a 5- to 10-cm (2- to 4-in.) Sched-
ule 40 PVC well (Figure 7.1) with a slot size allowing free soil gas flow
without transporting fines into the well."in applications where the ground-
water table is greater than 10 m (30 ft) below ground surface, a 10-cm (4-
inch) diameter well is recommended. The screened interval begins a maxi-
mum of 1 m (3 ft) above the water table and generally extends 3 or more
meters (10 or more feet) into the water table depending upon the thickness of
contamination within the saturated zone and the  seasonal fluctuation in
              •„!' , :»i   ,,     ,       	,.  „  " if  , r i 'f .n I 	   ,n    ' ,          i ,
groiindwater table elevation.
   Extraction well installation via hollow-stem auguring is recommended,
with the diameter of the auger hole at least two times the outside diameter of
the vent well.  The annular space surrounding the slotted section of the ex-
traction well should be filled with silica sand, with the annular space above
the screened interval completed with bentbnite grout to seal the well from
the surface and prevent short-circuiting within the well annulus. A minimum
1 m (3 ft) grout seal is recommended.
   A 2.5-cm (1-in.) PVC suction or slurper tube  is placed within the outer
extraction well and is sealed in some fashion with O-rings or compression
gaskets to produce an air-tight seal on me surface" at the wellhead. Figure
7.1 shows a typical bioslurper well with metal plates, rubber gaskets, and
compression screws to complete the airtight seal around the slurp tube. This
figure also suggests the  use of a "T" and valve on the wellhead so that a
variety of operating modes can be evaluated for the bioslurper well during
field treatability testing  (Section 7.2.2.5).

7.2.2.2  Liquid Ring Pump
                                             i
   Liquid ring pumps are suggested for bioslurper applications because of
their efficient performance at high vacuums, i.e., 74 cm (29 in.) of mercury,
and they are inherently explosion-proof.  This latter characteristic is very
important as bioslurping systems yield combinations of water, NAPL, and
vapors that potentially can be explosive]  A variety of pump sizes are avail-
able from a variety of vendors, ranging from 3-hp to 10-hp models, and
should be selected based on site-specific needs.  Single bioslutper well
- .   '"' -."        • '•        •       '	•	•                •
                                  7.12
-------
                                                            Chapter 7
 installations with groundwater depths less than 7.6 m (25 ft) would allow the
 use of 3-hp liquid ring pumps, while multiple wells and greater depths would
 require higher horsepower models. For example, the NAS Fallen site with
 groundwater depths ranging from 1.5 to 3.7 in (8 to 12 ft) utilized 48
 bioslurper wells over a 0.4 hectar (1-acre) area and required a 10-hpiliquid
 ring vacuum pump to drive the system (Hoeppel et al. 1995).

 7.2.2.3  Oil/Water Separator           i
   Operation of a bioslurping system will result in the recovery of a mixture
 of groundwater, LNAPL, and soil vapor that must be separated into its indi-
 vidual components prior to their treatment, storage, or discharge.  Because of
 high upflow velocities that may result from high vacuum applications and
 high flow rates required when groundwater depths are greater than 8 m (25
 ft), stable emulsion formation and problems related to its subsequent separa-
 tion can be problematic. All bioslurping  systems will require, at a minimum,
 a gravity oil/water separator (OWS) with a 1 to 2 gpm/bioslurper well capac-
ity. See AFCEE (1995) for recommended bioslurper system equipment.
   As indicated in Figure 7.1, recovered product drains by gravity into a
 holding tank, while the separated groundwater drains by gravity as a
 subnatant into an effluent transfer tank prior to its removal. These holding
 and transfer tanks must be sized based on site-specific limitations that gov-
 ern the frequency of product removal and the design flowrate of an aqueous
 treatment system if one is required.
   Connolly et al. (1995) describes the effluent treatment system which uses a
 conventional coalescing-plate OWS for the removal of >20(Jm size oil droplets,
 followed by an additional three-layer (medium-grade chopped fiberglass mat-
 ting, medium-grade steel wool, surgical cotton wool) to coalesce droplets of
 <20 (Jm in diameter. This OWS system is contained within a 200-L (52-gal)
 tank suspended within a 1,000 L (264 gal) holding tank. Water separated from
 the recovered oil is pumped into a biofilter for treatment prior to disposal. This
 system is designed to treat liquid from 11 recovery wells at an approximate
 flowrate of 5 L/min (1.3 gal/min).

 7.2.2.4 Groundwater Treatment
   When possible, groundwater that is recovered from a bioslurping system
 and which has gone through oil/water separation should be discharged di-
rectly to a local sanitary sewer. The volumetric flow rate of this recovered
groundwater should be low, typically a maximum of 0.5 to 1 gpm/bioslurper
well, and will contain less than 20 mg/L total petroleum hydrocarbons
                                 7.13
-------
Integrated Technologies
(TPH) leaving the OWS. If local ordinances will allow, direct sanitary sewer
discharge of this low volume, low organic loading waste stream is a cost-
effective alternative to on-site groundwater treatment that should have little
impact on the POTW receiving this bioslurper aqueous stream.
   Where a sanitary sewer is not readily available for discharge, or where local
ordinances do not allow such an untreated discharge, a variety of treatment
options can be used for this aqueous steam. Aboveground bioreactors de-
scribed in Section 5.4 are one option. Activated carbon can also be used, and is
the method of treatment recommended in the AFCEE (1995) protocol. Addi-
tionally, both water and extracted vapor treatment can be provided in a single
biofilter reactor as reported by Connolly et al. (1995) and described below.
                                 . ".   "/ J .   " ! ,1 l|  !'  •'   ":!'	    ' ,' ,1 '» "I  "| 	*
-------
                                                            Chapter 7


contaminated vapors.  The reader is referred to Section 4.2 for design details
of bioventing systems.
   Treatment of extracted vapors using vapor-phase activated carbon treat-
ment is also a viable option for bioslurping systems. Typically, a series of
two or three 91 kg (200 Ib) carbon canisters sire used to treat the bioslurper
offgas. Flow is passed through the canisters until complete breakthrough is
observed in the first canister. At this point, the first canister is removed from
service and disposed or regenerated, the second canister then becomes the
first canister in the treatment train, and a new, second or third canister is
added to the effluent stream. In this way, maximum use is made of the sorp-
tion capacity of the carbon, and essentially complete treatment of the vapor
stream is provided.  Figure 7.1 shows a typically activated carbon treatment
system used for bioslurper offgas treatment.
   The U.S. Air Force has evaluated the use of internal combustion engines
(ICEs) for the treatment of offgas from bioslurper systems (AFCEE 1995)
and has found them to be efficient and cost-effective, particularly if only
temporary offgas treatment (i.e., during initial operation of the bioslurping
system) is required. These ICEs have special computer-controlled carburetor
systems allowing them to run solely on combustible contaminant vapors
when extracted concentrations are high. Natural gas or propane is used as a
make-up fuel for these units during ICE startup and when concentrations fall
below combustible levels in the bioslurper offgas. Trailer-mounted units
from RSI, Inc. are available for bioslurper applications.
   An additional vapor treatment system that has been used in bioslurper
applications was described by Connolly et al. (1995) and consists of a
biofllter system designed to treat both the aqueous and vapor streams recov-
ered from a bioslurping system. Figure 7.2 shows this combined water and
vapor treatment system hi which vapor from a liquid/vapor separator is in-
jected into the base of the biofllter while the aqueous effluent from an OWS
was dispersed over the surface of the sand-media biofllter. The biofilter used
in this  application had a volume of 17 m3 (55.8 ft3), a depth of 1.75 m (5.74
ft), and a water contact time of approximately 60 minutes. The system oper-
ated in this mode produced non-detect levels of volatiles in the effluent va-
por stream and of TPH and BTEX components in the treated liquid.

7.2.2.6 Field Instrumentation and Monitoring
   Field monitoring requirements for bioslurping systems are similar to
those described in Section 4.2.2 of this monograph for bioventing systems
and include the use of soil vapor monitoring probes to measure of soil gas
pressure, and soil gas oxygen, carbon dioxide, and contaminant
                                         I
                                 7.15
-------
Integrated Technologies
concentrations. In addition, groundwater level and LNAPL thickness mea-
surements must be collected during bioslurper operation using an interface
probe and vacuum-tight well seal as shown in Figure 7.3.
         	   ,  .              <         ,                    ..  ,           •
                                 Figure 7.3
       Schematic of a Vacuum-Tight Interface Probe System Used for
     Groundwater and LNAPL Level Measurements in Bioslurper Systems
               Interface Seal
               with Teflon™
               2 in. Socket by
               2 in. Male Pipe
               Thread
                                                 Interface Probe Lead Wire
       Quick Connect Fitting
       for Vacuum Readings
                   Soil
               Fuel Layer
              Water Layer
                                               1 in. Aluminum Conduit
                                               1 in. Male Pipe Thread by
                                               1 in. Compression Fitting with
                                               Teflon™ Furrule
                                               Interface Probe Tip
Source: AFCEE1995
                                   7.16
-------
                                                             Chapter 7
 7.2.2.7 Field-Scale Treatability Testing
    Because of the significant effect subsurface site and soil conditions and
 the actual distribution of residual product have on the recovery of mobile
 free product and residual phase material, the Air Force (AFCEE 1995) devel-
 oped a field-scale bioslurping treatability test protocol designed to
 evaluate bioslurping effectiveness and a variety of other free product recov-
 ery methods for the recovery of free product on a site-specific basis. The
 reader is referred to the AFCEE (1995) document for details of the testing
 protocol. A summary of the testing method is provided below.
    As indicated in Table 7.4, the AFCEE bioslurping protocol consists of a
 sequence of operations with a single bioslurping well installation to evaluate
 the recovery potential for free product using a variety of recovery techniques
 ranging from simple bailing, to skimming, to drawdown pumping, and fi-
 nally, to high vacuum bioslurping operation.  The testing is designed to be
 conducted over a two-week period using portable vacuum extraction,
 groundwater pumping, and effluent treatment equipment.  Initial activities
 are used to assess the "recoverability" of LNAPL at the site through an ini-
 tial baildown test, and to assess the bioventing potential for degradation of
 contaminants in the unsaturated zone through monitoring point installation
 and limited soil gas sampling to collect evidence of bioactivity within the
 unsaturated zone, i.e., oxygen depletion and carbon dioxide production
 within the vadose zone.
   With completion of these preliminary site assessment activities, a 10-day
 field bioslurper pilot test is specified which involves 2-day operation of the
 system in a skimming mode without vacuum (Figure 7.4), 4-day operation in
 the vacuum-enhanced bioslurper mode (Figure 7.5), 1-day operation in a
 repeat skimming mode without vacuum, and finally, a 2-day groundwater
 depression operating mode (Figure 7.6). Soil gas composition, free product
 thickness, and groundwater elevation measurements are made throughout the
 10-day pilot test, as are the cumulative volumes of extracted free product,
 groundwater, and soil  gas. The latter measurements are required to  deter-
 mine the overall effectiveness of each free product recovery method and the
 contribution to overall free product of the bioventing component of the
 bioslurping system.
   Selection and design of a full-scale product recovery and site remediation
 system that is optimal  for the site can be made based on the outcome of this
field treatability test. From treatability test results, sites showing only mar-
ginal improvements in product recovery for bioslurping systems over simple
skimming methods (i.e., Boiling AFB, DC Bldg. 41 site in Table 7.5) can be
differentiated from those sites in which vacuum-assisted product recovery

                                 7.17   !
-------
Integrated Technologies
                                 Table 7.4
            Schedule of Activities for a Typical Bloslurper Field
              Treatabillty Study as Recommended by AFCEE
                    Pilot Test Activity
                                                     Schedule
         Sife-Specific Test Plan Completed
         Test Plan Approval (when required)
         Mobilization
         Site Characterization
          Product/Groundwater Interface Monitoring
          Baildown Tests
          Monitoring Point Installation
          Soil Gas Survey
          Soil Sampling
         System Installation
         Test Startup and Operation
          Skimmer Test (2 days)
          Soil Gas Permeability Test
          Bioslurper Vacuum Extraction Test (4 days)
          Skimmer Test 2 (1 day)
          Drawdown Pump Test (2 days)
          In Situ Respiration Test (4 days)
         Demobilization
14 days prior to approval
Prior to Pilot Test
Days 1-2
Days 2-3
Days 2-3
Day 4
    i
Days 4-5
Day 6
Days 6-10
Dayil
Days 12-13
Days 10-14
Days 13-15
Source: AFCEE 1995
and bioventing using bioslurping technology significantly accelerates source
treatment and product removal (i.e., Boiling AFB, DC Bldg. 18 and Travis
AFB, CA JFSA-1 sites in Table 7.5).

7.2.3  Process Flow Diagrams
                                                  ,|
   A typical bioslurper system process flow diagram has been presented in
Figure 7.1. All systems will have the same general configuration, with slight
variations in their layout depending upon whether vapor and liquid effluent
treatment are required, and on the specific  treatment option selected, if they
are heeded. The reader is referred to Section 4.2 for bioventing systems, 5.4
for aboveground reactors, and 6.2 for biofilter systems, all of which provide
options for offgas or recovered groundwater treatment using biological treat-
ment methods.
                                    7.18
-------
                                                                       Chapter 7
                                    Figure 7.4
       Schematic of a Bioslurper System Operating in a Skimmer fyiode
               Compression Screws
                Metal Plates •
                              "1

]

(



Tee
4 r
1 	 h




HValvf
Valve
T



j
                                                                    -6 in. Header

                                             • Rubber Gasket
                       2 in. Tee
              1 in. Suction Tube
        Free Phase Product -
1 in. Valve Open
                                                           Land Surface
                                            - 2 in. PVC Bioventing Well
                                                  Water Table
Source: AFCEE1995
                                       7.19
-------
          Integrated Technologies
T.I-.  :'!
                                            Figure 7.5
                           Schematic of a Bioslurper System Operating
                            in a Vacuum-Enhanced Bioslurper Mode
                                                                  Valve
                                                                   T
                        Compression Screws
                         Metal Plates •
                                                         Tee
                                                       1_JH
Valve
                                                                          •• 6 in. Header

                                                    - Rubber Gasket
                                2 in. Tee
                                                           2 in. Valve Closed
                       1 in. Suction Tube
                  Free Phase Product
                                                         Water Table
          Source: AFCEE1995
                                               7.20

-------
                                                                   Chapter 7
                                  Figure 7.<6
                Schematic of a Bioslurpef System Operating
                     in a Groundwater Drawdown Mode
              Compression Screws
               Metal Plates
                                                                -6 in. Header
                      2 in. Tee
             1 in. Suction Tube
     Free Phase Product
              Water
2 in. Valve Open
                                                       Land Surface
                                         - 2 in. PVC Bioventing Well
                                                  f
                                               Drawdown
Source: AFCEE1995
                                    7.21
-------
Integrated Technologies
                               Table 7.5
 Summary of Bioslurper Field Treatability Performance for Selected Sites


' Site Location
Boling AFB, DC, BIdg. 18
Boling AFB, DC, BIdg. 41
Travis AFB, CA, JFSA-1

Site Location
Boling AFB, DC, BIdg. 18
Boling AFB, DC, BIdg. 41
Travis AFB, CA, JFSA-1


Depth to Water
(ft)
23.65
19.06
8.7
Aver
2-Day Skimmer
Test
16.9
0.86
0
i
1
Initial Product
Thickness (ft)
1
-' -444' ": : •'
	 ""034
	 1
Skimmer
Active
age Fuel Recovery
4-Day
Bioslurper Test
	 - .1
59.8
1.14
3.85
Thickness After
24 hour
Baildown Test
(ft)
352
034
Skimmer
Active
Rate (gal/d)
1-Day Skimmer
Test
8.2
NA
0


Well Diameter
(in)
2
4
6

2-Day
Drawdown Test
312
0.13
3.76
       Site Location
  Soil Gas     Vadose Zone
  Radius of     Biqdegradatioii
Influence (ft)   Rate (mg/kg/d)
Boling AFB, DC, BIdg. 18

Boling AFB, DC, BIdg. 41

Travis AFB, CA, JFSA-1
                           454755.3
                                        NA
      0.86     12.9 to 15.3
            	.,'.  , - i   I
      0       61 to 82
NA = Test not performed

Source: Kittle at at. 1996
7.2.4 Process Modifications
   As indicated in Figures 7.4 through 7.6, a bioslurping well can be oper-
ated in a variety of modes by changing the vacuum applied at the wellhead
and by adjusting the depth of the slurper tube with respect to the LNAPL/
groundwater interface. The standard mode of operation is with a. high
vacuum applied at the well and the slurp tube located at the static product/
groundwater interface to encourage horizontal flow of product into the well
as indicated in Figure 75. When product recovery rates fall off as recover-
able product is removed from the subsurface over time, bioslurping systems
                                                ,|          „            , , ,,
                                 .  7.22 ..'	   '  ' "   '.    :    "  "."""';	;;'
-------
                                                            Chapter 7


 can easily be converted to combined groundwater depression and bioventing
 systems by operating in the groundwater depression mode as indicated in
 Figure 7.6. This operating mode can be highly effective in source
 remediation if a significant smear zone exists at a site. Groundwater depres-
 sion via the bioslurping system can effectively dewater the capillary fringe
 allowing oxygen to be transferred to this residual saturation, stimulating its
 aerobic degradation via bioventing.

 7.2.5 Pretreatment Processes      |

   Since this treatment technology is designed to provide groundwater and
 free product recovery and biovent the associated unsaturated zone, no spe-
 cific pretreatment steps are normally required this technology.

 7.2.6 Posttreatment Processes     j

   Because the bioslurping process is carried out in situ, soils are left in
 place following treatment, and posttreatment of soils is unnecessary. How-
 ever, as indicated in Section 7.2.2.3 through 7.2.2.5, there may be significant
 posttreatment process requirements for the various effluent streams based on
 local discharge requirements.  In addition, actions taken for the removal and
 disposal or recycling of recovered liquid product will be governed by local
 requirements and may vary from strict manifesting of the product as a haz-
 ardous waste to the contracting of recycling or disposal services with a local
 waste oil handler.                                           i         .
                                                         •
 7.2.7  Process Instrumentation and Control
   A typical process instrumentation used in a bioslurping  system is shown
 in Figure 7.1. In addition to vacuum pump performance, the mass of con-
 taminant removed in the free phase (LNAPL), aqueous phase (recovered
 groundwater), vapor phase, and via degradation in the vadose zone are also
 monitored over time.  The following describes process monitoring associated
 with a typical bioslurping system.
                                       i
 7.2.7.1 Flow Measurement           i
   Vapor flowrates from the bioslurping system are typically measured using
pitot tube flow indicators. Differential pressure across a pitot tube connected
to the inlet side of the liquid ring pump indicates the total flowrate through
the system. Vapor flow measurements can be estimated based on subtraction
of the product and liquid recovery rates, and/or by measurement of the vapor
stream generated following oil and water separation.

                                 7.23
-------
Integrated Technologies
Groumiwater extraction rates are generally determined using an in-line flow
totalizer on tlie aqueous discharge line from theOWS. Liquid product re-
covery rates are determined using manual or automated gauging of product
levels in the product holding tank or with in-line flow totalizers on the prod-
uct discharge line from the OWS.
             • i	     .    ,•   i     .  •.. i'.^  . '•	ii "i'        .1     . ii     ;.'••
7.2.7.2 Flow Stream Composition "'Monitoring
        • i!  '•••   ; • ''i   '    •  , i '  , . i" : '""•'  ''•*;!' ' "	 i":i  ||  •   i. •    "    r i  '•'  i .1	
   To quantify the mass of contaminant recovered as free product, dissolved
or vapor phase contaminant, or through bibdegradation, concentration data
must be collected for contaminants of concern for each recovery pathway.
             ""* :„."''?•! ',	ii.r	•'	,"» ""••''  ..-.•!'i •!,;;•	, 	i"!! "if •• .•:	!,.i,	t :;.,.n   , '• i	     ;/ •  -h »]'	
   LNAPL samples should be collected for analysis during each phase of the
field treatability test and routinely thereafter throughout the free product
recovery phase of remediation at a site.  Samples are collected using
Teflon™ bailers during the baildown test or from the product holding tank
when it is  used, in glass vials fitted witfi Teflon-lineH septa. Analysis of the
product samples should be carried out, as appropriate, for the analytes of
concern. These analyses are used to quantify the mass of specific contami-
nants recovered as free product and to track the "weathering" of the residual
product material over time in response to bioslurper treatment. The product
sampling frequency should be determined on a site-specific basis, but would
typically be  performed daily during the initial operating period of the
bioslurper system, then reduced to weekly or bimonthly as product recovery
rates decrease.
   Samples for specific compound analysis in the vapor phase are recom-
mended (AFCEE 1995) to be collected using evacuated, 1-L, stainless steel,
Summa polished canisters from the vapor discharge line of the bioslurping
unit.  If vapor treatment is required at a specific installation, both pre- and
posttreatment samples would be required. These data are used to quantify
the mass of contaminant recovered in the vapor phase during bioslurper
treatment and for the evaluation of offgas treatment system performance.
Continuous or semi-continuous vapor monitoring using a non-specific total
hydrocarbon or total halogen detector would be desirable for locations where
strict vapor emission limits are placed on the system. With a non-specific
field detector, a recommended non-continuous vapor phase sampling fre-
quency would be hourly following initial system startup, every 4 hours after
4 hours of operation, every 12 hours after 12 hours of operation, then daily
when system operation stabilized.  Specific compound samples for fixed-
base laboratory analysis would be collected to augment total contaminant
level field measurements at system startup, and at 4 hours, 12 hours,  1 day, 1
week, weekly, and then monthly — once the system operation stabilized —
                                 7.24
-------
                                                            Chapter 7
so that a representative estimate of specific; compound recovery rales could
be made.                                                   '.'
   Aqueous samples for specific constituent analysis are collected from the
bioslurper OWS aqueous discharge line, and from a water treatment system
effluent line, if one is required, using standard 40-ml VGA vials with Teflon-
lined septa. The composition of this flow stream would be expected to
change less rapidly over time than that of the vapor stream, and would be
sampled at a frequency suggested above for the LNAPL product samples.
These data are used to quantify the mass of contaminant recovered in the
aqueous phase during bioslurper treatment, and for the evaluation of aqueous
treatment system performance.          |
                                      i
7.2.7.3 In Situ Biodegradation Rate Determinations
   As indicated in Table 7.5, at some sites vadose zone biodegradation of
residual contamination can be a significant contaminant reduction process.
If contaminants of concern are aerobically biodegradable, quantification of
this contribution to overall contaminant removal can be provided through the
conduct of routine in situ respiration tests (Section 4.2.1.3) using vapor
monitoring points installed for bioslurper performance monitoring. In these
tests, the bioslurper system is turned off and oxygen and carbon dioxide
concentrations are measured (Section 4.2.7.3) over time from soil vapor
monitoring probes throughout the site. Reductions in oxygen concentrations
with parallel production of carbon dioxide indicate biological consumption
of oxygen during the degradation of contaminants within the soil matrix.
Data reduction procedures for degradation rate estimates have been de-
scribed in Section 4.2.9.1. The reader is referred to the Air Force Bioventing
Protocol (Hinchee et al. 1992) for a complete description of in situ respira-
tion rate test procedure.
   Routine monitoring of the oxygen and carbon dioxide levels throughout
the vadose zone and in the vapor discharge from the bioslurping system also
provide quantitative estimates of the mass of contaminant "recovered" in the
vapor phase due to biodegradation reactions. Oxygen concentrations in the
recovered vapor below (and carbon dioxide levels above) background soil
gas concentrations indicate of biological activity in the soil volume from
which the gas is recovered. A quantification of the mass of oxygen depleted
or mass of carbon dioxide produced in the recovered vapor can be converted
to an equivalent mass of indicator contaminant that has been biodegraded
using the stoichiometry for biodegradation known for this indicator com-
pound.  The reader is referred  to Sections 4.2.9 and 5.5.9, and Table 5.11 for
information regarding the stoichiometric conversion of oxygen use to

                                 7.25
-------
 Integrated Technologies
                                               :i       	        	
 equivalent contaminant mass, and to Section 7.2.9 for sample calculations
 related to this mass degradation estimate.
         11        ,          •         	    • i
                              	     ' '    	'  '  jl 1   ,    "       ",„'''.'
 7.2.8 Process and Instrumentation Diagrams

   Figure 7.1 provides a process and instrumentation diagram for a typical
 bioslurping system.
 7.2.9 Sample Calculations
   Calculations routinely carried out for bioslurper systems include: cumula-
 tive contaminant recovery rates in the product, aqueous, and gaseous flow
 streams; contaminant recovery associated with v'acjose' zone biodegradatiori;
 and data reduction for pneumatic pump tests conducted as part" of the field
 treatability study.

 7.2.9.1  Contaminant Recovery Rates
   Contaminant recovery rates within the product are easily determined
 based on the volume of product recovered per unit time, the concentration of
 specific contaminants within the recovered product, and the measured or
 estimated product density using the following relationship:
                _,,',,   _. i  •'     rt  ,   Vp[Contaminant]       ,_ ,,.
                Product Mass Recovery Rate = ——	       (7.1)
                              	 '•  " 	 " ' • '	| •'   t
 where:
             V  =     volume of product recovered per unit time t(m3);
             p   =     product density (kg/m3); and
 [Contaminant]    =     contaminant concentration (mg/kg).
   Contaminant recovery rates in the aqueous phase are calculated in a man-
 ner similar to that in the recovered product, with slight modifications to re-
 flect the aqueous nature of the flow stream as indicated in Equation 7.2:

           Aqueous Phase Mass Recovery Rate = Q[Contaminant]aq     (7.2)

 where:
             Q  =     aqueous stream flow rate (mVtime); and
 [Contaminant]   =     contaminant aqueous concentration (mg/m3).
   Finally, the contaminant recovery rate in the vapor stream is calculated using
Equation 7.2, with Q being the vapor flow rate and [Contaminant] being the
vapor stream concentration, both having the units as indicated above.
                                7.26
-------
                                                           Chapter 7
7.2.9.2 Contaminant Biodegradation
   Contaminant biodegradation rates are normally determined from routine
in situ respiration tests. These tests are conducted during periods when the
bioslurping system is not operating so that the rate of oxygen consumption
can be followed at specific soil gas probe locations over time. Data reduc-
tion techniques for in situ respiration tests were, described in Section 4.2.9.1,
along with an equation (Equation 4.1) to convert oxygen utilization rates
with units of %/hour into biodegradation rates with units of mg/kg/d. Total
contaminant removal due to biodegradation is determined from this method
by multiplying the biodegradation rate by the soil mass on a dry weight basis
for which this biodegradation rate is representative.
   Additionally, contaminant biodegradation can be estimated from oxygen
and carbon dioxide concentration measurements in the recovered vapor com-
pared to concentration of these gases in uncontaminated soil gas. The reduc-
tion in oxygen concentration below or the increase in carbon dioxide levels
above background levels times the vapor flow rate times the stoichiometric
relationship between oxygen use or carbon dioxide production (with proper
unit conversions) allows the determination of equivalent contaminant mass
that is being degraded within the unsaturated zone affected by the bioslurper.
Assuming an estimate based on equivalent hexane mass biodegradation, the
following stoichiometric equation allows the determination of the oxygen
requirement to degrade 1 gmol of hexane:

               CJH1A+19/20, ->6CCX+7H20             (7.3)
indicating that 9.5 gmol oxygen are required, or 6 gmol of carbon dioxide
are generated when 1 gmol or 86 g of hexane are completely degraded.
   Using this relationship, the following expression allows the determination
of the mass of hexane equivalent recovered from the system in the form of
depleted soil gas oxygen or elevated carbon dioxide levels:

             Hexane Equivalent Degraded! =
             f  Q(A%02)   .(1 gmol hexane)!                   (7-4)
             [(22.4 L/ gmol)   (9.5 gmol O2)  J
                                7.27
-------
 Integrated Technologies
              Hexane Equivalent Degraded =
             f  Q(A%C02)  .Ogmol hexane)!
             [(22.4 L / gmol)   (6 gmol CO2) J
                                                    (7.5)
where:
•  '• '•  •     i      •'':'•:'•  :",,,'  '  ""   :"     ;:! ! " . ""  "!"; "	' '!	"
     A%O2  =   change in oxygen concentration compared to background

    A%CO,
   MW,
       hexane
=   change in carbon dioxide concentration compared to
    background (%); and
=   molecular weight of hexane = 86 g/gmol.
7.2.9.3 Pneumatic Pump Test Data
   During bioslurper system field treatability testing, in situ air permeability
measurements are collected at several locations throughout the site to evalu-
ate vapor flow and soil conductivity, along with the radius of influence pro-
vided by each bioslurping well. One approach that has become a recom-
mended standard for in situ soil air permeability measurements was de-
scribed by Johnson et al. (1990) and is based on Darcy's Law and. equations
for steady-state radial flow at a vent well. The method entails the use of a
single vent well with soil vapor probes placed radially and vertically away
from it to monitor soil gas vacuum throughout the field site when air is ex-
tracted at a constant rate at the wellhead.
                                            '  1
   The governing equation for such a system, assuming one-dimensional
radial flow from the extraction well, is shown in Equation 7.6:
                P =
                 -0.5772  - In
                                            r2eji
+ ln(t)
(7.6)
where:
        P'   =  "gauge" pressure (g/cm-sec2) measured at the vapor
                probes some radial distance r (cm) from the vent well at
                time t (sec);
        m   =  vent well screen interval (cm);
         k   =  soil gas permeability (cm2);
         |i   =  air viscosity (1.8* IQr4 g/cm-sec @ 18°C);
                                7.28
-------
                                                             Chapter?
          8  =   soil air filled porosity (decimal %);
         Q  =   volumetric air flow rate at the vent well (cmVsec); and
       Patm  =   atmospheric pressure (1 atm - 1.013 x 106 g/cm-sec2).
   Soil gas pressure or vacuum data collected over time at various vapor
probe locations following initiation of vent: well pumping allow the determi-
nation of in situ soil gas permeability and its variability throughout the site.
Vapor probe readings are plotted as a function of the natural log of time,
generating a straight line with a slope equal to Equation 7.7:


                Slope = — Q
                                                                  (7.7)

   Rearrangement of this equation allows the determination of k directly as:

                       Oil
                k =
                    4 Slope ^m                                   (7.8)
   This approach to data reduction will not be possible if the assumption of
radial flow is not maintained at the field site.  Radial flow will not occur if a
significant vertical air velocity component exists due to shallow contamina-
tion and subsequently a small well screen interval (<10 ft), and if the soil is
coarse grained. Under these conditions, vacuum measured in the vapor sam-
pling points will reach constant values very quickly, requiring that the data
be reduced using Equation 7.9:
                                                                  (7.9)
where:
        Rw   =   the radius of the vent well (cm);
         H   =   the depth to the top of the well screen (cm);
         Rj   =   the minimum radius of vent well influence under steady-
                 state flow conditions (cm); and
        P    =   the absolute pressure at the well head (g/cni-sec2)
                                 7.29
-------
                                               , "I , ,.,'
Integrated Technologies
   Rj can be estimated from inspection of field data, or by extrapolating the
relationship of vapor probe vacuum versus log(r) to a 0 vacuum value.
                                        	i"" !i  ,      ,     '    '    ; ,!'!"!'"

7.2.10  Safely Requirements
   Safety requirements for a bioslurping system are similar to those specified
for bioventing systems (Section 4.2.1G) and bipsparging systems (Section
5.5.10), and involve: (1) safeguards for electrical equipment to prevent explo-
sions, (2) prevention of uncontrolled subsurface vapor migration into confined
spaces, and (3) precautions that must be taken when operating equipment at
high vacuum. In addition, the free product recovery aspect of bioslurping re-
quires that adequate precautions be taken to prevent exposure of site personnel
to this material.', Precautions include the use of personal protective clothing
(safety goggles, disposable gloves, coveralls, disposable boot covers, and appro-
priate respirator) by site personnel when they are sampling or handling free
product material. AFCEE (1995) has developed a general site Health and
Safety Plan for bioslurping field treatability studies that is useful in developing
site safety plans for full-scale bioslurper operations.
                                     	i
7.2.11  Specifications Development
   Specifications for vent wells and monitoring points will generally follow
state-specific drillers' standards for monitoring well installations.  General
specifications for size and materials of construction for bioslurper wells and
monitoring points are shown in Figures 7.1 and 4.4, respectively.
   Bioslurping system piping is generally constructed of PVC, with Sched-
ule 80 PVC generally preferred for shallow burial applications and for con-
nections to individual bioslurper wells. To improve system durability, galva-
nized piping and valves can be used for connections entering and exiting the
blower and for all aboveground piping. ASTM standards should be con-
sulted for general specifications for high-vacuum piping and valves used in a
bioslurping applications.
   Motors and blowers should conform to ASME standard PTC-9andthe
National Electric Code. For operation in potentially explosive atmospheres,
NEC Sections 500-505 also apply.
         •'    .if     ,'          '  " '|j "   	J'iii ,  " |i| "  ' , ':« i11 ' il  „   i;  '  '      ' '    '' r j '. ii.

7.2.12 Cost Data
   Only limited data are available describing the cost of bioslurping systems
for the recovery of free product and contaminated groundwater and
bioventing of vadose zone contamination under full-scale field conditions.
                                 7.30
-------
                                                            Chapter 7
The only study to date reporting cost for field-scale bioslurping systems is
that of Connolly et al. (1995) which described the recovery of gasoline and
diesel fuel from a site with fractured rock with a water table depth of ap-
proximately 4.9 m (16 ft). The system used eleven 2-inch bioslurper wells
with both offgas and groundwater treatment provided by a biofilter using the
system. With this system, 3,900 L (1,030 gal) of product have been recov-
ered for a total design and installation cost of $80,000, and annual operating
costs of $40,000. This product volume does not include vapor phase product
recovery nor product destruction provided by in situ biodegradation in the
vadose zone, so it is a conservative estimate of bioslurper performance at
this site. With a reported liquid volume recovery of 15% to 25% of the esti-
mated total release volume, the unit costs for this bioslurper system were:
$20.50/L ($77.67/gal) design and capital costs, and $10.25/L ($38.83/gal) in
annual O&M costs. Due to the general lack of cost data for this technology,
more data are required to improve these cost figures beyond the preliminary
values presented here.

7.2.13 Design Validation
   As with any engineering design, bioslurper design should be subject to
peer reviews of all assumptions and design calculations. The applicability of
a bioslurping system at a given site should be based on field treatability test-
ing results generated using procedures suggested by AFCEE (1995).  Selec-
tion of a specific operating mode for product recovery and/or bioventing at a
site should be made based on these field treatability data, and further system
design and operating refinements should be made based on results of ongo-
ing monitoring of liquid and vapor streams generated by the bioslurper unit,
and from soil gas and groundwater and product level data collected during
system operation. Decisions regarding the need for additional bioslurping
wells based on actual field determinations versus estimated liquid and vapor
phase radii of influence should be made as field performance data are col-
lected and evaluated. Additional system design modification or operating
mode changes should be implemented as necessary in response to changing
product, groundwater, and vadose zone conditions that develop as product
removal and contaminant degradation proceed at a site.

7.2.14 Permitting Requirements
   Permit requirements for this technology, as with others, can vary signifi-
cantly from state to state. In general, the construction of the bioslurper well
will require a standard well permit.  Some states also require a standard well
permit for vapor monitoring probe installations. In some states, disposal of

                                 7.31
-------
Integrated Technologies
drill cuttings that may contain RCRA-listed wastes can be an expensive and
time consuming byproduct of bioslurping field work.
   Discharge permit considerations for this technology include both air (Sec-
tion 6.2.14) and water (Section 5.4.14) discharges, and requirements for the
proper handling and disposition of free product recovered during
bioslurping. The appropriate permitting authorities should be consulted
regarding local restrictions on liquici ancl vapor discharges, and on acceptable
handling and permitting methods for recovered free product.

7.2.15 Design Checklist
   Table 7.6 provides a list of items that need to be considered in the design
and implementation of the bioslurping process for product recovery and
source remediation at a given site.
                                           , „ , •  l|    . .            •   i „ •
7.2.16 Implementation
                                               i
   It is recommended that implementation of a bioslurping system begin
with a field-scale treatability assessment of product recovery techniques as
described by AFCEE (1995). The primary requirements of the field treat-
ability study are the proper drilling, sampling, installation, and completion of
the bipslurper well and associated system monitoring points.  Next, electrical
service, equipment support pads, storage buildings, security, fencing, light-
ing, access roads, etc., should be provided as necessary to support short-term
field testing that is to take place. A system shakedown consisting of a brief
start-up test should be conducted to ensure that all system components are
operating properly. A system checklist should include:  the liquid ring
pump; aqueous effluent treatment pump;  OWS; vapor, fuel, and water
flowmeters; emergency shutoff float valve on the effluent transfer tank; and
all analytical equipment.  Finally, all appropriate local agencies should be
notified of planned field activities.
                            	i,    ;',     ',; : I II     ...               I "
                                   	j     ' •          .  .  :  . .
7.2.17 Start-up Procedures
   Bioslurper system start-up procedures should be initiated through a field-
scale bioslurping treatability study as described by AFCEE (1995).  This
field treatability test consists of three distinct product recovery tests  and
leads to the selection of tfie optimal operating mode for a given site. A de-
scription of each phase in the treatability testing effort are described below.
                                 7.32
-------
                                                                                    Chapter 7
Site Characterization
Site Features
Field Treatability Test Results
Well Design
Effluent Treatment
Free Product Handling   .
Fluid Transfer System
                                           Table 7.6
                          Bioslurper Process  Design  Checklist
Type and distribution of contaminants and free product
Soil characterization
Delineate contaminated soil, groundwater, and product volume
Aquifer characteristics
Property lines
Nearest power source
Building/confined spaces locations
Underground utilities
Concrete/asphalt surfaces
Normal use and traffic patterns
Adequate production rate of free product
Adequate vacuum enhancement to product recovery
Adequate vadose zone conductivity
Appropriate microbial activity in vadose zone
Product recovery rate adequate
Screened intervals
                                                       •
Well locations
Completion methods
Slurper tube location (operating mode)
AH regulated compounds treated
Precipitation potential addressed
Sufficient capacity
Residuals addressed
Discharge stream quality acceptance
Permitting acquired as necessary
Storage and handling procedures adequate
Storage volume adequate
Containment adequate
Personal  protection and monitoring adequate
Lines deep enough to avoid freezing/heat traced
Check valves, flow meters, control valves
Surge tank controls
In line filters
All prime movers explosion proof
Adequate compatibility with product being recovered
                                               7.33
-------
ii   i  in  	  ii	  "•  ..'	:?!":	"iiTr	ni:  "i"'1:'""   •  •'••  .     •   •  "r i"'"1,,11 '     ••     • n. '•!'-!:	i	n	,."ij	"" Ibi'i	;•	••	ifijs'
                                Integrated Technologies
                                7.2.17.1  Initial Skimming Test
                                   The first product recovery test recommended in the bioslurper treatability
                                test is a 48-hour skimming test. In this test, the slurper tube is set at the prod-
                                uct/groundwater interface with atmospheric pressure maintained on the well-
                                head by leaving open the wellhead ball valve (Figure 7.4).  Prior to starting the
                                pump test, the bioslurper pump and OWS are primed with diesel fuel, and the
                                flow totalizers for the product and aqueous effluent streams are zeroed.
                                                  .   :•  	•'       "  '	s	  <  •'•: ••!•	'•   	  " '   •	
                                   The liquid ring pump is started and free product and groundwater extrac-
                                tion rates are monitored on an as-needed basis throughout the test. Product/
                                groundwater levels are monitored periodically (every 1/2 hour for 2 hours
                                and on an as-needed basis from that: point on) over the 48-hour duration of
                                the test. Final product and groundwater extraction rates and product/ground-
                                water levels are made at the end of this portion of the field test.
                                7.2.17.2  Bioslurper Test
                                   When the initial skimming test is complete, the ball valve on the extrac-
                                tion wellhead is closed so that vacuum levels sufficient to provide
                                                     	      	    	i •      r
                                bioslurping in the well can develop (Figure 7.5).  The vacuum-enhanced
                                bioslurper test continues for 96 hours, during which time system perfor-
                                mance variables (product, groundwater, vapor flow rates; contaminant con-
                                centrations; product depth; etc.) are monitored.
                                   As part of the bioslurper test, a pneumatic pump test should also be con-
                                ducted.  Here, soil gas probe vacuum monitoring is carried out from through-
                                out the site at a high frequency during the first 20 minutes of bioslurper op-
                                eration, then less frequently after that point to monitor vacuum development
                                produced by the bioslurper well.  As indicated in Section 7.3.9, reduction of
                                this pressure versus time data allows the estimation of gas permeability at
                                each of the sampling locations where data are collected. Soil gas oxygen
                                and carbon dioxide concentrations and total or specific contaminant concen-
                               i* niii   n, inji',:        i       , f     ,1,1  v, ,i< III.M   ,,i ,	jii'n"1  , !•„ • J.
                                trations are also monitored at soil gas probe locations to evaluate the radius
                                of influence of the bioslurping well in terms of oxygen transfer, and to aid in
                                the evaluation of the distribution of biological activity taking place through-
                                out the site.

                                7.2.17.3 Secondary Skimming Test
                                  Following the 96-hour bioslurping test, a second skimming test is carried
                                out to provide additional data regarding sustained product recovery using
                                conventional techniques versus vacuum-assisted bioslurping. This test is run
                                for a 24-hour period with the wellhead valve open to maintain atmospheric
                                                                 7.34
-------
                                                           Chapter 7
pressure in the bioslurper well and the slurp tube maintained at the static
product/water interface.                  ,
                                       . i.
7.2,17.4. Dual-Pump/Drawdown Test
   This final product recovery test is designed to assess the effect of product
drawdown on its rate of recovery in the bioslurping well.  Groundwater and
product drawdown are produced by lowering the slurp tube to below the
static water table level. The depth below the; water table to which the slurp
tube is lowered is recommended to be equivalent to the vacuum produced at
the wellhead during the bioslurping portion of the field treatability test.  In
case of extremely high or low vacuum readings observed at a site, default
values of a maximum drawdown of 3 ft and a minimum drawdown of 1 ft are
recommended by AFCEE (1995).  This dual-pump test (Figure 7.6) is car-
ried out for a 24-hour period during which time the following process vari-
ables are monitored: product, groundwater, and vapor flow; groundwater
and product elevations; and contaminant concentrations.

7.2.17.5  In Situ Respiration Tests      I
   In situ respiration testing  should take place following oxygenation of the
unsaturated zone below the site in the bioslurper test portion of the field
treatability study. AFCEE (1995) recommend that the test be initiated con-
currently with the second skimming test usiiig procedures described above in
Section 7.2.7.3 to evaluate the contribution to overall contaminant recovery
at a site that can be attributed to aerobic biological degradation.

7.2.18 Performance Evaluation

7.2.18.1  Operations Practices
   Operation of a bioslurper  system is mechanically simple, and requires
only adjustments of vacuum  produced by the; liquid ring pump, and of
bioslurper tube depth during each phase of field treatability testing.  In addi-
tion, groundwater and product elevations; product, groundwater, and vapor
flowrates; and contaminant composition data must be collected over time.
Once an optimal operating mode is selected at a given site based on field
treatability test results, operating requirements primarily consist of checking
and adjusting operating vacuums and system flow rates, managing product
storage requirements, checking and adjusting effluent treatment systems, and
collecting flow stream data for performance evaluation and regulatory com-
pliance monitoring.
                                7.35
-------
 Integrated Technologies
7.2.18.2 Operations Monitoring
   As indicated in Section 7.2.2.6, operations monitoring requirements for
bioslurping systems are similar to those of bioventing systems, with the addi-
tional monitoring necessary for trie product/liquid stream generated in the
bioslurping process.  Liquid and vapor contaminant recovery rates and contami-
nant in situ biodegradation rates are important process performance variables
that determine which operating mode is best suited for optimizing remediation
of a given site. Quantification of the contaminant removal rates in each of these
pathways is crucial in selecting an operating mode for long-term use at a site.
Likewise, once a specific operating mode is selected, the long-term perfor-
mance of the bioslurping system must be based on data collected to describe a
mass balance for product recovery by a bioslurping system.
• "    \ '" ,    |       '           ': „ : :, 	"  ">, <• I11'"'' ""   I1'  "i  :i '   ' '"  " ,' ,'    '    i' '  !i • ,P  |i'i;
   Long-term operation monitoring must include:
        '	'	  "., i	::  	,,,;:• ;,  , ••;;,. ;...-•,  »:!. I',.'- ,:. •' '  .' ,   ,   '•  •    i ,;«  •;
         •  free product, groundwater, and vapor phase recovery rates;
         •  free product, groundwater, and vapor phase contaminant concen-
           tration data;
         •  free product and groundwater elevation data from throughout the
           site;  and
         •  contaminant removal provided by in situ biodegradation esti-
           mated ideally by both in situ respiration measurements and oxy-
           gen and carbon dioxide concentration measurements in the ex-
           tracted vapor stream.
     '.' !'         . .      '..•:,:'•  ;''  " ,).' I.',?	•'  '  I"1 1: ,:	 -.	 '   :  ' ! ,,T	
   Collection of these data over time allows an ongoing assessment of the
effectiveness of a bioslurper system for the recovery of product, extraction of
dissolved contaminant mass, and  destruction of residual contaminant mass in
the unsaturated zone. These data also enable a periodic modification of field
operating procedures to provide continuous process improvement as site and
contaminant conditions change.
                                               i'
7.2.18.3 Quality Assurance/Quality Control
   A Quality Assurance/Quality Control (QA/QC) plan should contain the
practices to be used during bioslurper operation to ensure the accuracy, pre-
cision, completeness, representativeness, and comparability of all collected
data as described inTest Methods for Evaluating Solid Waste, SW846 (US!
EPA 1986c, 1986d) and the Interim Guidelines and Specifications for Pre-
paring Quality Assurance Project Plans (US EPA  1980). The reader is re-
ferred to a general outline for such a quality assurance plan presented in
Table 5.15. QA/QC  practices applicable to the bioslurping process include
                                  7.36
-------
                                                            Chapter 7
practices common to other bioremediation technologies (i.e., the use of
blanks, blind duplicates, and spiked samples for laboratory and field mea-
surements).  In addition, engineering practices to ensure data quality and QC
should include:
                                       1
        • an ongoing review of health and safety practices;
        • review of sampling and analysis procedures;
        • ongoing training of new field personnel;
        • routine maintenance of all field process equipment; and
        • routine maintenance and calibration of all field monitoring and
          sampling equipment.
                                7.37
-------
••£:'
                                                                                                                                                                J!S:	][      Illliil'1"!,  ' :	i  ',     ,1   •  ';"'	'«	1  'Ml   i  •• ,'	illifin.il
-------
                                                         Chapter 8
                    CASE  STUDIES
 8.1  Biofliter Vapor Treatment

   The development, pilot testing, and implementation of a soil bioreactor
 for the treatment of vapors from the S.C. Johnson Wax Facility in Racine,
 Wisconsin are described. This project was conducted jointly by the US EPA
 Laboratory in Ada, Oklahoma, (Robert S. Kerr Laboratory) and S.C.
 Johnson & Sons, Inc. of Racine, Wisconsin. S.C. Johnson Wax (Johnson
 Wax) implemented an in-ground soil reactor based on the results of US EPA
 and their own testing. The soil reactor was, initially, successful in treating
 the vapors, but then failed in cold weather and was eventually replaced with
 a more traditional gas treatment method.

 8.1.1  Site Description
   Johnson Wax operates a manufacturing facility in Racine, Wisconsin,
 where a variety of consumer products are produced, some of which are pack-
 aged in aerosol cans.  During the filling process, some of the propellant is
released. The propellant consists of a mixture of light hydrocarbons, prima-
rily propane, n-butane, and isobutane. In  1986, the filling losses were col-
lected and vented directly to the atmosphere, the common industrial practice
at that time. Although a reasonably available control technology (RACT)
standard had not been developed as of 1986, Johnson Wax anticipated that
the state of Wisconsin would soon develop RACT regulations that would
require an 85% reduction in industrial emissions, such as those from their
aerosol can filling process.
                               8.1
-------
   Case Studies
   8.1,2  Technology Selection
      It had long been known that indigenous soil microorganisms could de-
   grade a wide range of organic compounds, including the hydrocarbons
   present in the propellant.  Studies in the Netherlands had shown that as much
   as 5 mg of methane/kg of soil/hour' could be degraded in a soil biofilter after
   a suitable acclimation period. Based""on this and other experience, largely in
   Europe, Johnson Wax and the US EPA researchers decided that biological
   vapor treatment using soil as the filter-bed matrix was a viable option for
   treating the propellant-laden gases.

   8.1.3  Technology Evaluation
      Laboratory tests were conducted at the Robert S. Kerr Laboratory in Ada,
   Oklahoma, to evaluate biofilter biodegradation potential.  These tests were
   conducted in serum microcosms using site soil maintained at 30% moisture
   content and 22 to 24°C. Propellant gases consisting of 29 mol% isobutane,
   44 mpl% n-butane, and 27 mol% propane were used in these laboratory
   studies. Further laboratory tests were conducted by the Ada laboratory to
   evaluate biofilter acclimation and to generate laboratory degradation kinetic
   parameters.  Propellant gas mixtures were introduced into
   approximately 60 times greater volume of air than the air-filled pore spaces
   of the prototype biofilter.  Aliquots of gas were periodically removed and
   analyzed by gas chromatography with a detection limit of approximately 20
   to 50 ppmv.  Tests were conducted in duplicate or triplicate, and data were
   analyzed for fit by regression analysis.
      Evaluation of the laboratory test results led to a prediction of between
   90% and 95% removal efficiency in the bioreactor at a soil temperature of
   24°C. Figure 8.1 presents the hydrocarbon removal rates achieved in the
   prototype reactor which corresponded well with the laboratory test results.
   Acclimation in the laboratory tests was slow, with little increase in degrada-
   tion rates over several weeks. Johnson Wax experienced acclimation times
   of approximately 20 days in the field unit.  Following acclimation, degrada-
   tion rates increased by approximately a factor of 100 over the initial rates as
   observed in the biofilter.
      As shown in Figjure 8.2, disappearance of the three hydrocarbons was
   rapid and extensive. Hydrocarbon disappearance could have been due to
   Variety of mechanisms: sorption to soil components; abiotic transformations;
   or physical loses through adsorption to glass or leakage from the serum mi-
'   crocosms. However, leakage from the microcosms was not observed.
                                    8.2
-------
                                                              Chapter 8
                                Figured.]
         Hydrocarbon Removal/Degradation in Prototype Reactor


g 90
I 92

S3 94
W
I 96
1 98



o
o
~ o _
o

— •

• «
- , v / „ 9l9

u
25 ~
20 g
15 I
fi
10 =§
CO




9/1 9/21 10/H 10/31 :
Date
• Percent of Injected Hydrocarbon Degraded
O Soil Temperature at Depth of Injection
Source: Kampbelletal. 1987. Used with permission.
   Preliminary laboratory investigations indicated that biodegradation might
be able to satisfy regulatory requirements for the reduction in volatile or-
ganic compound (VOC) emissions from aerosol can filling operations. To
verify the preliminary laboratory VOC destruction efficiencies and deter-  .
mine cold weather effects, Johnson Wax planned the construction of a pro-
totype soil bioreactor. After discussions with the Wisconsin Department of
Natural Resources (WDNR) to ensure acceptability of the data obtained
from the pilot operation, the bed was constructed. Initial startup occurred in
May 1984.
   At the request of WDNR, the prototype soil bioreactor (Figure 8.3) was
lined with an 0.8-mm reinforced  poly vinyl chloride (PVC) liner to prevent
groundwater contamination. The air pipes (10-cm [4-in.]) Schedule-40 PVC
pipe perforated with 3-mm holes) were installed above the liner and buried
in Rollin muck soil. This soil contained 6.5% organic matter and 45% clay
and had a pH of approximately 7.4. The surface area of the bed was 190 m2
(2,040 ft2), and the soil depth was 90 cm (3 ft).  The bottom of the bed was
                                  8.3
-------
                                            Case Studies
                                                                                 Figure 8.2
                                              Removal/Degradation of Hydrocarbons In Initial Laboratory Experiments
	Ill	1,	
                 V'
,4111
                                                             ISO Butane Initial Concentration 441-510 ppm (volume/volume)

                                                             m r-t-	———i-	—	Q
                       eo   80


                       1   60


                       |   40

                       (2   20

                            0
                                                                         1   1   1   1  0  i  I   1   1   1   1   1   1 - 1 - 1
      0     2     46    8     10   12    14    16
                            	  I
                          Time(hr)


  n-Butane Initial Concentration 399-470 ppm (volume/volume)

100


 80


 60

 40

 20
                                                                                                                   I
                                                                                     i
                                                                                                               1
                                                                   0     2    4     6     8     10    12    14    16

                                                                                  ..  .;,, Time(hr)


                                                               Propane Initial Concentration 248-285 ppm (volume/volume)

eo
1
3
£

100
80
60
40
20
n
-* T
^
: s
•
: i
i i l i T l i i i i l l l l 1 1 	 J-.
                                                                   0     2     4     6     8     10    12    14    16

                                                                                       Time (hr)
                                            • Soil from a Depth of 90 cm/Acclimated
                                            O Control Soil not Previously Acclimated

                                            Source: Kampball et al. 1987. Used with permission.
                                                                                     8.4
-------
                                                              Chapters


 sloped 0.75% to carry leachate to a sump.  The injected air contained approxi-
 mately 2,000 ppm (v/v) total hydrocarbons. The intended flow rate was 3.0 cm3
 air/cm2 of surface area/minute, making the residence time nearly 15 minutes.
 The pressure drop across the bed was approximately 85 cm (33 in.) of water.
                                Figure 8.3
             Schematic of Prototype Vapor Treatment System
                                        ^tj'r Injection Pipes
     Leachate Collection Pipe
                                            Balancing Valve
Source: Kampbell et al. 1937. Used with permission.
   To monitor bed performance, offgases were collected in a modified
55-gallon drum. The drum was open at the bottom, had opposing baffles at
the rolling hoops, and had a small vent hole in the top.  It was positioned on
the soil bed at random, and allowed to come to equilibrium before sampling.
Offgases in the bottom third of the barrel were withdrawn directly into a
portable organic vapor analyzer (OVA) (Foxboro Century OVA Model 128)
with a flame ionization detector (FID). The influent gas was sampled
through a valve on the distribution pipe. The soil temperature was moni-
tored with a thermocouple buried 75 cm (30 in.) in the soil bed.
                                  8.5
-------
       Case Studies
'! ' II •'
 ,1311,1
          It was anticipated that prior exposure of soils to hydrocarbon vapors
       would elevate the microbial density, particularly of those microorganisms
       able to utilize the hydrocarbons being tested. Based on a comparison with
       control reactors, it was concluded that biodegradation was occurring and was
       the primary mechanism for disappearance of the tliree hydrocarbons.
                               •> , .*!,"'. :,"•. y;K. V-i'1	'•••	'. ! tf':1!1 i:-.:	d  „•;''':j  " '• /' • '''»'
          No lag periods were observed, and the disappearance of hydrocarbons
       was proportional to their inlet concentrations, although at higher concentra-
       tions it appeared that the biodegradation rate was limited by the microbial
       capacity to metabolize the organic compounds. For the most part, degrada-
       tion followed first-order degradation rates described by:
                                                                         (8.1)
       where:
               K   =
                S  =
                 the biodegradation rate;
                 the maximum possible biodegradation rate;
                 an empirical constant representing the concentration
                 where biodegradation shifts from first-order to
                 zero-order kinetics (commonly called the
                 half-saturation constant); and
                 trie concentration of the organic compound in air.
   The removal rate of hydrocarbons was examined at 700,6,000, and
20,000 ppm (v/v) to cover the transition from first- to zero-order kinetics.
Figure 8.4 is a graphical representation of the prototype bioreactor test re-
sults. This plot indicates that the hydrocarbon degradation followed hyper-
bolic kinetics. The slope determines the half-saturation constant, Km, (5,680
pprriv).  The intercept of the y-axisindicates the maximum biodegradation
rate for the system, which was 6.8 mg propellant/kg of soil/hour. Hydrocar-
bon degradation rates should be first-order for vapor concentrations of less
than 500 ppmv.
   The biodegradation rates for the three hydrocarbons were similar, varying
only by a factor of 3.  As might be expected, the branched hydrocarbon,
isobutane, was the slowest to degrade m the pilot biofilter.
   It was anticipated that degradation would be fastest within the deepest
portion of the prototype soil bed. However, the opposite was occurred. The
primary reason for this apparent anomaly; may have been that the higher
moisture content within this interval restricted movement of hydrocarbon
vapors through these soils.  The moisture content was elevated in this inter-
val because of the liner that prevented adequate drainage from the biofilter.
                                         8.6
                                                                     	I	
-------
                                                              Chapter 8
                               Figure 8.4
     Hydrocarbon Removal/Degradation Kinetics in Prototype Reactor
   1.4


f  "

1  L0

5  0.8

|  0.6

M
M  0.4

^0.2
                                         Slope =
    km
   Vmax
                                         6.8
mg Hydrocarbon
  kg Soil • hr
                                        5 680 ppm
                            0.0005      0.0010      0.0015
                          1/S (ppm Hydrocarbon in Air1)
Source: Kampbelletal. 1987. Used with permission.
   The probable byproducts of alkane biodegradation are alcohols and ke-
tones.  Both classes of compounds are biodegradable so disappearance of the
alkanes should not be accompanied by the production of stable intermedi-
ates, particularly since the feed gas contained oxygen far in excess of the
amount required for biodegradation of the hydrocarbons.  Consistent with
these observations, no hydrocarbon or intermediate byproducts were ob-
served in the biofilter leachate.
   Concern over performance during the cold winters in Wisconsin led to
tests that evaluated degradation rates at 12 and 24°C. As shown in Figure
8.1, the test results indicated that performance would not be significantly
impacted over the temperature range tested. The effects of temperatures
below 12°C were not investigated, so it was not possible to know the lowest
temperature at which satisfactory performance could be expected.

8.1.4  Implementation
   Subsequent to the laboratory and prototype reactor tests, Johnson Wax
constructed and field tested a full-scale soil Injection bed (SIB). The bed
                                  8.7
-------
Case Studies
was approximately 57 by 49 m (185 by 160 ft). The active treatment zone
consisted of 1 m (3 ft) of soil underlain by slotted PVC pipe that introduced
air into the soil from a manifold that delivered aerosol-laden air from the
aerosol production line gas house. The PVC pipe was contained within 0.3
m (1 ft) of gravel overlaid by 0.7 m (2 ft) of soil. Beneath this soil was a
leachate collection system consisting of slotted sewer drain pipe that trans-
ferred leachate to the process sewer.
   Because the SIB was a nonstandard air pollution control device, demon-
stration of performance was required. Influent air flow velocity and con-
stituent levels were easily measured using standard methods.  Because the
effluent from the SIB discharged over a large surface area at very low flow
rates, compliance demonstration required methods specific to the design of
the system.  Effluent gas was collected using a special hood which was
placed on a portion of the SIB. The hood collected effluent gas from a bed
surface area of approximately 10 m2 (110 ft2). The discharged effluent was
directed through a 5-cm (2-in.) diameter Schedule-40 FVE open-ended dis-
charge pipe. A calibrated Kurts Model 505-9B-00 Mass Flow Meter was
used to measure  the discharge rate which was recorded on a strip chart re-
corder.  Samples of influent and effluent gas were collected and measured on
a gas chromatograph fitted with an FID.
   The hood was moved around the bed to numerous locations where flow
rates and effluent quality were measured.  Statistical methods and correction
loss factors for air flow rate loss within the hood and discharge pipe were
applied to calculate total bed flow rate and percent conversion of the three
hydrocarbons within the active biofilter.
   Analysis of flow revealed a fairly complex pattern. Further, a good deal
of the effluent air was unaccounted for.  Sheet piling was installed on three
sides of the bed along with a clay barrier on the fourth side to minimize
lateral air losses.
   Further investigation using the hood identified several high flow zones.
To correct this problem, the bed was tilled to yield a more uniform air flow.
Regrading of the bed eliminated a swale that had been engineered into the
original design to facilitate runoff control. Even after these modifications,
only 40% of the  air flow was accounted for.
   Additional testing indicated very little leakage to the sides of the bed
where the sheet piling had been placed. However, the side with the clay
barrier indicated air leakage that extended a substantial distance outside the
formal limits of the Sffi. This increased the volume of the soil bed and,
fortunately, the degree of treatment provided to the waste stream gases.
                                  8.8.
                                                                  K.I1 ,1,:, "I ' '	\I
-------
                                                           Chapter 8
   The percent destruction and removal efficiency (DRE) decreased dra-
 matically when the air flow through the filter bed exceeded approxi-
 mately 0.015m3/sec/100m2 (3 cfm/100 ft2). Had the actual bed volume
 not been larger than the design volume, adequate treatment would not
 have been achieved. The measured DRE for the three areas ranged from
 85% to 95%.  However, because not all of the air flow could be ac-
 counted for, even including areas outside the  SID, the official reported
 overall treatment efficiency was 60%; just over half of the reduction was
 attributed to the designed reactor bed volume.
   The regulatory requirement for treatment was 52% DRE so the reported
 system performance was adequate. Had all of the air flow been accounted
 for, the actual DRE probably would have been significantly higher than 60%,
 possibly  close to the 88.5% DRE achieved for that portion of the air flow
 that was  accounted for.
   Although Johnson Wax demonstrated compliance with the required 52%
 DRE, performance problems during cold weather resulted in Johnson Wax's
 decision  to replace the SIB.

 8.1.5 Conclusions
   Johnson Wax achieved 85% to 90% DRE of hydrocarbons contained in an
 aerosol-laden air using the SIB.  They were able to identify air flow losses
 and take  appropriate actions to contain some of the lost air flow through
 their biofilter. Treatment met regulatory requirements during warm weather.
 However, year-round use of the SIB was terminated because of decreased
 performance during winter months. Reduced performance levels during
 winter, combined with the need to report lower performance values than
 actually achieved because it was not possible to account for all of the air
 flow, meant that the reported efficiencies were not satisfactory year round.
   This case history demonstrates that despite pilot tests, which incorpo-
 rated both the filter bed matrix that was used at full-scale and the spe-
 cific gas  compositions to be treated, problems can still arise.  These
 problems can relate to control and measurement of air flow, uniformity
 of the filter bed with respect to air flow velocities, and achievable degra-
 dation capacity at full-scale. The need to iaccount for nearly all of the
effluent volume can be critical when attempting to validate process per-
formance in biofilter systems.
                                8.9
-------
Case Studies
8.2  Biosparging
                                         it I T I1 'i 'lii!111
   This case study outlines design and implementation of an integrated
sparging and vapor extraction system to achieve site closure (numerical stan-
dards) at an industrial facility in the eastern United States.

8.2.1 Site History
   The site is a former manufacturing facility that operated from the 1940s to
1988. No operations have been conducted at the facility since 1988.  A site
plan of the facility is presented in Figure 8.5.
   The background information collected during hydrogeologic investiga-
tions identified nonaqueous-phase liquids (NAPLs) and elevated concentra-
tions (several hundred ppm) of VOCs in the unsaturated and saturated zone
soils . The highest observed concentrations of'VOCs in the subsurface soils
were between 1.5 to 2.5 m (5 to 8 ft) below ground surface; the average
depth to groundwater at the site was 1.25 to 2 m (4 to 6 ft). VOC concentra-
tions in the groundwater up to mg/L were also identified. The primary VOC
detected was toluene.
   The source of the VOCs is believed to be several solvent transfer lines
that ran in an underground trench from the abbveground storage tank located
at the west end of the site. The transfer lines and the aboveground tanks
were ramoVfeH from the site in 1^89. The site investigation data suggested
that approximately 4,500 kg (10,000 ib) of toluene were present in site soils.
   The geologic materials are relatively consistent across the site.  Fine to
medium sands and silts are present from 0 to 0.6 m (0 to 2 ft) below grade
and is underlain by medium to coarse sands and gravel to a depth of at least
5.2m,(17ft).   	      ;	'  i(
   The site investigation results dictated implementation of an interim rem-
edy. Based on the contaminant mass and its volatility and biodegradability,
the approved interim remedy included the implementation of a full-scale in
situ soil vapor extraction (SVE) and air sparging (AS) system.
8.2.2  System Design and Implementation
    1 , . ,'        	     MI •  '   i .'i. ii	 	  • , .in1 h ' . ' i,, "• i1 nM i' '""I ' •»' ' i: ''j«  i '' '  	•:• r ' : ' 	i: 'i. i|i» " /
   A pilot study design was developedI following' a review of pertinent soil
sampling and analysis data. This design called for the installation of two
SVE and AS test well networks,  each consisting of one vapor extraction
well, one air sparging well, four  shallow vapor probes, and two sets of nested
piezometers.  Each test well network was used to evaluate the critical
                                 8.10
-------
                                                            , Chapter 8
                               Figure 85
                                Site Plan
                             (not to scale)
  WHl



!/
*r
i
Former NY, NH and HRR

parameters used in the design of the S VE and AS systems. The test areas
were chosen to provide an evaluation of the design parameters over the vari-
able site conditions that were expected during operation of the full-scale
system. One test area, which was paved, had elevated VOC concentrations
                                8.11
-------
                                                                             1 ''i11;1
-------
                                                               Chapter 8
   Pre- and post-AS test soil gas and helium tracer concentrations were mea-
sured in all local vapor probes for each AS test. Soil vapor discharge
samples were collected from the SVE system periodically and analyzed for
total hydrocarbon concentration using the OVA and field GC.
                                Figure 8.6
                            Remediation Area
                              (not to scale)
                                                        r
      /Site Access Gate	 _^_L_J—,
                                                         Site Access Gate
                                                    Parameter
                                                    Evaluation
                                                    Test Area 1
     Former Aboveground
      Storage Tank Area
                  N
                 A
Parameter
Evaluation
Test Area 2
                                 8.13
-------
                                                                   I	"
                                                                   	III!, «:!(,'.
                                               ll .1.
Case Studies
   The results of the SVE field testing and subsequent air flow modeling
indicated that the physical characteristics of soils (i.e., the permeabilities of
the soil strata and air flow potential) across the site were within the range
considered optimal for the application of SVE technology. Data analyses
indicated the effective radius of vacuum influence  (based on radially- depen-
dent pore volume exchange rates) of a vapor extraction well in the unsatur-
ated zone in unpaved areas to be approximately 3 to 8 m (10 to 26 ft) at an
air flow rate of approximately 8 to 17 m3/hr (5 to 10 scfm). (A pore volume
exchange is defined as a volume of clean air which moves through a unit
volume of contaminated soil.  Pore volume exchange rates are essential to
determine of an SVE well radius of influence for the remediation of VOCs.
Observed vacuum alone does not indicate pore volume exchange rates).
   Discharge VOC concentrations during SVE tests indicated that the highest
VOC levels were observed in the northern area of the site where vadose zone
soils contained VOCs as a residual and/or nonaqueoiis phase.  Lower levels
of VOCs detected in the southern area of the site indicated lower levels of
groundwater and soil contamination.
   Several short-term AS  tests were conducted to evaluate the feasibility of
using air sparging technology to remediate VOC-contaminated saturated
zone soils at the site and to develop a conceptual full-scale design.  The re-
sults of the field testing indicated that the physical characteristics of soils
(i.e., the permeabilities of the soil strata and air flow potential) across the site
were within the range considered optimal for the application of AS technol-
ogy. Data analysis^ computer modeling", arid previous experience at similar
sites indicated an effective radius of influence of 3 to 8 m (10 to 26 ft) at an
air injection flow rate of 7 nVYhr (4 scfm). Due to  the estimated mass of
VOCs present, the design radius was selected to ensure rapid mass removal
via volatilization with system polishing through biodegradation.

8.2.3 Full-Scale Design and Installation

8.2.3.1 Full-Scale Design
   The conceptual, full-scale SVE/AS design was  developed under the as-
sumption that the goal for the site was to reduce the VOC sources contribut-
ing to groundwater contamination to state-mandated standards (maximum
contaminant levels [MCLs]) in 18  to 24 months of system operation.
   The full-scale SVE system design was developed to remove vadose zone
soil VOCs and to capture AS vapors across the delineated remediation area.
The full-scale AS design  was developed to remove VOCs from saturated
                                                                   \- •	
                                      ' ji	! i. '    ,„! - I ••'•   •    '  .  ." •    .1:1 !»	'
                                 8.14
-------
                                                             Chapter 8
 zone soils and groundwater to a depth of 2.5 m (8 ft) below grade across the
 delineated remediation area.
    Injection of air into saturated zone soils displaces groundwater and forms
 a random distribution of air channels in the vicinity of an AS well. At this
 site, plume migration due to water displacement was expected to be mini-
 mal, based on the observed low air entry pressures, minimal groundwater
 mounding during the pilot testing, and the jproposed pulsed-mode 0f system
 operation. However, to control the minimal displacement and natural migra-
 tion of contaminated groundwater, a series of defensive AS wells (a sparging
 curtain) was designed around the eastern, western, and southern portions of
 the remediation area.
    Based on a review of site conditions, pilot test results, and contaminant
 distribution, it was determined that approximately 67 SVE/AS well pairs and
 three single SVE wells were required to remediate the target area. Design
 vapor extraction well (VEW) flow rates were in the range of 8 to 17 nvVhr (5
 to  10 scfm) at 50 to 75 mm (2 to 3 in.) of water vacuum.  Design air sparging
 flow rates per air sparging well (ASW)  were in the range of 5 to 7 m3/hr (3
 to 4 cfm) at 35 to 55 kPa (5 to 8 psi) pressure.

 8.2.3.2  Full-Scale Installation         i
    The 67 SVE/AS well pairs and three single VEWs were installed in
 spring 1993. All borings were completed using standard rotary auger tech-
 niques. Auger cuttings were visually classified for soil type and screened for
 VOCs in the field with a portable OVA. Site soils were visually classified as
 medium to coarse sand and gravel.
   The ASWs  were constructed of steel well points with 0.6 m (2 ft) screens
 attached to steel rises. The installation of the well points consisted of first
 advancing a 160-mm (6.25-in.) inside-diameter hollow stem auger to ap-
 proximately 3  m (10 ft) below grade.  The well points were then driven to a
 depth of approximately 5.2 m (17 ft) below grade using a 140-kg (300-lb)
 drop hammer.  The open borehole was then backfilled with bentonite from
 approximately 3 m (10 ft) below grade to approximately 2 m (6.5 ft) below
 grade.  For this particular site, driving ASWs proved to be cost-effective
 compared to installation by standard hollow-stem auger techniques, but this
 ASW installation technique may not be applicable to all sites. All ASWs
 were finished with 15 to 30 cm (0.5 to 1 ft) of riser above grade to allow
manifold connections to be made at a later date.
  The VEWs were installed by standard hollow-stem auger or
wash-and-drive techniques and were constructed of 50-mm (2-in.)
                                 8.15
-------
i  Si!
        Case Studies
        Schedule-40 PVC well screen with 0.5-mm (0.02-in.) slots and a 50-mm
        (2-in.) Schedule-40 PVC riser. Screen lengths for each well were 1.2 to 1.5
        m (4 tp15 ft) .The borehole wasthen backfilled with silica sand to the top of
        the well screen. A 6.6-m (2-ft) thick bentomte seal was placed above the
        silica sand, The VEWs were finished with 15 to 30 cm (0.5 to 1 ft) of riser
        above grade to allow manifold connections to be made at a later date. The
        SVE and AS well layouts are presented in Figures 8.7 and 8.8, respectively.

          Following installation of the SVE and AS wells, manifold piping and
        equipment required to complete the full-scale SVE/AS system were in-
        stalled.  Approximately 1,500 m (5,000 ft) of 50- and 75-mm (2- and 3-in.)
        Schedule-40 PVC manifold line were installed aboveground from the SVE
        and AS wells to the equipment staging area. All manifold[lines were di-
        rectedinto the 4-m6y4.8-m(l2-ft by 16-ft) equipment building where the
        connections were made to their respective blowers or compressors.
          Three;204-m3/hr (120-scfm) explosion-proof SVlb^
        Each blower was fitted withan air/water separatbFw
        shut-off switch, an m^
        valve. Each of the SVE systems was manifolded to separate sections of SVE
        wells. Systems 1 and 2 were manifolded to extract vapors from interior SVE
        wells; System 3 was manifolded to extract vapors from the perimeter SVE
        wells (refer to Figure 8.7).
          Also installed at the site were one 170-m3/hr (lOp-cfm) and one 85-m3/hr
        (50-cfm) AS compressors. Each AS compressor was fitted with intake and
        exhaust silencers and an adjustable pressure relief valve.  The AS systems
        were configured to direct the air flow from the 170-m3/hr (100-cfm) AS
        compressor to the perimeter AS wells, and to direct the air flow from the
        85-m3/hr (50-cfm) AS compressor to a valve bank where it could be  distrib-
        uted to any one of three groups of interior AS wells. The AS  valves in the
        valve bank were electrically actuated and controlled by a timer to allow au-
        tomatic operation (i.e., cycling of AS well groups). The 170-m3/hr
        (100-cfm) compressor was also connected to a timer to allow it to be auto-
        matically cycled. Refer to Figure 8.8 for AS well groupings.
          Three 170-m3/hr (100-scfm) catalytic oxidation units (CATOX) were
        installed to control the VOC emissions fromittie SVE systemdischarge. The
        CATOX units were installed adjacent to the equipment building in a 4-m by
        6-m (12-ft by 20-ft) fenced:area witn a locking gate". The CATOX units
        discharged treated.air^to tne atoosphere through a l50-mm (6-in.) diameter
        stack at a height approximately 4'rh"(f2 ft) above grade.  The CATOX con-
        trol units were connected to a phone modem that was programmed to notify
        operations personnel if any of the systems shut down. The CATOX control
                                        8.16
-------
                                                               Chapter 8
 units were also tied into the SVE systems and the AS compressors. When
 any of the CATOX units shut down, the associated blower and both AS com-
 pressors were also automatically stopped.
                                Figure 117
                             SVE Well Layout
                              (not to scale)
                                                         Site Access Gate
    Equipment Shed
    and Catox Area
                                              Site Building.
A SVE Well Group 1
O SVE Well Group 2
• SVE Well Group 3 (Perimeter)
                                  8.17
-------
Case Studies
                                   "figure 8.8
                                  AS Well layout
                                  (not to i
                                                                 / Site Access Gate
     Equipment Shed
     and Catox Area
       / Olic m.l.(,aa VJaif
                 N             •
                 A
                                                      Site Building
 •  Perimeter AS Group
 V  AS Well Group 1
 O  AS Well Group 2
 A  AS Wet! Group 3
                                        8.18
-------
                                                           Chapter 8
 8.2.4 System Operation

 8.2.4.1 Year 1
   A properly designed and implemented SVE/AS system provides flexibil-
 ity of operation so that system operators can adapt the S VE/AS systems to
 maximize VOC removal at a given site. At this site, valves were used to
 allow control of the air flow through S VE and AS system manifolding and at
 individual SVE and AS wellheads. Additionally, timers were installed to
 automate cyclical operation of the AS wells. Data obtained during routine
 system site checks allowed system optimization on an ongoing basis.
   In late spring 1993, SVE Systems 1,2, aind 3 were activated and initial VOC
 concentrations prior to treatment by the CATOX units were 1,250,1,100, and
 50 parts per million on a volume basis (pprnv), respectively. During subsequent
 site checks and SVE system optimization, SVE System 1 and 2 VOC discharge
 concentrations exceeded 2,000 ppmv, and SVE System 3 concentrations ex-
 ceeded 100 ppmv. (S VE/AS systems installed in the presence of NAPL nor-
 mally commence with SVE operation only. Once the VOCs in the SVE dis-
 charge and the potential for NAPL migration has been sufficiently reduced, the
 AS system is started. This operation allows efficient offgas treatment selection
 based on the life-cycle of the S VE/AS system.)
   Initial estimates of SVE-only operation at the site were 1  to 2 months.
 During summer 1993, historically low groundwater table elevations were
 recorded at the site. This low water table condition exposed more VOCs
 (i.e., a "smear zone") to the influence of the SVE system. VOC mass re-
 moval is more efficient through SVE than by AS since SVE  will normally
 provide more effective contact between the air and contaminant than AS.
 Therefore, to maximize VOC removal, SVE-only operation was imple-
 mented to take advantage of the low water table.
   After two months of SVE-only operation, the perimeter AS  system was
 activated and operated in a pulsed mode. Pulsing AS systems are more effi-
 cient for a number of reasons. First, the capital  and operation costs are re-
 duced because less total system air flow isi required and smaller compressors
 can be used. Second, as discrete AS channels are formed during steady-state
 operation, continuous injection will generally impact water, primarily in the
 vicinity of these channels. Pulsing mixes the groundwater, thereby enhanc-
 ing air/water contact. Third, through pulsing, spatial changes in channel
formation have been observed (on a site-specific basis). This spatial vari-
 ability in channels also enhances mixing of contaminated groundwater in the
vicinity of the AS wells, thereby enhancing contact of contaminated ground-
water with AS channels over time.

                                8.19
-------
Case Studies
   The duration of the perimeter AS well cycle times was determined from
pilot test data.  Because the perimeter AS wells were designed to prevent
downgradient migration of dissolved-phase VOCs, frequent AS cycling
times (i.e., 2 hours on and 3 hours off) were used.
   The interior AS systems were activated in fall 1993. SVE-oniy VOC
discharge concentrations had dropped to concentrations of less than 1,000
pprriv that permitted the activation of the interior AS systems. To determine
optimal initial AS cycling times for the interior wells, the system was opti-
mized to maximize VOC removal while remaining within the operating con-
straints of SVE offgas treatment (i.e., the CAFOX units).  An increase in
S VE VOC levels due to AS activities in the source area leveled off after
approximately  3 hours of air injection at concentrations exceeding 2,000
ppmv. However, after approximately 1 to 2days of source area sparging
(i.e., operating each AS well group for 3 hours twice a day), the CATOX
units shut down due  to exceedance of catalyst temperatures.  It was deter-
mined mat the  S VEsystem could not sufficiently purge the VOC-laden va-
pors released by the AS system between cycles. This  created a buildup of
VOC levels hi the SVE systems that could not be handled by the CATOX
units. Therefore, AS cycle time was reduced to 2'hours twice daily.
   During AS system operation in the winter 1993/1994, the interior AS
wells had to be deactivated for short periods (i.e., less than 7 days) to prevent
water uptake by the SVE system due to high groundwater table elevations
and groundwater mounding by AS operation. Although groundwater
mounding was negligible for a single AS well, the combined effects of mul-
tiple active AS  wells created a temporary mound in the groundwater. Al-
though the groundwater mounding was temporary (dissipated in 2 to 3
hours), the temporal change in depth to groundwater did cause unplanned
SVE system shutdowns due to water uptake.
   In addition to recording historically low groundwater table elevations at
the site during summer 1993, historically high groundwater table elevations
of 0.6 m (2 ft) below ground surface were recorded during spring 1994. This
abnormally high water table elevation prevented SVE/AS system operation
in March 1994. During this period of high" water table elevation, the state
agency was petitioned to allow operation of the perimeter AS system without
the SVE system. Supporting documentation of estimated VOC discharge to
   ,i ' . i, ' "", , i •% irii ,  IT    , CV i,' •, i •' Ji1 i '  f i ' !• • ini' "	S 'H'liNi' ' 	!!'!! if 'rtJi',, 	'innra,! : ,   	 ,,   ,, ,  	 i
the atmosphere during perimeter AS operation (0.07 kg [0.15 Ib] to 0.23 kg
[0.5 Ib] per day) was presented to and approved by the agency. Based on
this approval, the perimeter AS system was activated without SVE.  Site
groundwater table elevations receded by late spring 1994, after which the
SVE system was reactivated.
                                8.20
-------
                                                            Chapter 8
   Over the first 12 months of SVE/AS system operation, approximately
 5,660 kg (12,450 Ib) of VOCs as toluene were removed, primarily via vola-
 tilization by the SVE/AS system. This mass removal estimate exceeded the
 initial 4,500 kg (10,000 pounds) estimate of toluene present at the site by
 37%. Due to the rapid mass removal by volatilization, the mass removed by
 bioremediation was not accurately quantified; however, oxygen uptake data
 suggested up to 10% additional mass removal through biodegradation.
   To further evaluate the effectiveness of the SVE/AS system, groundwater
 quality data were collected on a quarterly basis. Figure 8.9 illustrates
 isoconcentration lines for total dissolved-phase VOCs measured in site ground-
 water prior to SVE/AS system startup. Figure 8.10 illustrates isoconcentration
 lines for total dissolved-phase VOCs measured in site groundwater in April
 1994, approximately 11 months after SVE/AS system startup. These figures
 illustrate that the impacted groundwater plume was being contained and that the
 magnitude of the dissolved VOC contaminant plume significantly decreased
 during the first year of SVE/AS remediation at the site.

 8.2.4.2 Year 2
   Over the second year of operation, the mass of VOCs removed was increased
 to approximately 10,000 kg (22,000 Ib). As expected, the percent of the mass
 destruction due to biodegradation was increased (based on oxygen uptake
 analysis) as the cleanup progressed. At the end of the operating period, it was
 estimated that 50% of the mass removal was attributable to biodegradation.
   At the end of the 2-year operation period, the VOC concentration in
 the soil gas and SVE  discharge had dropped to below detection limits
 (less than 1 ppmv).

 8.2.5 Site Closure
   A closure program,  including extensive groundwater and soil sampling
 and analyses, was initiated. Fourteen groundwater samples were collected in
 and around the remediation area using existing monitoring wells (MW101,
 102, and 103) and Geoprobe points. The groundwater samples were col-
lected immediately after system shutdown and several months following
shutdown (to allow for rebound).  "With one exception, VOC levels detected
in groundwater were measured at or below MCLs in the treatment area, and
ho significant (above MCLs) rebound was observed.  The exception was one
area under the corner of a building that was not accessed by the SVE/AS
 system.  Subsequent access to that location provided rapid improvement in
the groundwater quality to below MCLs.
                                8,21

          Case Studies
                                           Figure 8.9
                            VOC Plume — May,1993 (Units of
                                         (not to scale)
                , Site Access Gate
mg/L)
                                                                      Site Access Gate
                / *1US "^^ ualc           	street_	.	
 HI!1,  nllMII
't!!   iiii!)1
                                             8.22
-------
                                                      Chapter 8
                        Figure 8.10
          VOC Plume — April, 1994 (Units of mg/L)
                      (not to scale)
                                                 Site Access Gate
Site Access Gate
                          8.23
-------
                                Case Studies
111 PI l
                                   Approximately 60 soil samples were collected for analysis from within
                                the remediation area approximately half each from the vadose zone and half
                                from the saturated/smear zone.  A subsample of the soils was sent for confir-
                                matory laboratory analyses for VOCs. The laboratory results indicated that
                                the levels of VOCs in the soils had been reduced from residual levels to
                                nondetectable levels (less than 166 ppb)in all areas except the limited
                                nonaccessable area under the comer of a building.  This area was subse-
                                quently remediated within 1 to 2 months after the system was expanded to
                                encompass that location.   ,

                                8.2.6  Conclusions
                                ";        ',;"'"  '"'' , ",. '  mil h  , , ," .; u.  •  '   '' '"	' i,,*!1' •'  .  i '! ' f • !•' ,,/r ,' "'" , •", '#,! , .,   ,  , "',i ,. 'i';,;,,;	j nil
                                   A closely-spaced AS/SVE system was successful in  the remediation (to
                                MCLs) of impacted soils and groundwater at a manufacturing facility in the
                                eastern United States based on posttreatment soil and groundwater data col-
                                lected from throughout the site. The AS/SVE system operated, as designed,
                                to remove approximately 10,000 kg (22,000 Ib) of VbCs over a 2-year pe-
                                riod. The problems encountered were mostly associated with the physical
                                plant and with fluctuating water table levels impacting SVE system perfor-
                                mancel The role of biodegradation increased significantly over the duration
                                of the project.
8.3  Bioventing

   During early February 1990, a ruptured pipe at a Burlington Northern
Railroad (BNRR) fueling pumphouse in Alliance, Nebraska, resulted in over
230 m3 (60,000 gal) of #2 diesel fuel being released onto the surrounding
SfijpL Several months later, a subsurface
were contaminated to a depth of over 20 m (65 ft) below ground surface
(bgs) and could impact groundwater which was observed at approximately
22 m (70 ft) bgs. State regulatory agencies requested that BNRR develop
and implement a remedial action plan to treat these 3iesei-contaminated soils
and protect local groundwater This case"stuidy summanzes activities carried
out by Parsons Engineering Science, Inc. (Parsons ES) to evaluate a variety
of remediation technologies, and highlights the soil bioventing system used
to remove volatile BTEX compounds and provide long-term biodegradation
of all fuel residuals.
                                                                 8.24
-------
                                                            : Chapter 8
   Bioventing pilot testing was used to determine soil properties, such as air
permeability, and assess the potential for both volatilization and long-term
biodegradation of diesel fuel residuals at the site. Pilot-test results con-
firmed that bioventing was feasible for the remediation of this site* so a
full-scale bioventing system was installed in September 1991. This system
operated continuously for over four years. System performance and a review
of site characterization test results is summarized below.

8.3.1  Site Description
   The site is located south of a diesel fuel pumphouse, where the pipe rup-
ture occurred (Figure 8.11).  An estimated 13,000 m3 (15,000 yd3) of soil
was contaminated to a depth of approximately 22 m (70 ft) bgs. In prepara-
tion for the pilot test, a single  10-cm (4-in.) diameter vent well (VW) and six
5-cm (2-in.) diameter vapor monitoring poiints (VMPs) were installed. The
results of the pilot test and a more detailed description of the methods used
to conduct the test were reported by Parsons Engineering Science (1991).
                                       I
8.3.1.1  Groundwater Conditions
   Groundwater was observed at depths of 21.5 to 22 m (68 to 70 ft) bgs.
Dilute levels of BTEX compounds were detected in groundwater beneath the
site; however, only benzene  and total petroleum hydrocarbon (TPH) concen-
trations exceeded their clean-up levels of 5 |Hg/L and 2 mg/L, respectively.
Due to the large vadose zone above the water table, the vast majority of the
diesel fuel spill was  adsorbed and occluded! in these unsaturated soils. With-
out soil treatment, soluble BTEX compounds could continue to percolate
downward toward the groundwater, creating a larger and more concentrated
plume of hydrocarbon contamination than existed at the time.

8.3.1.2 Soil Conditions
   Soils at this site were characterized during the construction of the VMPs
and the VW in April 1991. The general lithology in this area, shown in Fig-
ure 8.12, consists of fine- to medium-grained silty sands from the ground
surface to approximately 9.5 to 11 m (30 to 35 ft) bgs (upper sand zone),
interbedded sand and silt/clay  lenses that extend from 11 to 16 m (35 to 50
ft) bgs (interbedded zone), and another layer of fine- to medium-grained silty
sand that extends to  a depth  of 22 to 24 m (70 to 75 ft) bgs (intermediate
sand zone).  Soil moisture varied from 2% in the intermediate sand to 11%
in the interbedded silts and clays.
                                 8.25
-------
                                      Case Studies
                                                                                  |i!	.,':
111
:ili |,  "'I!-
                                       1
                                        a

                                       I
                                        0
                                       CO





                                       I
                                       a
                                     «
                                          -2
                                       Q
                                        D>




                                       !
                                       JD
                                        D
           £

            I
           c


           al


           £
           CO
                                                                              I
                                                                              §>*
                                                                              •C-n
                                                                                                              1
                                                                                                                <
                                                                                                              .§*
                                                                                                                        ra

                                                                                                                        •i
                                                                                                                       £

                                                                                                                       •5
                                                                                                                       1


                                                                                                                       I
                                                                            8.26
-------
                                                                    Chapter 8
                                  Figure 8.1:2
          Geologic Section — Full-Scale Bioventing Demonstration
                   Pumphouse Spill Site, Alliance, Nebraska
                                 (not to scale)
  4
  en
                                      (Vapor Monitoring Point)
        BNW2 (Monitoring Well)                VMP3  VMP4        B6 (Soil Boring)
                •           P             A1    A              B
       North            VW1 (Vent Well)       .  j                        :  South
     3,960 -
     3,950 -
     3,940 -
     3,930 -
     3,920 -
- 3,960
                                                                        - 3,950
     3,910
     Clay
     Silt
 PF1 Clayey Sand
 EFI Sitty Sand
38,300 Initial SoilTRPH Concentration (mg/kg)(Samples oplleoted in April 1991)
 	Groundwater Elevation
 	Geologic Contact, Dashed where Inferred
  £3 Screened Interval and Total Depth of Well
                ,
Reproduced courtesy of Parsons Engineering Science (1991)
   Soil gas permeability was quantified through vacuum response tests con-
ducted as part of the pilot test in April 1991.  Vacuum response and oxygen
concentrations were measured at the VMPs and nearby groundwater moni-
toring wells (Figure 8.12) while soil gas was being extracted from the VW
                                     8.27
-------
                                  Case Studies
III  ill,   .  ,  	f     I	.'[-.!:   :     1;    ,'|l  „       ~":    :
                                  using a 10-hp vacuum blower. The soil responded rapidly to the vapor ex-
                                  traction system, with measurable vacuum response and increases in oxygen
                                  concentrations occurring in all soil zones, including the interbedded silt/clay
                                  zone. Using test methods described by Hihchee et al (1992), the average
                                  soil gas permeability was estimated at 5 Darcys.  Because the contaminated
                                  soil zone was entirely oxygenated using only the pilot-scale VW, no addi-
                                  tional VWs were required for the full-scale system.

                                  8.3.2  Initial Soil Contamination and Nutrient Availability
                                    The diesel fuel contamination in the soil appeared to be localized within a
                                  19-m (60-ft) radius of the pumphouse (Figure 8.11J.  Initial total recoverable
                                  petroleum hydrocarbon (TRPH) concentrations are plotted on the geologic
                                  section shown in Figure 8.12. Based on initial soil analyses and observa-
                                  tions made during drilling, it appeared that diesel fuel migrated rapidly
                                  downward at the spill site until it encountered the interbedded sand and silt/
                                  clay zone at approximately 9.5 to 11 m"(30~ to 35 ft).  At this depth, the fuel
                                  spreadlaterally and continued its downward movement through more perme-
                                  able sand lenses hi the interbedded layer.
                                    • '  I"'!; :. "r '	<•     "' '"	<  ' '• •'  	t .. , .',*< ,,.. ',	•„ 1 I. 'Ill • •  .1	 .1   '     ••  ..! 	•
                                    Initial soil samples collected from the screened intervals of VMP4 and
                                  VMPS were analyzed for ammonia- and nitrate-nitrogen, total Kjeldahl ni-
                                  trogen (TKN), and water-soluble phosphates. These analyses were per-
                                  formed to determine the concentrations of naturally-occurring nutrients
                                  available in the soils. Ammonia-nitrogen was found at concentrations of 204
                                  mg/kg in the upper sand zone (VMP4) and 4.2 mg/kg in the intermediate
                                  sand zone (VMP5), while nitrate-nitrogen levels ranged from 4 mg/kg in the
                                  upper sands to 11 mg/kg in the intermediate sand zone. TKN levels were
                                  found to be 4 mg/kg in both sand zones. Water-soluble phosphate concentra-
                                  tions ranged from ill mg/kg in the upper sands to nearly 6,000 mg/kg in the
                                  intermediate sands. The relatively low nitrogen concentrations found in
                                  these soils may have been limiting biodegradation rates. To determine if
                                  nitrogen addition could improve natural biodegradation rates, an
                                  ammonium-nitrate solution was added to several VMPs after several months
                                  of bioventing. However, the addition of the nutrient solution did not signifi-
                                  cantly accelerate fuel consumption rates.
                                       1 ,  iM,,      '  	: i ,  """    ' '! ii/.1, i, T ' 	 I	 I,,,'1',,  , , 'VJI "„',  ,    	  	 "ii ,i " li, ,,*!'
                                                        ,	   „ „	„ ii	I,, ;,:	    i	|
                                       • I.  • ' „    •    . •   I '- ,       ,  ,	;i,, •       ( I'  >  .    >        . '   n'TSi
                                  8.3.3 Remedial Goals
                                    The objective of .this remediation project was to reduce the potential im-
                                  pact of soluble diesel components on local groundwater by removing BTEX
                                  from the soil through a combination of short-term S VE and continuous
                                                                  8.28
-------
                                                           Chapter 8
bioventing. A secondary objective was to remediate this large site without
disruption to existing rail lines or the fuel pumphouse where the spill oc-
curred and to obtain regulatory closure for the lowest possible cost.,

8.3.4  Full-Scale System Design and Operation
   Based on the air permeability and oxygeri influence observed during the
pilot test, an extraction rate of 3.2 mVmin (100 scfm) was selected for
full-scale bioventing operations. A 7-hp regenerative blower system capable
of producing this design flow rate was installed at the site, plumbed to the
existing VW;  operation began in September 1991. Soil gas extraction was
selected over  air injection due to the immediate need to remove volatile and
soluble BTEX compounds from the soil before further groundwater contami-
nation occurred.  In December 1994, the system was switched to injection
mode, and the flow rate was decreased to 2 m3/min (60 scfm).  This modifi-
cation was made because much of the soluble BTEX had already been re-
moved/biodegraded, and because bioventing systems require much less
maintenance when operated in an air-injection mode.  The six VMPs in-
stalled during the pilot test were used to monitor pressure response and en-
sure that aerobic conditions were maintained throughout the contaminated
soil volume.
   The full-scale bioventing system operated continuously for over 4 years
with minimum maintenance downtime. The monthly electrical cost for the
system was estimated at approximately $280. BNRR personnel were re-
sponsible for  weekly system checks and monthly blower filter changes.

8.3.5  Long-term Performance Monitoring
   The fuel-consuming capability of native sioil bacteria was examined dur-
ing seven in situ respiration tests conducted by Parsons ES over the four-year
operation period. Emissions of volatile hydrocarbons to  the atmosphere
were monitored to ensure regulatory compliance and estimate the amount of
hydrocarbons physically removed from the soil. Three soil sampling events,
including the  initial sampling event in April 1991, were conducted to docu-
ment the removal/biodegradation of petroleum hydrocarbons at the site.

8.3.5.1  In Situ Respiration Testing
           •
   Continuous air injection into contaminated soil zones  provides the neces-
sary oxygen for aerobic biodegradation. When the blower is turned off,
oxygen is no longer delivered and soil bacteria consume  the available
oxygen. Oxygen consumption and carbon dioxide production were moni-

                                 8.29
-------
                                                                 ••• li1'-!"!
 Case Studies
 tored at each VMP using a portable O2/CO2 gas analyzer. The rate at which
 soil bacteria consumed oxygen was an important indicator of the viability of
 the fuel-degrading organisms in soils near each VMP. Uncontaminated
 background VMPs were also monitored during the initial respiration test.
 Oxygen levels at these background VMPs remained relatively constant at
 approximately 18% by volume, indicating that biological oxygen consump-
 tion of natural (nonfuel) organic carbon and abiotic oxygen consumption
 were not significant in these uncontarninated soils.
         ..   : •  .       .•:	    :: :  , ,""   , •	': •  ':,:.(.,: •, , ;,   ,;, , ,:„  	  ,." ;: : "T  .
   The results of respiration tests at the VW and VMPs 2 and 3 are shown in
 Figure 8.13 and indicate that the apparent rates of oxygen utilization de-
 creased significantly over time. Very low fates of respiration (<0.001% by
 volume oxygen/min) were measured in VMP3 during the last three tests,
 indicating that little fuel remained for biodegradation near this monitoring
 point Soil sampling at VMP3 in September 1995 confirmed that TRPH
 levels were reduced from 194 mg/kg to less than 3.3 mg/kg over four years
 of system operation. Respiration rates also decreased significantly at VW1
 and VMP2.  Although TRPH concentrations were still somewhat high at
 fuel-impacted depths at these locations, respiration rates were minimal.  Low
 respiration rates at VW1 and VMP2 indicate that the majority of the
 readily-biodegradable fraction of the diesel fuel had been eliminated. The
 remaining TRPH in site soils is likely high-molecular weight material, with
 corresponding low water solubilities,  tow  bioavailability, and low biodegra-
 dation rates.
   Using a conservative ratio of 3.5 kg of  oxygen per 1 kg of hydrocarbon
 consumed, the fuel biodegradation rates were estimated for soils immedi-
 ately surrounding VW1 and each VMP. Using this estimation method, the
 average biodegradation rate at the center of the spill (near the VW) was ap-
 proximately 3,400 mg of TRPH degraded/kg of soil/year during the first: two
 years of system operation (September 1991 to November 1993). During the
 final respiration  testing event in September 1995, the average fuel hydrocar-
 bon consumption rate was estimated to be 130 mg of TRPH degraded/kg of
 soil/year.  This represents an order of magnitude decrease in the rate of fuel
 consumption and indicates that the bioventing system successfully accom-
 plished its objective.
8.3.5.2 Volatilization
          1 ' •  '•	 ••• •    :    "•••;   :'"',-"'   " ''1	:'      :	 :  :   : •;'"
   The removal of volatile and water-soluble compounds, such as BTEX,
from this large diesel fuel spill was also a key objective of this remediation
project. Regular sampling of extracted soil gas indicated that approximately
0.3 kg (0.66 Ib) of BTEX and 14.7 kg (32 Ib) of diesel vapors were removed
                                 8.30
                                      ;}'	:•:;'	:; U'i	I"	•:.	:::
-------
                                                               Chapter 8
                               Figure 8.13
           Biological Oxygen Uptake During Respiration Tests —
              Full-Scale Bioventing Demonstration Pumphouse
                       Spill Site, Alliance, Nebraska
  •p 8
  oO u
  w
U.U1DU
0.0140
0.0120
0.0100

0.0080
0.0060

0.0040
0.0020
nnnnn

-
-
-
2
_
_

-
i
i
t^t
i
3








•™
-

_
—

-



_2_



1
5
1
1

%
^
•>X
1




4
3
Is i
                       VWl            VMP2
                       17,600            1,7:!0

                                  Monitoring Point
                            Initial TRPH Concentration (mg/kg)
VMP3
 194
ES3 Respiration Test 1 (Nov 91)
EigSl Respiration Test 2 (Dec 91)
CT3 Respiration Test 3 (April 92)
I-"*' Respiration Test 4 (April 93)
BUB Respiration Test 5 (Nov 93)
CZ3 Respiration Test 6 (Dec 94)
•• Respiration Test 7 (Sept 95)

Reproduced courtesy of Parsons Engineering Science (1991)
per day of bioventing operation while the system operated in extraction
mode (September 1991 through December 1994). During this time, an esti-
mated 21,400 kg (47,000 Ib) of total volatile hydrocarbons were removed via
extraction by the bioventing system. This mass of hydrocarbon removal
could account for approximately 10,000 mg/kg reduction in soil TRPH.

8.3.5.3 Soil Sampling
   To more accurately assess remediation progress, soil sampling programs
were conducted in November 1993 and September 1995.  In each sampling
event, soil samples were collected from a number of boreholes located
within a 6.5-m (20-ft) radius of VWl.  This area was selected for sampling
                                  8.31
-------
                                                  	':'!!	"'"ir I1. "flLIII!
I iillll	   ""!i,«l! ' ,!! "I"
                                  Case Studies
                                                                               ii
                                  because it represented the most contaminated portion of the site based on
                                  initial soil sampling.  Sampling depths matched those of the initial site char-
                                  acterization in April 1991. A total of 18 samples were collected from four
                                  boreholes in September 1993, and 19 samples were collected from five bore-
                                  holes in September 1995. All soil samples were analyzed for TRPH by US
                                  EPA Method 418.1, and six soil samples from each sampling event were also
                                  analyzed for BTEX by Method 8020 (US EPA 1986d). The samples ana-
                                  lyzed for BTEX compounds were chosen based on elevated field hydrocar-
                                  bon analyzer readings.
                                     Figure 8.14 illustrates the general reduction in diesel fuel concentrations
                                  from initial April 1991 levels. These soil sampling results indicate that an
                                  overall TRPH reduction of 75% occurred over the four years of bioventing.
                                 	 • ,,-y   .1 ••,,,•   ,   .  , :	    •.  ,,|- nurt'iii1; oau, I'jsf.:1 "tut	HAHIM, :'*'	<•  ,	i	• i  ,;. -A >,*i-	, >.	••
                                  Only the 11-m (35-ft) depth interval showed a significant increase in fuel
                                  concentrations from the 2-year to 4-year sampling event, likely caused by
                                  the nonuniform distribution of hydrocarbon contamination in the interbedded
                                  zone at this depth. Concentrations of fuel contamination are likely to be
                                  greater in clay lenses in the zone because of higher capillary retention of
                                  liquids, and because oxygen is more difficult to deliver in these soils.  The
                                  soil sample collected during the September 1995 event could easily have
                                  contained a higher percentage of fihe-gramed sofls'lh'^ ttie"sample collected
                                  during the November 1993 event, thereby yielding a higher fuel concentra-
                                  tion.
                                   	           .,  ,..	,,	!.,	.,.,..  .,;	  ,	:	:
                                     The overall TRPJH reduction observed during the first four years of sys-
                                  tem operation was approximately 75 %. Fifty-five percent of the TRPH was
                                  removed in the first two years of system operation, with an additional 20%
                                  removal between November 1993 and September 1995. These results indi-
                                  cate that the rate of TRPH removal is slowing and the benefits of continued
                                 -system operation are decreasing over time.
                                  • ''  ',"  1 r  " i»  ", > '»'" ,, r' ••!	mint   , ",;.ir n, •• •  , • y	,  i,,,'V  i;!,^, i.vMHH :„,•' ?'  »l	;ll	Hi* Ml I1!	i"!1	i " , i 
-------
                                                                ChqpterB
                                Figure 8.14
              Four-Year Petroleum Hydrocarbon Reduction —
              Full-Scale Bioventing Demonstration Pumphouse
                        Spill Site, Alliance, Nebraska
           10.8


           12.2


           13.8


        I  15.2
        .5

        J"  16.8


           18.2


           19.8

B8BBBB8BB88888R 16,900


8888888888888888881
!:!:!:!:::!:!:!:!:!:! 12.100


NB8888888E
23.5
3884,100
.15

BS88B88S5
I;:::::::;:::;::::

388888888
1:1,111

S: 2,300


110,650
•••••i:

,650
),503

38112,850
0

B8W 24,200


,800




X)


,800


I


8,509
j
•:::-::::45,OOC

!,300









50,000


                      10,000    20,000   30,000    40,000    50,000   60,000

                                  TRPH(mg/kg)
All 2-year samples represent an average of two or more sample locations per depth.

•• Initial (April 1991)
S8 2-Year (November 1993)
:•:•: 4-Year (September 1995)

Reproduced courtesy of Parsons Engineering Science (1991)
cause 96% of the risk-driving BTEX compounds were removed/biodegraded,
the health risk at this site has been substantially reduced.


8.3.6 Conclusions

   Full-scale bioventing at this large diesel fuel spill site produced encourag-
ing results during the four years of operation. Remediation took place
throughout a 22-m (70-ft) soil profile with no disruption to railroad opera-
tions or facilities.  Specific indicators of progress include:

        •  a 75% decrease in TRPH in the most contaminated portion of the
           site during the four-year demonstration (based on soil sampling
                                   8.33
-------
Case Studies
           results): Also, a 75% reduction in soil BTEX concentrations
           occurred in the last two years of system operation (between No-
           vember 1993 and September 1995). The low levels of remaining
           BTEX should pose little or no risk to me local groundwater or to
           workers during any potential soil excavation at the site; and
         •  significant decreases in respiration rates across the site. These
           reductions in respiration rates indicate a significant reduction in
           the concentration of readily-biodegradable, low- to
           medium-molecular-weight; petroleum hydrocarbons in the
           fuel-impacted soils.
                                Figure 8.15
     Total BTEX Reduction in the Last Two Years of System Operation •
              Full-Scale Bioventihg Demonstration Pumphouse
                       Spill Site, Alliance, Nebraska
       B


       I
          10.8
          13.8
16.8
          19.8
                             30,000            60.000
                             ./.	'V1,, "I,, sir.,;,,,,",; :•  t,i "	;
                             Total BTEX Concentration (fig/kg)
                                                   90,000
Number of samples Indicated in parentheses.
   2 Year (November 1993)
   4 Year (September 1995)
Reproduced courtesy of Parsons Engineering Science (1991)
                                   8.34
-------
                                                              Chapter 8
                                Table 8.1
             BTEX Concentrations in Soil Samples Taken Within
              6 m (20 ft) of the Vent Well — September/1995
                              Concentrations in. (4g/kg
Sample I.D.
BN-AL-SB8-65
BN-AL-SB8-35
BN-AL-SB7-65
BN-AL-SB7-55
BN-AL-SB7-45
BN-AL-SB7-35
Depth
m (ft)
19.8 (65)
10.8 (35)
19.8 (65)
16.8 (55)
13.8 (45)
10.8 (35)
Benzene
ND
140J
ND
6.3J
2.6J
88J
Toluene
ND
300J
ND
2.8J
ND
270J
Ethylbenzene
ND
550J
2.3J
NO
ND
1,500
Xylenes
ND
27,000
4.8J
ND
ND
15,000
 J-Detected, but value is estimated because it is below the practical quantitation limit.
 ND-Not detected.
 Reproduced courtesy of Parsons Engineering Science (1991)
8.3.7 Cost Summary
   The total cost to date of bioventing at this site including pilot testing,
full-scale installation, and four years of operation and maintenance is ap-
proximately $146,000. Figure 8.16 provides a breakdown of total cost and a
cost per volume of soil based on an estimated contaminated soil volume of
13,000 m3 (15,000 yd3). These totals include: all project administration and
reporting costs, but do not include electrical costs or BNRR labor costs for
system checks.
   Because the majority of the volatile and water-soluble contaminants (i.e.,
BTEX compounds) have been volatilized and biodegraded, and because the
high-molecular weight compounds remaining in the soil are not expected to
cause further deterioration of groundwater quality, it was recommended that
a risk-based closure of vadose zone soils be pursued at this site.  Early in
1996, the state of Nebraska reviewed soil data from the site and agreed that
the site could be closed based on the significant risk reduction that was
achieved using bioventing technology.
                                 8.35
-------
                                     Case Studies
                                                                                 1 t
                          (I.  	ii;!!;!
                                                                    Figure 8.16
                                                Cost Summary — Full-Scale Bioventing Demonstration
                                                      Pumphouse Spill Site, Alliance, Nebraska
                                                                                       Four-Year Monitoring
                                                                                      and Sampling $4.28/yd3
        Four-Year Monitoring
         and Sampling 44%
                                                                               Pilot Testing
                                                                                $3.06/yd3
Pilot Testing
   31%
Full-Scale
Installation
$2.39/yd3
Full-Scale
Installation
  25%
                                                Total Cost id Date - §146,000
                                      •Based oq an estimated 15,000 yd*

                                      Reproduced courtesy of Parsons Engineering Science (1991)
                                      8.4  Raymond Process

                                        This case study describes successful application of enhanced bioremedia-
                                      tion of an aquifer at the French Limited Superfund Site in Crosby, Texas
                                      (Adapted from Dey et al. 1993; O'Hayre et al. 1093; Thomson et al. 1995;
                                      and Biotreatment News 1993,1994).  Specifically, the study of this complex
                                      site illustrates the following:

                                              •  observational approach to achieving an effective remedial system;

                                              •  use of more than one type of electron acceptor;

                                              •  incorporation of an alternate water supply;

                                              •  integration of bioremediation with the use of a barrier and an in
                                                situ slurry reactor;

                                              •  use of intrinsic remediation as a polishing step to achieve a
                                                shorter active treatment period; and

                                              •  planning and budgeting for significant system modifications that
                                                require capital expenditures.

                                        The approach described in this case study reduced the time of
                                      remediation, perhaps by as much as 50%, without having to over-design
                                      the system.
                                                                        8.36
imf^l	i»!l,	Ililiii, ,=	,	i;:.1 < liiiiiiiiiliiliiii	i	.';	ail	Jli,.;	IB	 f Tim,,,,	I	f	•	;	•	'i-„..''ffii1•:	:	•	-t«A-, n	ii	,„ni	;	i.i.!!.i,il;,;	;!!:	:.i.:	^. •„„ f	£•	iti,,	....•iiiiii	i,..mi'il
                                                   'i	i'Mf >s
-------
                                                             Chapter 8
 8.4.1  Site Description

    The French Limited Superfund Site is located within the flood plain of the
 San Jacinto River as shown in Figure 8.17.  For many years, the site and
 another nearby Superfund site (Sikes) were used as sand pits. Both sites
 were converted for use as disposal lagoons for liquid chemical waste.  At the
 French Limited site, this practice continued from 1967 through 1972.  Resi-
 dues from the waste materials formed a sludge at the bottom of the lagoon.
 This material, which contained elevated levels of organic constituents  and
 metals, infiltrated the soils adjacent to and beneath the lagoon.
 Dissolved-phase liquids and NAPLs  migrated into the underlying aquifer
 and impacted soil and the shallow alluvial groundwater beneath and
 downgradient of the lagoon.
   In 1982, the site was placed on the National Priorities List and was  desig-
 nated for remediation under the CERCLA. The potentially responsible par-
 ties (PRPs) formed the French Limited Task Group (FLTG) to manage all
 activities at the site, including the remedial investigation (RI); feasibility
 study (FS); remedial action plan (RAP); and! engineering design, construc-
 tion, and operation. FLTG maintained control over the site and provided the
 primary contact with US EPA. Several contractors and consultants were
 employed on the site; Applied Hydrology Associates (AHA) provided  the
 primary consulting services related to aquifer remediation.
   The unique site geology determined the design features of the aquifer
 bioremediation system. Shallow alluvial deposits extend to a depth of about
 17 m (55 ft) and consist of sands, silts, and clays. This interval contains four
 units whose properties dictated the well configurations discussed below.
 These units, from the surface downward, are an unconsolidated zone (UNC),
 a sandy zone (SI), a discontinuous clay layer (Cl), and an interbedded silt
 zone (INT). The shallow alluvial deposits are underlain by a second, thicker
 clay unit (C2) which averages 20 m (70 ft) and contains minor sand and silt
 lenses. A silty sand unit (S2) is found below the C2 unit and averages 7 m
 (20 ft) in thickness.  The UNC unit consists of clay, silt, and sand over  an
 interval of 0 to 3 m (0 to 9.5 ft). The S1 unit extends from 3 to 9 m (9.5 to
 28.5 ft) and consists of coarse-grained sand with occasional gravel layers.
 The Cl unit ranges from 0 to 3 m (0 to 9.5 ft) in thickness and serves to
 separate the SI  unit from the INT unit where it exceeds 1 m (3 ft) in thick-
 ness.  The INT unit consists of interbedded silty sand and silt with variable
clay over a vertical interval of 5 to 8 m (16 to 25 ft) beginning at 9 to 12 m
(28 to 38 ft) below the groundwater surface.  Groundwater is first encoun-
tered within the UNC zone as depicted in Figure 8.18.
                                 8.37
-------
                                          •I!'!! •"' ' ''''"C •!'i1 I!	!/* ' "! 	I!'1, '
                     Case Studies
Hi
           111 illr '   " 1'
                              Figure 8.17
                   French Limited Site location Map
                             (not to scale)
                     Source: O'Hayre, Day, and Thomson 1993. Used with permission.
                                                                 •	I-
   The lagoon was excavated to the base of the SI unit. The contamination
subsequently addressed by the remediation was largely contained within the
SI and INT units. The SI unk is significantly more permeable (k= I0-3cm/
sec) than the INT unit (k = 10'4 cm/sec). The INT unit is underlain by the
Beaumont clay aquitard (C2) whose permeability (k = 10'7 cm/sec) and
thickness has served to prevent migration of dense nonaqueous-phase liquids
(DNAPLs) and dissolved phase contaminants to the deeper sandy unit (S2).
   The principal constituents found in the soils and groundwater on which the
groundwater clean-up criteria (Federal MCLs) were based were monoaromatic
hydrocarbons and chlorinated ethenes and ethanes. Concentrations of benzene,
toluene, 1,1-dichloroethane (1,1-DCA), 1,2-dichloroethane (1,2-DCA), and
vinyl chloride, as well as total organic carfion (TOO) were used to track
remediation progress. Benzene, 1,2-DCA, and vinyl chloride concentrations
were of particular interest because: (1) the health-based clean-up criteria asso-
ciated with these compounds was low, (2) they were the most widespread at the
                                                      8.38
-------
                                                                     Figure 8.18
                                                 French Limited Project Geologic Cross-Section
                                                                           Gulf Pump Road
                                                                                                                                   South
po
58
                                                                                                    Pie-Operation (SI and INT)
                                                                                                    Potentiometric Surface
                                                                                                C2
                                                                                                                                    Middle
                                                                                                                                     Clay
                                                                                                                                     Zone
              -50 -1
                                                                                                                                                         O
                                                                                                                                                         Q

                                                                                                                                                        f
                                                                                                                                                         CO
Source: O'Hayre, Day, and Thomson 1993. Used with permission.
-------
 Case Studies
 site, and (3) all three compounds are particularly mobile. These compounds
 were also the only VOCs that exceeded their MCLs outside the property bound-
 ary within the INT unit at the time intrinsic remediation was implemented.
   DNAPL residuals, including DNAPL pools, were present at the site.
 DNAPL was found in both the SI and WT units, mostly within the steel
 sheet-pile wall. The sheet-pile wall was driven through the shallow alluvial
 sediments into the underlying clay unit, C2, as a barrier to subsurface
 groundwater flow, containing the'lagoon source area. A few smaller occur-
 rences of DNAPL were observed south of the eastern end of the lagoon be-
 tween the sheet-pile wall and Gulf Pump Road. DNAPL samples collected
 from these areas contained up to 44% VOCs, which, by weight, consisted
 mainly of higher chlorinated VOCs such as chloroform, carbon tetrachloride,
 perchloroethene (PCE), and trichioroethene  (TCE).
   The dissolved-phase plumes within the SI and INT units extended
 downgradient of the lagoon south and southwest across Gulf Pump Road.
 Prior to implementation of in situ bioremediation, the plume covered an area
 of approximately 60,000 m2 (15 acres). Concentrations of 1,2-DCA in
 groundwater were as high as 8^0,000 jig/L near DNAPL sources and much
 higher in soils (380,000 ing/kg), soil'leacHate'^dO.^'')^^), and recov-
 ered DNAPL (14% by weight). Similar high concentrations existed for car-
 bon tetrachloride and chloroform. Very significant levels of vinyl chloride
 and benzene were also detected. The highest concentrations of 1,2-DCA and
 the other halogenated VOCs were detected immediately downgradient of the
 lagoon. In addition to the VOCs, other major contaminants, including poly-
 chlorinated biphenyls (PCBs),polynuclear aromatic hydrocarbons (PAHs)
 and heavy metals, were present in the chemical-rich sludge located at the
 bottom of the lagoon. Leaching of the lagoon  sludges and dissolution from
 DNAPL-impacted soils were identified as the sources of the dissolved-phase
 plume and, along with the properties of the contaminants, explains the distri-
 bution and composition of the dissolved-phase plume throughout the site.

 8,4.2  Regulatory Considerations
   Pilot tests of bioremediation of the lagoon sludges and subsoils (effec-
tively, a large in situ slurry reactor) were conducted in 1987 as part of the
FS. Based on the success of these tests, US EPA granted approval of in situ
bioremediation as the preferred remedial action for the lagoon  sludges. This
 •»   ' |,|  '"' •   ,     '"'"',  '  t' ,.  Ill" 	"illnil,1  'W  '7,,I .1, !!'i  " . ,. I. Hi''i ' ll "i !'  "• ' i   ,'n,' i.i ,"       	 .• j"" ,iii.
remedy was incorporated into the Record of Decision (ROD) in 1988.  The
ROD also specified a pump-and-treat remedial action for the contaminated
groundwater and associated subsoils surrounding the lagoon.
                                8.40
-------
                                                            ChapterS


   The alluvial deposits are considered to be, a drinking water aquifer and
 thus were required by the ROD to be in compliance with federal drinking
 water standards. The compliance boundary was defined in the ROD as the
 site boundary, approximately 32 m (100 ft) to the south of the lagoon.
                                                                    . -
                                        •I
 8.4.3  Design Approach

   The remedial program for groundwater and subsoils at the French Limited
 site had two major objectives: (1) isolate contaminant source areas, such as
 DNAPL zones, so that they could not provide a continuing supply of dis-
 solved constituents to groundwater; and (2) remove or degrade dissolved and
 adsorbed contaminants in the affected groundwater and subsoils outside the
 source areas.
   The sheet-pile wall completely surrounded the lagoon disposal area and
 was keyed into the underlying clay unit, providing containment of
 DNAPL-impacted subsoils in the lagoon soiirce area. In addition, the well
 pumping and injection network was operated so that hydraulic containment
 of the source area was maintained. One isolated DNAPL zone was identi-
 fied outside the sheet-pile wall on the basis of direct investigation and indi-
 rect groundwater quality data. This zone was hydraulically contained under
 the groundwater operating scheme and was eventually permanently isolated
 by an extension of the lagoon sheet-pile wall.
   The focus of the groundwater and subsoil in situ bioremediation program at
 the French Limited site was placed on the affected aquifer units outside the
 known DNAPL source areas. Enhancement of pump-and-treat systems was
 considered because  of the well-documented failure of such systems to
 remediate soils and groundwater in all but the most ideal conditions. Pump-
 and-treat systems fail as remedial technologies; because most contaminants have
 limited solubility and are easily adsorbed by soils, and effective groundwater/
 contaminated soil contact is rarely possible under field conditions.
   As contaminated groundwater passes through the pore space of the aqui-
 fer soils, dissolved contaminant constituents ,are transferred to the soil matrix
by the processes of adsorption and diffusion. In most cases, the mass of a
 contaminant constituent adsorbed onto soil particles or diffused into
 low-permeability lenses far exceeds  that dissolved in the groundwater. Slow
desorption and diffusion from soils into groundwater can act as a continuing,
long-term secondary source of groundwater contamination. Similarly, zones
of DNAPLs can compromise remedial activities because they also provide a
continuing source of pollutants to groundwater unless they can be removed
or isolated from active groundwater flow.
                                 8.41
-------
                                                                	I	——

                                                                 	•'•  i.
Case Studies
   The existence of soil contamination and/or DNAPL zones limits the abil-
ity of pump-and-treat remedial schemes to achieve stringent remedial crite-
ria. One of the advantages of in situ bioremediation, as discussed in Section
5.2, is that contaminants adsorbed onto the aquifer matrix are directly ad-
dressed. Direct bioremediation of DNAPL zones has not been demonstrated
because, in part, the very high concentrations of organics in DNAPL are
likely to be toxic to microorganisms.

8.4.4  In Situ Bioremediation Design Basis
                                 ,«,,"' ' "'^	i' "     i i • '     •  • "' '• •
   Considerable information was gathered during the lagoon pilot studies in
1987 to verify that aerobic bioremediation would be effective in meeting
remedial action criteria for chlorinated and nonchlorinated volatile organics
in the lagoon sludges and subsoils. As might be expected, the more mobile
organic constituents found in lagoon sludges were also found in the ground-
water outside the lagoon "source" area. Unlike the lagoon, however, aggres-
sive mixing and blending  of contaminated aquifer solids with
biologically-stimulating ingredients was not feasible. The challenge was to
establish subsurface conditions conducive to stimulating bioremediation by
injection and distribution of appropriate electron acceptors and nutrients
(i.e., to implement the Raymond Process).
   The in situ bioremediation system at the French Limited site incorporated
the addition of both oxygen and nitrate as electron acceptors. Historically,
oxygen has been the most widely used electron acceptor for in situ bioreme-
diation programs.  Nitrate was included as an alternate electron acceptor
because oxygen has limited solubility in water and would be rapidly con-
sumed by the high concentrations of TOC in the alluvial soils and groundwa-
ter. Nitrate is more soluble than oxygen and, therefore, can be distributed
more widely within the aquifer system. Stimulation of different consortia of
microbial populations under aerobic and denitrifying conditions would result
hi direct oxidation of monoaromatic hydrocarbons (with the exception of
benzene), under denitrifying conditions and promote co-metabolic biodegra-
dation of some chlorinated compounds. This is believed to be the primary
mechanism for bipdegradation of chlorinated organic constituents under
these conditions. Reductive dechlorinatlon of the chlorinated compounds
would be anticipated in the absence of nitrate and oxygen and in the pres-
ence of sulfate and areas of methanogenic activity.
   In the presence of nitrate or oxygen, degradation processes involving
other electron acceptors would be inhibited.  Nitrate was added to the injec-
tion water stream, initially in the form of concentrated aqueous solution of
ammonium nitrate. Its addition was controlled so that the concentration of
                                 8.42
-------
                                                             Chapter 8
 nitrate in the receiving water would not exceed the drinking water standard
 of 10 mg/L as nitrogen. Ammonium nitrate is also a source of nitrogen nu-
 trient. During the course of the remediation, the nitrogen source was
 changed to potassium nitrate based on concerns that ammonium could be
 exerting an oxygen demand in the aquifer during nitrification.
   Oxygen was added to the injection water stream by direct injection of gas-
 eous oxygen. The oxygen was stored on-site in liquid form and converted to
 gaseous form in vaporizer units. Liquid oxygen was particularly attractive
 because it was already being used on-site for lagoon biorerriediation in much
 larger quantities than would be required for (he aquifer remediation. ,
   Phosphate addition (as potassium tripolyphosphate [KTPP]) was evaluated
 through a series of precipitation tests using site water from several wells. The
 tests indicated that precipitation could lead to plugging of the injection wells
 and the adjacent formation if KTPP were used to amend recycled groundwater.
 The recycling of treated groundwater was a concern with respect to potential
 precipitation of calcium, magnesium, and iron if phosphate were added and
 possible precipitation of iron oxy-hydroxides following injection of oxygen-rich
 water. The solution to this potential problem was to discharge treated water to
 the lagoon and the San Jacinto River and use low total dissolved solids water
 from a deeper aquifer for injection purposes.
   Four test areas, two in each of the S1 and INT units, were constructed to
 establish operational parameters for the in situ bioremediation system before
 going full-scale. The major purpose of the test areas was to confirm that addi-
 tion of chemicals and oxygen would not have detrimental effects on the aquifer
 system. Injection water was amended in phases. First, water with no nutrient
 or oxygen amendments was injected. This "(blean water" front had the effect of
 flushing groundwater containing relatively high concentrations of iron and other
 cations away from the injection well screens. The second phase of injection
 involved amendment of injection water with ammonium nitrate and KTPP. The
 last phase of injection included the addition of oxygen.
   Despite a thorough hydrogeologic investigation, it was not possible to
 locate all injection and extraction wells in the optimum locations.  Therefore,
 an inherently flexible approach to well placement was adopted. Injection
 and extraction wells were initially placed in fairly regular patterns similar to
 a five-spot array.  Because of the order-of-magnitude difference in hydraulic
 conductivity between the SI and INT units and the semicontinuous Cl clay
layer that separated the two units, separate injection well/recovery well sys-
tems were designed for each unit. The spacing between wells was larger for
the more permeable SI unit than for the Cl layer. After the system was
operated for a sufficient length of time to assess performance with respect to

                                 8.43
-------
                                           1	'SHi	I	•	.•••	••
Case Studies
the time needed to remediate individual areas, additional wells were added to
accelerate remediation of "hot spots" and/or areas of relatively low hydraulic
conductivity. This empirical/observational approach allowed for the place-
ment of some wells based on aquifer response and for flushing and electron
acceptor delivery to those portions of the aquifer that would otherwise re-
quire longer times for remediation.
   The design called for a phased sequence of injection water amendments
to be applied to the entire injection well field. The sequence of amendments,
which took into account the higher mobility of nitrate compared with oxygen
in the subsurface^ was designed so that three zones of subsurface activity
would be developed. The leading zone, developed farthest away from the
injection well, was a zone where flushing enhanced conditions for in situ
bioremediation by reduction of high concentrations of organic and inorganic
constituents.  This zone was followed by an anaerobic denitrifying zone
where nitrate in the injection water was the primary electron acceptor. Clos-
est to the injection well was the aerobic zone. Rapid utilization of nitrate
and especially oxygen, retarded penetration of the electron acceptors relative
to the clean water front.
                                               i      ... 	   ,,
   The injection of clean water, combined with groundwater recovery, served
to flush soluble organic compounds (to remove contaminated mass) and
inorganic ions (iron, calcium, magnesium, etc.) from the aquifer to reduce
complications (i.e., reduced permeability) from nutrient and oxygen addi-
tion. Removal of unidentified organic compounds was not only beneficial
because of the reduced TOC levels, but also because removal of such mass
could reduce the demand for electron acceptors.
        -       '     ..     •   . .•/	'   •   ":;..     .:•  I'. "" :• ;.  :v     ;. ,•. - ! v
   The in situ bioremediation sequence of flushing, denitrifying conditions,
and finally aerobic conditions was designed to take advantage of the natu-
rally different rates of metabolism occuring under different electron acceptor
conditions and to stimulate different consqrtia of microorganisms at different
times within the aquifer. The strategy was to establish beneficial
co-metabolic biodegradation processes and to maximize the biodegradatioh
of the wide variety of chlorinated and nonchlorinated organic constituents
found at the site.
   Recovered groundwater was pumped to an equalization tank and then
treated in an aboveground biological reactor. The reactor also treated water
pumped from the lagoon to maintain water levels in the lagoon.  Microbial
cultures obtained from the lagoon were maintained for periodic addition to
the reactor. The discharge from the bioreactor was polished using carbon
adsorption, discharged to the San Jacinto River and, when necessary, re-
cycled to the lagoon.

      .'   '          .      •        8.44 J   "''f'"   :   '        '   ''  J?
-------
                                                           Chapters
   Initially, groundwater was recovered from 49 SI wells and 33 INT wells
and amended water was injected through 11 SI wells and 13 INT wells. As
a result, 700 m3 (200,000 gal) of groundwater containing approximately 400
mg/L of TOC as well as VOCs and metals were being removed from the
aquifer and treated on a daily basis. This corresponded to. the removal of
approximately 300 kg (670 Ib) of TOC per day.
   During operations, groundwater samples were analyzed for TOC, VOCs,
pH, nitrate, dissolved oxygen, ammonium, ortho and total phosphate, and
other inorganic parameters.

8.4.5 Remedial Progress          \
   Remedial progress at the site was primarily gauged on the basis of
changes in dissolved constituent concentrations in groundwater over time as
measured by periodic sampling of monitoring wells in both SI and INT
units. The primary VOCs used to determine remedial progress were ben-
zene, 1,2-DCA, and vinyl chloride.      j
   1,2-DCA was representative of the chlorinated solvent compounds found in
the DNAPL in the source areas. Dissolved concentrations of 1,2-DCA up to
800,000 fJg/L were observed near DNAPL source areas, and benzene exceeded
1,000 pg/L throughout most of the area immediately downgradient from the
lagoon. The concentration range of these compounds in the INT groundwater
unit before remediation was similar but slightly less widespread.
   After 15 months of system operation, monitoring indicated reduction in the
dissolved concentration of all chlorinated organic compounds in the Si ground-
water to below the 5 jug/L detection limit, except in the immediate vicinity of
known DNAPL areas. Continued monitoring confirmed that these reductions
continued through December 1994. During the same period, significant reduc-
tions in benzene concentrations were observed. By January 1994, benzene was
the only VOC in S1 groundwater that exceeded its drinking water clean-up
criteria at the point of compliance. The reduction in constituent concentrations
in SI groundwater through time is illustrated by the results from monitoring
well S1-106. Denitrifying conditions have prevailed in the vicinity of this well
throughout most of this time period.  This was indicated by relatively low con-
centrations of nitrate in the groundwater compared with ammonium and potas-
sium, which were injected with the nitrate.
   A significant portion of the reduction in dissolved VOC concentration
observed in the SI groundwater can be attributed to flushing by injection
water.  However, it is not possible to explain the reduction entirely by this
mechanism.  Indirect evidence indicates that biological degradation under
                                8.45
-------
Case Studies
denitrifying conditions played a significant role in the reduction of the chlo-
rinated solvent constituents.  Numerous studies have indicated that benzene
does not readily degrade under denitrifying conditions, hence it can be used
as a conservative indicator of flushing effectiveness when denitrifying condi-
tions prevail. Given appropriate adjustments for different sorptive character-
istics of the compounds of interest, the projected change in concentration of
a constituent due to flushing only can be estimated on the basis of observed
reductions in benzene concentrations at a monitoring well. Flushing is un-
likely to have been the only mechanism that reduced VOC concentrations in
the S1 groundwater in the vicinity of well S1 -106.
   The reduction of benzene concentrations by in situ biological mechanisms
is not likely to occur under denitrifying conditions.  DO monitoring indi-
cated that oxygen in the injection water was rapidly consumed within a fairly
short distance of the injection wells. Oxygen utilization probably occurred
during both biological and chemical subsurface reactions.  As a result, the
migration rate of the aerobic injection water "front" was between 0.45 and
1.2 m (1.5 and 4 ft) per month compared with a fluid migration rate of up to
30 m (100 ft) per month.  Aerobic bioremediation became more efficient as
flushing and denitrifying bioremediation continued to decrease the total
organic carbon concentration in the aquifer ahead of the aerobic front.
   The extent of oxygenated groundwater (arbitrarily set at greater than 5
mg/L) steadily increased in both the SI and INT units over the course of
remedial operations. Reductions in benzene and vinyl chloride concentra-
tions in groundwater to below analytical detection limits coincided with the
"breakthrough" of oxygenated conditions. This indicated that bioremedia-
tion was a major process  influencing the breakdown of these organic con-
stituents in the subsurface.
   The reduction of VOC concentrations in SI groundwater to below the clean-
up criteria over some areas of the site allowed certain parts of the active reme-
dial system to be turned off in September 1993. These areas were monitored to
evaluate the potential for rebound of groundwater concentrations that might
result from slow desorption or diffusion of contaminants from soils into ground-
water.  More than a year after shut-off, there was no indication of constituent
concentration increases in monitoring wells. Slight increases in VOC concen-
trations at pumping wells were attributed to the small volume of dewatered
SI unit in the immediate vicinity of the well mat was less effectively remediated
under continuous pumping conditions. These increases were addressed by the
initiation of a "pulsed" pumping program over a few months. This program
was successful in achieving remedial objectives at the wells, and the pumps
were eventually turned off in March 1994.
                                 8,46
-------
                                                            Chapters
   Predictably, remedial progress in the lower permeability INT unit was less
rapid than in the SI unit. As with the SI unit, affected areas outside of
known DNAPL zones showed significant reductions in constituent concen-
trations as a result of the in situ bioremediation system operation, but attain-
ment of clean-up criteria took considerably longer than in the SI.
   To avoid continuing remediation activities for an extended period in areas
where achievement of remedial goals was slower due to either higher con-
taminant levels or lower permeability, additional INT wells were installed.
A total of 66 additional INT injection and production wells were added dur-
ing the course of active operations. Installation and use of these wells corre-
sponded with removal of some SI and later INT wells from the active sys-
tem, and thus, the pretreatment and posttrentment requirements remained
relatively constant over most of the operating period.
   An important aspect of the project was the removal of a large portion of
the organic mass (TOC) through groundwater recovery. This greatly reduced
the demand on electron acceptors, allowing electron acceptor consumption
to be more beneficially and cost-effectively applied to VOC biodegradation.
During the project, approximately 182,000 kg (200 tons) of TOC were re-
moved from the aquifer via groundwater recovery.

8.4.6 Intrinsic Remediation
   The ROD for the site stated that groundwater recovery and treatment was
to continue until modeling showed that the concentration of volatile organics
would be reduced to a level consistent with the 10'6 human health criteria
through natural attenuation in 10 years or less. This allowed FLTG to inte-
grate intrinsic remediation at the end of the process as a polishing step. This
approach also provided the potential to meet the human health and environ-
ment protection objectives in a manner consistent with the ROD, while re-
ducing overall project costs by eliminating a significant portion of the
long-term O&M costs.

8.4.6.1  Intrinsic Remediation Modeling
   The basis for determining if intrinsic remediation was viable was the use
of modeling techniques to simulate future groundwater quality at the French
Limited Superfund Site 10 years after shutting off the active remediation
system.  To achieve agency approval to terminate active remediation, it was
necessary to demonstrate that continued dq^radation/attenuation would re-
sult in site clean-up criteria being met at and beyond the compliance bound-
ary within 10 years.
                                8.47
-------
Case Studies
   As discussed in Section 5.3 of this monograph, the modeling of intrinsic
remediation is based on a combination of transport mechanisms, advection
and dispersion, retardation (adsorption/desorption of the contaminants by the
aquifer matrix), and degradation (biodegradation and, potentially, chemical
reactions).  To the extent possible, site-specific data should provide values
for the parameters used in the model.
         ;       	           . -     	 ; ;_, • •   "  j.	;        ; „ 	!•-- •
   Where practical, numerical models are preferred over analytical models
for evaluation of complex sites. In this case, visual MODFLOW® was se-
lected from among several well-established codes to simulate post-
remediation groundwater flow. The effects of hydro-dynamic dispersion,
retardation, and biodegradation (natural attenuation) were superimposed on
the visual MODFLOW® output using Biotrans®.  BioTrans® was selected
because it simulates oxygen- (or oxygen equivalent-) limited degradation of
multiple chemical species.
   For this site, a large volume of data was available with respect to ground-
water quality and aquifer characteristics. Groundwater quality, nitrate, and
DO data were obtained from the results of the November 1995 sampling
event and were supplemented with data from earlier sampling events where
necessary.  The data were assigned to a grid and contoured using SURFER®,
with appropriate controls for peripheral areas.
   Individual contour maps were prepared for benzene, 1,2-DCA, and vinyl
chloride. As previously discussed, these compounds were selected as indica-
tor parameters because they had the lowest health-based clean-up criteria of
the VOCs present in the aquifer, are the most mobile of the VOCs, and were
the only VOCs exceeding their MCLs outside the property boundary. Other
VOCs were present in very limited areas and thus were not included in the
modeling process. Individual contours were also generated for TOC and for
total electron acceptors; a concept comprising a combination of DO and
nitrate concentrations, the parameter "DO+," was derived as follows
(Borden, Gomey, and Becker 1995):

               DO+ = DO + 2.5(NO3 - N)                        (8.2)

All three types of contour maps were prepared for both the Si and INT units.
                                   .., ." 4,1 'I • ,  "I'll1 >,iil '!'  I !'      " .' I-"  . , ' i "•"',. ."' l"'illi,, ' '
   Visual MODFLOW® was used to model the regional San Jacinto basin
shallow alluvial aquifer to produce long-term, steady-state, hydraulic gradi-
ents and flow velocities. Upgradient and downgraclient boundaries were
fixed-head boundaries based on long-term hydraulic gradients measured at
the site before active remediation commenced. Crossgradient boundaries
were based on mapped geological contacts  with the underlying and adjacent
                                     -•'•    .' .'i          '    '    "  •'•
                                ' 8.48
-------
                                      ;                      Chapters

 Beaumont Clay aquitard which were modeled as impermeable barriers. It
 was assumed that long-term vertical groundwater migration between the
 upper SI and lower INT shallow alluvial aquifer units would be insignifi-
 cant. The steel sheet-pile flood wall was also treated as an impermeable
 barrier. The aquifer properties in Table 8,2 were used.
                               Table 8.2
                  Properties Used in Aquifer Modeling
Unit
UNC
SI
Cl
INT
C2
Top Contact
m (ft) MSL
4.6 (15.0)
1.5 (5.0)
-4.6 (-15.0)
-6 (-20.0)
-12 (-40)
Horizontal
Thickness Permeability
m (ft) mid (ft/day)
3 (10.0)
6 (20.0)
1.5(5.0)
6 (20.0)
modeled as
0.3 (1.0)
6 (20.0)
0.3 (1.0)
1.5(5.0)
impermeable base
Storage
Coefficient
Type
specific yield
specific
storage
specific
storage
specific
storage

Storage
Coefficient
0.1
0.00003/m
(0.00001/ft)
0.00003/m
(0.00001/ft)
0.00003/m
(0.00001/ft)

 Source: O'Hayre, Day, and Thomson 1993. Used with permission.
   Groundwater elevation measurements collected as part of the long-term
monitoring program for the site were used to provide ongoing flow model
calibration. Groundwater flow models of four subareas selected for natural
attenuation modeling were developed using the head (groundwater elevation)
data generated by the regional San Jacinto shallow alluvial aquifer model.
Each model covered one area that did not meet compliance criteria based on
VOC plume results.
                                     i
8.4.6.2 Model Input and Results
   The model input parameters are shown in Table 8.3.  The soil
bulk-density, 1.7 g/cc, and grid spacing were kept constant.  However, grid
spacing varied among the four areas modeled.  The other parameters were
varied over the ranges shown in Table 8.3 to determine model sensitivity to
parameter variability.                  i

                                 8.49
-------
Case Studies
                                      1, • • 4 i«    I , • I
                                Table8.3
                         Model Input Parameters
                                      -;  • •.", Base
          Input Parameter              Unit     Case     Uncertainty/Sensitivity Range


Grid spacing                       .   ft      5/l5
Time step                           day      5>             0.1-800

Half life"                            day      60             30-50

Initial DO/nitrate (as equivalent "DO or
DO+")b

Dispersivity
                                              1 
-------
                                                            Chapters
initially used for the base case, 1.5 m (5 ft), was not well supported, whereas
the values of 5.3 m (17.5 ft) (INT unit) and 19 m (61 ft) (SI unit), obtained
from applying the l/10th rule are strongty defensible.
                               Table 8.4
               Transport-Related Constituent Parameters
Chemical
Benzene
U-DCA
Vinyl chloride
TOC
*.
83
33
25
1.25 • 106
Kd at fOT
= 0.12%
0.0996
0.0396
0.0030
1.500
KM at £•„
= 0.15%
0.1245
13.0495
0.0038
1,875
KdatfTC
= 0.18%
0.1494
0.0594
0.0045
2,250
Source
Howard
Howard
M&W
ES&T
Source: O'Hayre, Day, and Thomson 1993. Used with permission.
   Consideration of nondegradable TOC did not support the use of a
single value for the entire site; further evaluation of the French Limited
on-site bioreactor studies strongly supported the use of 50% of the mea-
sured TOC value.
   The model outcome is presented in Table 8.5.  The model demonstrated
that vinyl chloride, 1,2-DCA, and benzene in all simulated areas will de-
crease by between 96% and 100% in 10 years and that no areas of VOCs
exceeding site clean-up criteria will exist at or beyond the compliance
boundary within this time frame.

8.4.7 Conclusions                 i
   Based on the modeling results, it was determined that natural attenuation
would meet the requirements of the ROD. As an extra measure, ONE month
prior to shut-off of the active remediation systems, target SI and INT unit zones
were dosed with nitrate, dissolved oxygen, and phosphate to provide elevated
levels of nutrients and electron acceptors to further facilitate degradation during
the early phases of intrinsic remediation. On December 15,1995, active
remediation was terminated and intrinsic site restoration was initiated.
                                 8.51
-------
Case Studies
• :"i, , • ''''•,. , " i ." , .-,..;' • , ;- '.i 	 ••; . - : 	 : 	 ';:. i •-.,'.
Demonstration Runs —
Parameter
Grid spacing
Time step
Halflife
Final DO/nitrate
Dispersivity
DOstoichiometry
:'': ' : :, . " ' '" '";. f°°
Effective porosity
Nondegradable TOC
Benzene max. cone. % red.
i nun in i ' ' ' • ', , "
1,2-DCA max. cone. % red.
Vinyl chloride max. % red.
TOC max. cone. % red.
DO+ max. cone. % red.
Benzene % reduction
1,2-DCA % reduction
Vinyl chloride % reduction
TOC % reduction
DO+ % reduction
Benzene distance
1,2-DCA distance
Vinyl chloride distance
•with INT-60 converted to Injection (1 1/95).
Ill IK
III • "" " , , •
• : llh , , . 'i.
Table 6.5
Input Parameters
INT
Unit West
m(ft) 4.6(15)
day 50
day 60
9/95
m 53
(ft) (17.5)
''' - '" 2.75
% 0.15
02
% 50
% 82
,1,1' ' • 	 • ,
% »
% 95
% 3
% 20
% 91
' • " % 99
% 95
% 55
% 13
m(ft) 0
m(ft) 0
m(ft) 0

1 , ,,
8.52
and Summary Results
INT
Central
3(10)
50
60
9/95*
5.3 .
175
0.15
"0.2
50
55
100
98
40
18
93
,r,w
96
94
2 "
'•'V
.1
0
0
li "
: i... • . •
., : t i.
I
i
i
i
INT
Wall
1.5 (5)
50
60
9/95
53
2.75
0.15
0.2
50
100
100
100
100
14
100
100
100
100
0
0
6
0

u. ' !|" '".' ,
SI East
3.7 (12)
50
60 " ' ' ' 	
9/95
19
1 Jl/il." , .,, ! 	 ' I'll,
Z75
0.15
035
50
98
100
loo 	 "
79
50
99 ''"' ' ! ' 	 " '
loo 	 	
100
92
25"'": 	 J":: ' '
0
o"1" ' !il 	 ''
0
:i . '» "' .u ,.' i
'I1,! 'LI ' ' . i""i.!"' I
' -1 '!' •' " ii
1 	 	 ",; . . i.1 II
r 'II i ' .'III
1 i . •„,: • j|
i ,'S1 ' ", i '. j
	 • \ *" '••
ii ii 'i; 	 .M. 'film ! J,, jiki'Lilllillll'iili1 . iW"',1 ' ., II'
-------
                                                         Chapter 8
8.5  Bioremediation of
Explosives-Contaminated Soil Using
Composting Technology*
                                    i
   This case study provides an example of the application of composting
technology, a modification of soil cell treaitment as discussed in Section 4.4,
for the treatment of explosives-contaminated soil. The composting tech-
nique was developed by the Army Environmental Center and implemented at
the Umatilla Army Depot Activity (UMDA) near Hermiston, Oregon, by
Bioremediation Service, Inc., of Portland, Oregon. Agricultural waste prod-
ucts (or amendments) were blended with the contaminated soil during treat-
ment. Specialized soil turning equipment mixed the compost for optimum
biological action and homogeneity.  Homogeneity of the compost mix en-
sured rapid degradation of all contaminants. Physical and chemical proper-
ties were closely monitored to ensure that thermophilic bacteria played a
dominant role in the degradation process.

8.5.1 Site Description
   UMDA is a nearly 8,000-hectare (20,000-acre) munitions storage facility
located in northeastern Oregon. Past activities performed at UMDA in-
cluded the demilitarization of conventional munitions. The facility used a
pressurized hot water system to remove arid recover explosives from a vari-
ety of ordnance including 227- and 340-kg (500- and 750-lb) Composition B
bombs (60% royal demolition explosive [RDX] and 40% 2,4, 6-trinitrotolu-
ene [TNT]) and 90-mm projectiles.  The explosive-laden hot water was al-
lowed to cool in various containers, and the RDX/TNT sludge, which pre-
cipitated out, was removed and reclaimed in a variety of ways.
   Weekly flushing and draining of the plant's storage tanks and vats pro-
duced 567 m3 (150,000 gal) of contaminated water that was flushed down an
open trough outside the building into two evaporation lagoons covering ap-
proximately 930 m2 (10,000  ft2). These "washout" lagoons were used inter-
mittently; one lagoon was filled with the week's flush of "pink water" (from
its characteristic color), while the other was being emptied of its evaporation
residue. These residual solids were then transported to another area of
UMDA and burned.
•Adapted with permission from the paper by Emery and Faessler (11996) of Bioremediation Service, Inc., submitted to the
Engineering Foundation for publication in the Annals of the New York Academy o1Sciences.
                               8.53
-------
	'I
. j i
Case Studies
                                               1 .                 •	 ; . :

   Although this wastewater disposal technique was acceptable practice at
the time the plant was in operation, the site was placed on the National Pri-
orities List in 1987 because contamination was affecting not only the soils
under the lagoons, but also the groundwater.
   The soil from the munitions washout lagoons contained high concentra-
tions of TNT and RDX, as well as lower levels of high melting explosive
(HMX). Also found at the site were lower levels of tetryl-dinitrotoluene
(DNT), trinitrobenzene (TNB), dinitrobenzene (DNB), and nitrobenzene
(NB) contaminants as impurities or degradation products of TNT.
        ' ' ,   '  '. '     '  i  '  '.     • ;J'!'' .'rfX i Vfti/'! I "'  I" Hif1' '  " , j J' "'   ,  """ •' '	'Jj11'"1'"
   Of the several possible remediation techniques that could have been used
to decontaminate the site, biological remediation was the method of choice.
The Army Corps of Engineers, Army Environmental Center, US> EPA, and
Oregon Department of Environmental Quality sponsored the major innova-
tive composting remediation project at UMDA.
   As indicated in Section 4 A of this monograph, composting is not a new
technology, but if properly designed and implemented, it has been shown to
produce dramatic degradation and transformation of recalcitrant contami-
nants, such as nitroaromatics and nitramines, in contaminated materials. The
resulting treated soil is rich in organic matter and nutrients. The Army has a
research program underway to evaluate possible re-use or recycling of the
treated material

8.5.2 Pilot-Scale  Remediation
   A comprehensive trial test was performed for equipment and process
optimization before full-scale production was authorized. Many bench-scale
and pilot-scale tests performed by various organizations preceding the trial
test showed favorable results for the degradation of munitions waste com-
pounds  in contaminated soil using a compost system.  The trial test contrac-
tor used this information to optimize field-scale efficiency, as well as con-
firm that all contaminants and contaminant byproducts were degraded to
acceptable clean-up levels. Initial pilot studies demonstrated that
munitions-contaminated soils could be aerpbically composted in 30 to 40
days. During the comprehensive trial test at UMDA^ more than 280 Method
8330 tests were performed, which demonstrated that the clean-up goal of 30
mg/kg TNT and RDX could be achieved in less than 22 days (Figures 8.21
and 8.22), while clean-up levels for HMX occurred more rapidly (Figure
8.21). Daily turning of the compost pile produced the fastest degradation
results.  Two degradation byproducts, 2-Am-DNT and 4-Am-DNT, tempo-
rarily accumulated (Figures 8.24 and 8.25), but were subsequently degraded
                                        8.54
-------
                                                            Chapter 8
to below detection levels following further composting treatment. All other
byproducts (i.e., 2,4-DNT; 2,6-DNT; 1,3-DNB; NB; 1,3,5-TNB, and
methyl-2,4,6-trinitrophenylnitramine) were also shown to be below their
detection levels.

8.5.3 Full-Scale Remediation

   From July through December 1995, over 3,800 m3 (5,000 yd3) of
explosives-contaminated soil were successfully treated. Analytical results
for this treated soil show that 93% of all results for TNT were below the
detection level of 4 mg/kg, while the RDX level in 68% of all samples were
below its detection level of 2 mg/kg. All sampling points were below 30 mg/
kg for both TNT and RDX. Three hundred thirty Method 8330 analyses
confirmed these results.
                              Figure 8.19
    Pilot-Scale Results of TNT Degradation in Windrow Composting Units
      350*
    I
      3001
      250
      200
      150
           \\v
            \  \\
—E— Turned Daily
— A. - Turned Tri-Daily
—•••• Tlimed Variable
Each data point on the average of 12 EPA Method 8330 analyses

Source: Emery and Foessler 1997. Used with permission.
                                 8.55
-------
l|	''''Winiill'.'i1	IP'l'.fiu	.' ill "if " :	llii11 l1 ' 	|i
                        n i nil in inn   i in 11 '"'i is1'
                                 Case Studies
	 1 	 : 	
Pilot-S(
350
300 '
250
200*
1
150
100
50
0
(
Figure 8.20
sale Results of RDX Degradation in Windrow Composting Units
	 ' • 	 i • • • • '• " 	
^
l\
,\\
\\\
A v
A k
\\ ^
\\ ^
V_ >s*-~
i
) 5 10 15 20 25 30
Day
i
' , V:: ' '' 'j 1' i ' " ' H1"
— O— Turned Daily
— A, . Turned fri-Dally
....... Tbrned Variable
Each data point on the average of 12 EPA Method 8330 analyses
Source: Emery and Fbessler 1997. Used with permission.
                                    The rate of degradation during full-scale remediation improved by over
                                 25% from the initial trial test. Required action levels were achieved in 12 to
                                 15 days of treatment.
                                                                                „ ,|
                                 8.5.4 Composted Soil
                                                      ,    '           "   	:'     , i ,  I" .l.'..1  ' :,',	'• _  \ • '  I '.
                                    Composted soil is very rich in humic material and is an ideal candidate
                                 for recycling and reuse.  An accepted method of evaluating bioremediation
                                 treatment residue is the use of toxicity and teachability''tests. This approach
                                 is consistent with the Superfund National Contingency Plan objectives of
                                 evaluating the toxicity, mobility, and volume reduction effects of innovative
                                 treatment technologies.
                                        1   •' !    • ;         '      '   	   "   .   ' • • I • nil , '    .,..:„   .11      . . I I'	
                                                                  8.56
-------
                                                              Chapter 8
                               Figure 8.21
    Pilot-Scale Results of HMX Degradation in Windrow Composting Units
       35
       30
       25
       20
       15
       10
                           10
15
Day
                                               20
                                                         25
                              30
—Q— Turned Daily
— A, . Turned Tri-Daily
•••••••• Turned Variable
Each data point on the average of 12 EPA Method 8330 analyses

Source: Emery and Foessler 1997. Used with permission.
8.5.4.1  Previous Toxicity and Leachability Studies
   Toxicity and leachability studies have been performed on composted soil
with known TNT-biotransformation products. Results suggest that the final
biotransformed product of TNT may be a polymeric species of very limited
solubility.
                                       i
   One study showed that there was practically no leachable 14C activity after
a simulated 1,000-year leaching test (Griest et al. 1991). Based on the
1,000-year acid rain leaching simulation, any release of transformed explo-
sive from composted soil into the environment would not be significant.
   Toxicity tests that have been used on bioremediation treatment residues
include:  (1) Microtox™; (2) Ames assays for mutagenicity; (3) aquatic
toxicity tests on soil leachates; (4) oral rat feeding studies; and (5) earth-
worm toxicity tests.
                                  8.57
-------
i 'ill 1" '!"! , "'III;11
         Case Studies
                                        Figure 8.22
         Pilot-Scale Results of 2-Am-DNT Degradation in Windrow Composting Units
                                    10
15
Day
                                                        20
                                                                  25
                              30
         —-CJ— Turned Daily
         —A.- TurnedTri-Daily
         •••••"• "Ilimed Variable
         Each data point on the average of 12 EPA Method 8330 analyses
         Source: Emery and Fbessler 1997. Used with permission.
                                                                              	!;•	It
           The final conclusions developed from this study on previous pilot-scale
         compost residues from UMDA are:
            •  i  ,       •   • . .	 ..; ,  ",„ •', '     -i..  ;, i:  .  ' , ; -iJ.,, ,    • ; • ,,„•!,	i  '. ;	
                 •  composting is a safe and effective process for decontaminating
                   and detoxifying explosives-contaminated soils; and
                 •  the compost product should allow the re-establishment of plant
                   and animal populations in land application, although some in-
                   hibitory effects were noted for some plant types.

         8.5.4.2 Plant Uptake Studies in Composted Soil
           The degradation results achieved during the first full-scale composting
         effort at UMDA were better than previous pilot-scale treatability tests. The
         composted soil provided an excellent opportunity to demonstrate the reuse
         and recycling potential of the soil.
                                          8.58
-------
                                                              Chapter 8
                               Figure 8.23
 Pilot-Scale Results of 4-Am-DNT Degradation in Windrow Composting Units





i





3

?,.")


1.5



	 * 	 : 	 	 - 	
/ \
i \

/ \
/ \
/ *


/ *









L
05 10 15 20 25 30
Day
     Turned Daily
-A - TurnedTri-Dally
•••••••• Turned Variable
Each data point on the average of 13 EPA Method 8330 analyses
Source: Emery and Foessler 1997. Used with permission.
   From late July through late October, several plants were successfully
grown in the treated soil. Those plants included indigenous plants (sage-
brush); agricultural varieties of fruits, vegetables, and grasses (statesman
perennial ryegrass, chewing fescue, creeping red fescue, Kentucky blue-
grass); carrots (scarlet nantes); turnip (purple top white globe); strawberry
plants; as well as flowers common to landscaping (carnations, chrysanthe-
mums, marigolds).  In November, these plants were harvested and prepared
for analysis. Analytical results to date for TNT, RDX, 2,4-DNT, and
2,6-DNT showed nondetectable levels in all plant varieties investigated.
                                        I
8.5.4.3 Additional Testing
   Studies that augment previous evaluation of the potential toxicity of
composted soil were also commissioned. One involves "503 Class A"
sludge pathogen testing (40 CFR 503 Standards).  The Code of Federal
Regulations sets standards for pathogenic bacteria present in compost. The
treated soils from UMDA met or exceeded all criteria that include Helminth
ova, Salmonella sp., Fecal Coliform and Enteric Virus.  A second
                                  8.59
-------
          >"w	•flSjy'"	:'  '.;	 ' "i      ' ;	'"•'•"•'<<    '•   :.',.. ;..•:	,• .jfj;••'•"? tfjyn i"!'" {;/•;"-;?' ^	,/:?"i/!".' ill	;j|,5.;£jti
:if VI        "  <  • .I;1

                    Case Studies
                    comprehensive study involving the following tests is currently underway to
                    further define the toxicity of compost-treated munitions-contaminated soils:
                                                          	  	I "'    	         	
           '>	W    •        •  Plant Toxicity Tests:  .    	'
                                 •  Root Elongation, and
                                 •  Early Seedling Growth;
                            •  Earthworm Toxicity Test;
                            •  Freshwater Elutriate Tests:
                                 •  Algal Growth Test,
                                 •  Invertebrate Survival and Reproduction Test, and
                                 •  Fish Survival and Growth Test; and
                            •  Freshwater Sediment Tests:
                                 •  Amphipod Survival Test, and
                                 •  Invertebrate Bioaccumulation Test.
               ||,          '    •  ' • '   ", ,   '      '" ''i "     «••'','<•"•    .1 :  j ' •  y '»  .•)•''. -    ' ;  ' '  '  "'
                    8.5.5 Remediation Costs
                      Table 8.6 summarizes projected compost treatment remediation costs per
                    ton of munitions-contaminated soils based on pilot- and full-scale
                    remediation trials conducted at UMDA, These costs include installation and
                    removal of the composting system components, asphalt pad construction,
                    runoff containment, asphalt recycling, and final grading. Also included are
                    treatment/containment tents designed to reduce volatile emissions and pro-
                    vide positive moisture control at the site. These costs were based on a 2-year
                    life. It is important to note that the reported costs include mobilization, de-
                    mobilization, site preparation, special facilities, trial tests, and bond insur-
                    ance costs, a significant portion of which would not be necessary in future
                    projects.

                    8.5.6 Conclusions

                      Nearly 3,800 m3 (5,000 yd3) of soil have been successfully treated, and
                    more than 70% of all analyses indicate nondetectable levels of both TNT and
                    RDX in the compost-treated soil. The U.S. Army Corps of Engineers esti-
                    mates that over $2.6 million is being saved using bioremediation at UMDA.
                    All indications are that the composted soil contains no intermediate
                    byproducts of explosive contaminants, and the original primary contami-
                    nants have been degraded to well below the clean-up levels established in the
                    baseline risk assessment for protection of human health and environment at

           ••   ::;• ;     ,   '.,   •  '      .     . ;  ,   '    8.6Q  ":  "   '•-"  !":
-------
                                                               Chapter 8
the site. Furthermore, the composted soil is suitable for recycling or reuse since
it complies with the National Contingency Plan requirements as follows:
        •  the soil residue is nontoxic;
        •  the soil residue contains no leachable contaminants;
        •  transformation products that may exist are less toxic than TNT,
           and the covalent bonds that fix these metabolites to-soil or com-
           post are extremely rugged. Repetitive aqueous leaching of the
           composted soil and also ultraviolet light irradiation followed by
           leaching suggest that the product should not be released appre-
           ciably by acid rain or sunlight (Major, Bollag, and Ames 1994);
        •  the composted soil meets the same requirements as a Class A
           biosolid under US EPA 503 regulations;
        •  the composted soil is very rich in humic material and nutrients;
        •  the soil residue can support a wide variety of plant species; and
        •  plant species tested could not remobilize transformed compounds
           existing in the treated soil or uptake any residual TNT, RDX, or
           intermediate compounds.
                                Table 8.6
          Composting System Remediation Costs* for Munitions-
               Contaminated Soils ($/ton contaminated soil)
                 Treatment                          $150.00
                                                  •
                 Analyses:                  I
                         Method 8330                   $7.73
                         Field screening tests              $3.21
                 Treatment/containment tents
                                  .
                         (>60,000 ft2)                  $34.00
               .  Miscellaneous                         $61.56
                 Total                              S256.50

 Treatment costs are aggregated costs which include contaminated soil excavation and materials handling
 (screening, mixing, pile preparation, etc.) costs            |
                                   8.61
-------
I'll; ,:j''.  W, 1'iF!.."  •::,','      ..  ,  I';1";
   This case history demonstrates the use of laboratory- and pilot-studies and
 the design of a remedial system which syngeristicaliy integrated several
 bioremediation technologies. The remedial system is unique in that the en-
 tire subsurface system had to be installed during construction of a retail fa-
 cility with limited future access to the subsurface, the system included the
 Raymond method, air sparging, bioventing, LNAPL recovery, and a soil cell.
 A programmable logic control system provided pulsed flow of the air
 sparging system and coordination of bioventing and air sparging within indi-
 vidual areas of the site.

 8.6.1 Site Description
   The subject property was a former paint factory that was investigated as
 part of a property transfer agreement. At one time, the property contained 16
 underground storage tanks (USTs) used to store gasoline, heating oil, and
 varsol (a gasoline-like solvent). Many of these tanks had been removed in
 prior years as well as during the months immediately preceding the property
 transaction. All of the UST locations were identified as areas of environ-
 mental concern and were included as part of the initial site characterization.
   The investigation revealed petroleum hydrocarbon compounds (PHCs) in
 the subsurface. Light nonaqueous-phase liquids (LNAPLs), varsol, gasoline,
 and fuel oil were found on-site as liquicf-pnase PHCs, dissolved-phase
 PHCs, and adsorbed-phase PHCs in both saturated and unsaturated  soils.
 Fuel oil was largely present in the unsaturated soils. Varsol  represented the
 majority of the PHCs present. (See Figure 8.24).
                                              ,1
 8.6.1.1 Geology and Hydrogeology
   The site consisted of unconsolidated soijs from elevation 32.3 m  (106 ft)
 City of Philadelphia relative site datum (RSD) to approximately 28.9 m (95
 ft) RSD and was comprised of varying fill material. This material is of vary-
 ing permeability, but is generally characterized as being of low permeability
 and is underlain by gray to tan silts and clay with thin interbedded fine sand
 lenses. This silt and clay extends from the base of the fill material to varying
 elevations ranging from 28.4 to 27.4 m (93 to 90 ft) RSD. the silt and clay
 sediments are underlain by brown sands with little silts and gravel. This
 sand tends to grade into a weathered schist mix and, in some locations, is
interbedded with sand. The schist becomes less weathered and more compe-
        1 ""          .     '...  "" •	-" 	" '   .'	•:	   !':•(,    	       	 ••
tent with depth.  The weathered bedrock soil types and sand horizon are
                                                                 8.62
-------
                                                                   Chapter 8
                                  Figure 8.24
  Building Locations and Petroleum Hydrocarbon Compound Boundaries
                                (not to scale)
                                                                    V
 Existing Buildings
   (Fanner Paint
  Manufacturing)
mmimmmm Adsorbed-Phase PHC
m mm mm Dissolved-Phase PHC
•——mm Liquid-Phase PHC

Total Area of Impact = -150,000 standard ft
                                    8.63
-------
                                 Case Studies
                        f :  "I  •,       "   '•                '      ,:   '  ,   "",  •    III','".    '  •    • ''    ;-  | .'•''' V
                                 considered part of the unconsolidated zone, and groundwater appears to exist
                                 in this zone under water table conditions. Although the interface between
                                 the unconsolidated material and bedrock occurs at various elevations beneath
                                 the site, the interface was assumed to exist uniformly at 9.2 m (30 ft) below
                                 ground surface or at 23.2 m (76 ft) RSD for purposes of the hydrogeologic
                                 analysis.
                                    The shallow groundwater beneath the site exists under water table condi-
                                 tions within the unconsolidated silt and sandy sediments and into the weath-
                                 ered bedrock schist. The monitoring weir measurement and liquid level
                                 gauging data indicated a groundwater flow direction to the south with an
                                 estimated average groundwater gradient /; of 0.004.
                                    The aquifer was tested to assess the feasibility of in situ remediation.
                                 Testing included two aquifer pump tests, seven step-drawdown tests, and 20
                                 slug tests. Based on evaluation of data from these tests, the following aqui-
                                 fer characteristics were identified:
•'"III •   t  i,i';     '            ,           .   ! "•                    •  • " ,    "'       ,,„ •! i.. | '» '  „  !i«      "•    '»' i 1 ..'i ,"1 i M"1
                                         •  fransmissivity (T) = 134 m2/day (1,444 ft2/day); and
                                         •  Hydraulic Conductivity (K) = 29 m/day (96 ft/day).
                                    Capture zones used in the final design with each recovery well pumping
                                 at 15.1 Lpm (4 gpm) were as follows:
                                         •  Crossgradient—  15.2 m (50 ft); and
                                         •  Downgradient— 12.2m (40 ft).
                                    Average groundwater flow rates were calculated as follows based on the
                                 hydraulic conductivity value presented above:

                                 Vp(pore velocity) = Ki / ne  = 29(0.004) / 0.20 = 0.58m / day (1.93ft / day)(8.3)
-•   • :  ;      :    ,      •:•  ..,'  '.      • ;       '           ,:  •;    •  •'• • "'   ;•  '   ". •.!••             •  '•: ••:•••{ » 	
                                 where:
                                          ne   =   the soil porosity (unhless) 0.20 at this rate.
                                    The flow of groundwater is affected by hydrodynamic dispersion and the
                                 flow of dissolved-phase contaminants in the groundwater is affected by ad-
                                 sorption to soil as well as other attenuation factors (biological degradation
                                 and redox reactions).  As an initial assessment of contaminant transport, the
                                 effective velocity for contaminant flow can be estimated. For this site,
                                 where it is assumed that soil adsorption is the only attenuation mechanism,
                                 the migration rate of benzene was calculated as follows:
                                                                                i
                                 Ve (effective velocity of the contaminant) = Vp / R = 0.58 / 2.98
                                               = 0.2 m/day (0.65 ft/day)

	s   ,.         i..»          vs	: <*           .      . •  i; .     "          ,^8.64	
-------
                                                           Chapter 8
where:
         R  =   the retardation coefficient = 2.98 for benzene at this rate.
   The heterogeneous nature of the subsurface stratigraphy, both laterally and
vertically, and the corresponding variation in hydraulic conductivity of the vari-
ous materials were considered during design and analysis of the aquifer tests
and the subsequent design of remedial system components. Design of specific
remediation system components using conservative input values allowed for
operation of the remediation system at or above design specifications.

8.6.1.2  Contaminant Profile         '
                         •
   Soils, PHCs were detected in the unsaturated and saturated soils beneath
the site.  These PHCs were identified as gasoline, varsol, and small quanti-
ties of fuel oil using modified US EPA Method 8015 (US EPA 1986d) and
GC/MS analyses.  The VOC, benzene, toluene, ethylbenzene, and xylenes
(BTEX), naphthalene, and 1,2,4-trimethylbenzene, were identified as spe-
cific PHCs of concern relative to regulatory remediation guidelines.
   The horizontal and vertical extent of PHCs was assessed during two
phases of site characterization. The PHCs extended vertically from ground
surface (elevation of about 32.3 m [106 ft] RSD), through the water table
(about 27.7 m [91 ft] RSD), and to the unconsolidated zone and bedrock
interface (about 23.2 m [76 ft] RSD). The concentrations of PHCs detected
at the unconsolidated bedrock interface were below remediation guidelines.
Based on the known lateral extent of TPH and total VOC soil contamination,
the total lateral extent of the soil adsorbed-pfaase PHCs covered approxi-
mately 7,157 m2 (77,000 ft2).
   Groundwater. Groundwater sampling was conducted via a series of
monitoring wells that were installed during Phase 1 (MW-1 through -8) and
Phase 2 (MW-9 through -17) site characterization activities.  Results indi-
cated PHCs in groundwater at concentrations ranging from 0.16 to 430 mg/
L. These PHCs were identified as gasoline, varsol, and lower concentrations
of fuel oil.  Results also identified specific chemicals including BTEX, naph-
thalene, and 1,2,4-trimethylbenzene at concentrations ranging from 0.232 to
670 mg/L.  BTEX compounds indicate a light petroleum distillate fraction
representative of gasoline, and in this case, also of varsol.
  The highest concentrations of dissolved-phase PHCs  and BTEX were
detected near the eastern corner of the property. Only this area consis-
tently exhibited of liquid-phase PHCs which were identified as varsol
constituents using a GC/MS "fingerprint" analysis as compared to vari-
ous petroleum standards.
                                8.65
-------
                                           	'	''	p	;	;"
                                                                  I 	,
                                                                  I .;•
Case Studies
   Monitoring wells installed during Phase 2 site characterization were used
to further assess the groundwater contaminant plume, especially off-site
south and west. Results indicated that the| higHest contaminant concentra-
tions in groundwater were located in the southeast corner of the property. So
this area became the target of remedial efforts.
   The extent to which the plume extended beyond the property boundary in
the south and east directions was not delineated (but is currently under inves-
tigation). The lateral extent of the on-site dissotved-phase hydrocarbon
plume was approximately 13,940 m2 (150,000 ft2).
   The presence of liquid-phase PHCs suggested that any remedial action
should incorporate free-phase product recovery. The volatility of varsol and
gasoline and the biodegradability of the PHCs present in the subsurface
suggested that bioremediation, air sparging, and soil vapor extraction/
bioventing were also appropriate remedial technologies for this site.
  "         '             '          ,•'';'  ?'- .'   ! !'  |.i!:    !,,  ':'.'  ,   :    •  I","I
8.6.2 Initial Remedial Goals
                                   	'	  i              ,    	i -	
   Due to the potential extent of the contamination and the planned use of
the property as a retail strip mall, the property buyer was concerned about
the time required to remediate the site in conjunction with plans to remove
existing buildings and develop the property. To meet financial commitments
as well as lease the property, the buyer required a defined course of action
for site remediation.
   The Commonwealth of Pennsylvania Department of Environmental Pro-
tection (PADEP) has oversight responsibility for tne site. The PHCs in the
subsurface were not in compliance with the Pennsylvania Clean Streams
Act. PADEP agreed that the UST corrective action regulations (Corrective
Action Process for Owners and Operators of Storage Tanks and Storage
Tank Facilities and Other Responsible Parties, PA - Title 25, Chapter 245,
Subchapter D, 245.301) would be used as a procedural guideline to bring the
site into compliance.  According to these  laws and regulations, the respon-
sible party conducted a site characterization which was followed by a reme-
dial alternative analysis (feasibility study) since soils and groundwater were
found to contain chemical substances in excess of regulatory limits. All
remedial action plans had to be reviewed  and  approved by PADEP.
   Based on development plans for the site, a  significant portion of the con:
laminated soil and groundwater was directly beneath the area that was to be
used as a concrete pad for one of the developer's future tenants. Due to the
extent of the contamination and the lime frame established by the  developer/
tenant lease contract, a remedial action plan had to be put into effect quickly.

                                 8.66
-------
                                                             Chapter 8
Part of the plan was to remove a limited ambunt of contaminated soil, but the
majority of the remediation was to be accomplished via an in situ system.
   Due to time limitations, the remediation plan had to be designed and ap-
proved in conjunction with plans by the developer. Buildings and foundations,
as well as soils, were removed to allow for the installation of remediation
system's pipes and tubing. To ensure a controlled operation, the same contrac-
tor was chosen to perform soil and macadam removal as well as trenching for
the subsurface remediation system. Since the area to be remediated was di-
rectly under a planned concrete pad, scheduling of work had to be coordinated
with the future tenant's contractor. To remediate the area under the planned pad
and monitor the impact of the remediation process, air sparging points, water
reinjection points, vapor monitoring points, and monitoring wells had to be
placed within the footprint of the tenant's store.
   Within 20 weeks, the system piping had been laid and covered. A 40-mil
geomembrane was laid on top of the area to be remediated. The contractor
was made aware of critical areas, especially those near sewer lines and utility
installation points.
   The imminent conversion of the impacted area to a retail facility required
that any long-term remedy be unobtrusive and require minimum activity
across much of the impacted area. Plus, upgrading or modifying the
remediation system after installation would not be possible.

8.6.3  Design  Approach

8.6,3.1  Identification of Target Elements
   Prior to identification and selection of remedial technology alternatives,
target elements were established. A target element was an operational zone
within which a single remediation alternative could be applied to achieve
remediation.  The establishment of target elements was based on macro-
scopic properties, such as overall matrix permeability, relative humidity/
moisture content, accessibility, depth below ground surface, and contami-
nant state and concentration.
                                        I
   Three target elements for the former painit manufacturing site were estab-
lished, based on  the following site properties!:
        • liquid-phase PHCs floating on or residing near the water table;
        • PHCs above the water table adsorbed on soil particles,  present as
          liquid product retained in soil interstitial spaces, present as soil
          vapor, and dissolved in soil moisture; and
                                 8.67
-------
                                    Case Studies

                                            • PHCs below the water table dissolved in groundwater, adsorbed
                                              onto soils below the ; water table, arid present as liquid product
                                              retained in soil interstitial spaces below the water table.
                                       The feasibility study focused on identifying and evaluating appropriate
                                    and applicable technologies, and eliminating inappropriate remedial tech-
                                    nologies from further consideration. Remediation alternatives were gener-
                                    ated by combining retained technologies for target elements into integrated
                                    systems. The alternatives were then subjected to screening, detailed evalua-
                                    tion, and selection for final remedial alternative identification.
  : '.,    '          ;   ,  ,   •  .  ' ,„      •     '   , .....  '     ,    , ,,  ;  . •. • ;;;    I :. ;!"!! ,  .  ,   I .'.•>'.<  .'"'' !"">.    •  : '  \ •;*• \>
                                    8.6.3.2  Remedial Objectives
                                                   ......              '  •    '   I          "        •"  :
                                       The remedial objectives for this project were:
  I*-.:   ; 1  !'      " '              !i          ..  ...... ' , ,    if,   ' .  •,,  '  ;',, ' v, ,1!'! - .f;; . , ••( i!  |. ;„ " '..   If   ' , '. .;  ., ,.."i I .„,
                                            • to reduce the potential risk to human health and the environment.
                                              This included the following minimum goals:
      • ;        .-• :          •        '          '         "              "','• \ •,''"' 1'-' '     •"":":,;! ...... '•   , .'  ' '..  .'! •   :
                                                 •  eliminating recoverable liquid-phase PHCs;
......     ,        '                      .,       ", .  ,    . ' ....... "      "  ',..   .     . \ ..... |, ; . " ;  " ', '..  |... ,,  ; .;' .  , ...... , '.   . ,;;' , ,;, ;, ;,, \ , ,
                                                 •  reducing hydrocarbon concentrations within the vadose
                                                    zone by 95%;
                                                 •  reducing soil contaminant levels by 90%;
                                                 •  reducing groundwater contaminant levels by 90%;
                                                 •  negotiating achievable site-specific clean-up criteria accept-
                                                    able to PADEP;
                                            • to prepare the site for redevelopment operations and return the
                                              site  to productive use; and
                                            • to reduce future environmental liability associated with the site.
                                                                                  •i
t  I"!     i  "      III ..... i "        ''"                  ,         '              ' '" ...... *"  '"•   .;"'i'|| i 'M i'l1'!'1 '" "i,1 i '   .1- :    ' .'  ;,;v'' ..... (l^ "iLii1' # ,|
                                    8.6.4 Remedial Design
;  -    ^ ; ...... •     ^         •      ...... •:  " •  • •  '       ' •   '   '•   '  ..'•  • :" - -  ;;:   .;;'Ji;         - ..... _i'.  .'H^'f
                                    8.6.4.1 Technology Selection
                                       Selection of  specific technologies for remediation of this site was dictated
                                    by the properties of the petroleum hydrocarbons, their distribution within the
                                    subsurface, and the planned use of the site.  Special consideration was given
                                    to the planned imminent development of the property, including consider-
                                    ation of the health and safety of pedestrians following site redevelopment.
                                    Any in situ remedial system needed to account for the potential migration of
                                    PHC liquids or  gases. The selected remedial system ."was required to achieve
                                    immediate remediation of shallow soils to accommodate site redevelopment
...........             1|" "              '                   '                         '  '    '   1 ' '      '         -  "  : ::•" ,  ,
                                                     "           '     ' '        "    1        '            .....

  :' '           >              :                              '      '  '"  8.68
-------
                                                             Chapter 8
 construction and remediation of deeper soils in a manner that did not inhibit
 the commercial re-use of the site.
   Initially, a broad range of technologies was considered following the stan-
 dard remedial investigation protocol specified by CERCLA. Remedial tech-
 nologies were considered for each target element.  A combination of in situ
 aeration and biodegradation techniques was selected as the final remedial
 option for the hydrocarbon-contaminated soils at the site. Additionally, re-
 covery of liquid-phase hydrocarbons in conjunction with groundwater recov-
 ery was selected because this option was compatible with other remedial
 components and was the fastest and least expensive method of removing a
 significant portion of the hydrocarbon mass at the site.
   Impacted shallow soil, which contained primarily fuel oil, was excavated
 for on-site treatment.  This option was compatible with installation of other
 remedial system components, which required soil excavation anyway, and
 would ensure that the upper several feet of soil were clean prior to construc-
 tion of new buildings.
                                        i
   Excavated soil was treated with a technology compatible with that used in
 the in situ system in aboveground treatment cells constructed on an adjacent
 parcel that remained under the seller's ownership.  Biological soil cells were
 selected for treatment of excavated soil based on: (1) the biodegradability
 and low volatility of most of the hydrocarbon mass present  in these soils; (2)
 the potential to use the same offgas treatment system used for the in situ
 treatment system; (3) the availability and proximity of a large area for con-
 struction of the cells; (4) the benefit of on-site reuse of the treated soil; and
 (5) the relatively low cost of this process.  The soil cells were constructed
 and treated using methods described in Section 4.4.
   The soil and groundwater at depth were largely impacted with varsol,
 which is both volatile and biodegradable, and is only slightly water soluble.
 Based on these properties, remediation through a combination of biodegra-
 dation and physical removal in the vapor phase was the most appropriate
 approach. Bioventing has been used successfully at numerous sites and is
 easily applied in, sandy soils, such as those at this site. The  sandy soils
 within the upper aquifer also made the site conducive to biosparging (air
 sparging to enhance biodegradation as well as physical removal of VOCs).
The use of biosparging with bioventing provides much more air flow
through the highly impacted capillary zone than can be provided by
bioventing alone. Other positive factors for the use of bioventing at this site
were that the area was to be capped and the depth to water was greater than 3
m (10 ft) belowground surface. As a result, the radius of influence of the air
                                 8.69
-------
Case Studies
capture system would be large, making bioventing a cost-effective
remediation option.
   Biodegradation, subsurface aeration, and aquifer performance feasibility
studies were performed. The aeration, bioremediation, and aquifer pilot
study tests were generally designed and performed consistent with the meth-
ods described in Section 5.5. Biodegradation was generally limited by the
supply of electron acceptor (oxygen), as adequate levels of nutrients (nitro-
gen and phosphorous) were found throughout the site. Iron  and magnesium
precipitation was identified as a potential problem, as they were present at
concentrations up to 100 and 37 mg/L in groundwater, respectively. Subsur-
face aeration provided adequate oxygen to support bioremediation via both
ah- sparging from below the water table and through soil vapor extraction
(SVE) in the unsatiirated zone immediately above the water table.
                                           	i

8,6.4.2 System Design
                                             i
   The subsurface aeration system was designed to achieve the following:
        •  oxygen addition to support bioremediation;
        •  direct soil vapor withdrawal for gross hydrocarbon removal; and
        •  vapor-phase process treatment prior to atmospheric discharge.
   The subsurface aeration system is comprised of three subsystems includ-
ing the air injection (AI) system, the SVE system, and the vapor treatment
system. The AI system delivers an aggregate 1.89 «10'2 m3/sec (40 scfm) at
138 kPa (20 psig) to the AI manifold. One hundred thirteen AI points were
installed in the remediation target area, with the points operable through
dedicated manual flow rate control valves and 17 automatically-controlled
valves directed by the central computer for simultaneous control of groups
of six to seven AI points.  Figures 8.25 through 8.27 show a site plan and a
cross-section diagram of the site geology and remedial system installation.
   The SVE system applies an aggregate 3.7 • 10'2 standard  m3/min (78
scfm) at 12 A kPa (50 in. of water) vacuum at 25 SVE points. Vacuum is
dkected to each point through dedicated manual flow rate control and auto-
matic valves. Typically, three SVE points are operated in conjunction with a
set of AI points for complete capture of injected air.
   Vapor treatment included a catalytic incinerator during the initial
high-VOC loading period, and vapor-phase activated carbon during subse-
quentlower VOC recovery periods^ The vapor treatment system handles the
offgas from an air stripper (a wastewater treatment system component).  The
       „.       ,             »|'   "i| ..i  "      	'!'       ,            i I i	
                                 8.70
-------
                                                                           Chapter 8
                                     Figure 8.215
                          Site Plan Remediatipn System
                                    (not to scale)
                 Air Injection Flow


                      IN-1
      Soil Vapor
      Extraction Flow   /
                  OIN-2
>MW18
  New Store
I Monitoring Well    O Water Injection Well       °  Air Injection Point      	   Impermeable Liner
I Recovery Well     A Vapor Monitoring Probe   —	 Vapor E-xtraction Point  A-—A' Cross-Section Location
                                        8.71
-------
Case Studies
                               Figure 8.26
                            Cross-Section A-A1
                              (not to scale)
                     MW-16
• Asphalt B Asphalt MW-7 MW-9













*]
i*
''•
s
'$
••;:
s

V-
•:!'•
1
FiU

Silt



Silty Sand
Silt


Silty Sand
Weathered
Bedrock


" - : 	 ; • • ; ;••
"
-
1
i
2
C__
;.'"»
;''::
'•• **.'
•••:
• •'••
'.•/.
• !
• :
:|
1
|
|
U«i
Fill




Silty Sand




Sand

Z


-s»
'sJ
s
§



^
I
^
Fill




Silt


Sand


Weathered
Bedrock
g
i
>
*x
>C
^
1
i •
3"!
i»
1
i


FiU



Silt
Sand



Weathered
Bedrock

Silty Sand
Weathered
Bedrock

- 105.0


- 100.0


- 95.0

£•
I
- 90.0 &
- 85.0
- 80.0


- 75.0
' I - ' ' 	 ;
phased treatment design is based on a cost-effectiveness analysis indicating a
"break-even" vapor-phase VOC concentration of 375 ppm.
   One concern of air sparging is that the air typically moves through distinct
channels. As a result, removal of dissolved-phase constituents or provision of
oxygen for biodegradation requires movement of the constituents or oxygen by
advection, dispersion, or diffusion. These processes can be relatively slow and
can control the rate of remediation. Both pulsed-air injection and recovery and
reinjection of groundwater increases the rate of advection and thus the rate of
                                  ••::;.„ ti
                                  8.72
                                       1  -i
-------
                                                                                       Figure 8.27
                                                                 Subsurface Aeration System Components
                                                                                        Finish Grade
00
          Water
          Extraction
          ur.1i
                      6 in. Schedule-40 PVC
                                                                                        40-maeHDPE
                                                        Manifold Pipe-
                                                                                    Excavated Backfill
2 in. Schedule-40 PVC


       10-mile Polyethylene »
                                                                                                                                               Native Fill

/
H— 	 r
4in.by2in.Tee\




N
                                                                                                             '\		  ._
                                                                                                             -     in, wasnea sione
                                             Air Injection Point
                                             (2 in. Schedule-40 PVC, 0.040 in. Slot)
                      6 in. ScheduIe-W SS. 0.040 in. Slot
                                                                                                                  Vapor Extraction Pipe
                                                                                                                 '(4m. Schedule-40 PVC 0.040 in. Slot)

                                                                                                                 . Dewatering Pipe
                                                                                                                  (4 in. Schedule-40 PVC. 0.040 in. Slot)
                                                                                         50ft
                                                                                                                                                                     6 in. Schedule-40 PVC
                                                                               Water

                                                                               Wen"""
                                                                               6 in. Scheduled) SS,
                                                                               0.040 in. Slot
o
of
V.
(D
00
-------
                                    Case Studies
''I	 III1"!1' "
Jill, ''"<
                           	Jill'	I  ,        '	.	—.	"	—	—
                            ''""ill  .'•',,:'    :         .        .: ''  •'    '. >i	!'.'
                           	!* «!       i       ,'       .  <      ',     .,     i •  ' . Ai'iv
                       'Hi   ,                                 * 1*1 i    • ,' ,    '   '  f     ,    '  , '' I  '"
                      remediation. The relatively permeable soils within the aquifer allow for ad-
                      equate grbuhdwater recovery rates and consequently, groundwater flow rates
                      that will significantly enhance advection between air sparging channels. Be-
                      cause of me need to recover liquid-phase PHCs, groundwater recovery provides
                      the benefit of enhanced tiquid-phase PEE recovery, recovery of a significant
                      fraction of the mass for surface treatment, enhancement of air sparging, and
                      with reinjection, the opportunity to add nutrients to ensure that nutlient levels
                      are not rate limiting within the saturated zone.
                        The planned removal of soil for on-site treatment allowed horizontal va-
                      por recovery lines to be used to provide an additional method to reinject
                      water and nutrients, if necessary. The horizontal lines provide more efficient
                      vapor recovery over the relatively narrow vertical interval where
                      hydrocarbon-impacted soil remained after partial excavation.
                        The water management system was designed to achieve the  following:

                              • hydraulic control of theshallow aquifer downstream of the
                                remediation target zone via groundwater recovery through a se-
i,       i •   .,.'!.!    .        '          , '   • •, i. n*?    mi	' "n,'i "'' ,1'irai .„   ,. v i i " ii,        • •        • • ,"i i "'Hi!
                                riesof recovery wells (RWs);
                                       	            	i.,         	
                              • direct liquid-phase PHC recovery at RWs where present;
                              • groundwater recovery (dewatering) in the vicinity:bf SVE points in
                                dewatering wells (DWs) to prevent flooding of the SVE points;
           .'. ,,   •;,!!;  i .'                 	" F.    ,   '•.	,	:	v	• ,„' ,: i !•' ',!!;""' c 	'-, ,  ,n, , ,    .••;•  ••,,.,• HIE  „	•
                              • process water treatment prior to discharge to the city sanitary
                                sewer or on-site reinjection; and
                                            '!	» '  '...'i  ." •:  !.""":.	    :'".!'   .' :  '."       •   '.'  ='  •.'"
                              • water injection and distribution of nutrients required for support
                                of the saturated zone bioremediation system.
                          ,  •;•    ,  :   ;:": - ,:'":    '  '• ;  '.';•.",,'•• ••?-.'•;(  :  i
-------
                                                            Chapter 8
 concentrations, and the metered injection of amended water to a series of
 seven injection wells. Based on an aggregate long-term flow rate of 57 Lpm
 (15 gpm), the groundwater turnover period is approximately 1 year.
   Materials management provided by the subsurface aeration and water
 management systems established the basis for the final remedial element: a
 combination of limited area air sparging and intrinsic remediation.  Hydro-
 carbon mineralization will occur over a 2-year period at design aeration and
 oxygen delivery .flow rates assuring a 20% utilization of injected oxygen.
 This time frame may be shortened based on the improved distribution cre-
 ated by the continuous enhanced groundwater flow through the target zone
 and intermittent air injection, allowing greater time for distribution through
 dispersion and diffusion.
   The selected in situ remedial system is highly integrated, consisting of
 groundwater recovery with aboveground treatment, injection of treated and
 nutrient-amended groundwater, liquid-phase PHC recovery, biosparging,
 bioventing, and aboveground treatment of offgasses. The remediation com-
 ponents were designed to be either underground or located in a treatment
 facility adjacent to the tenant's building. Electronic controls located in the
 treatment facility along with other design features permit operation with
 minimal need to enter the tenant's building;.
                                  '
 8.6.4.3 Automatic Control System  j
   The automated control system is driven by a programmable logic control-
 ler (PLC) which provides standard relay-type operation of water level con-
 trollers and flow control valves.  Air injection, vapor extraction, and dewater-
 ing system operation are controlled through the PLC timer.
   The PLC is used to operate the remediation system controls and the sole-
 noid valve primary control unit.  The PLC system is comprised of hardware
 (physical equipment) and software (programmable features). PLC hardware
 includes components required for electricall connections to the power  supply,
 connection to and control of valve solenoids, and the user interface. All
 hardware is housed in the water system sub-panel and is adjusted or main-
 tained only by specialized personnel trained in the use of the PLC. PLC
 software includes components that are preset by the supplier and a compo-
 nent that allows field operator adjustment.  Preset components include those
 to establish basic PLC configuration, parameters that may be changed by the
field operator, and field operator interface format. Of these, all except those
pertaining to the field operator interface can be adjusted or maintained by
specialized personnel trained in the use of the PLC.
                                8.75
-------
Case Studies
   The preset software configuration is a computer program that allows the
field operator to change the solenoid valve operation schedule by download-
ing an array template provided by a user-friendly DOS support program. In
general, the DOS program allows: (1) selection of a desired number of "se-
quence slots," (2) assignment of a time duration for each of the sequence
slots, and (3) assignment of one or multiple valves that are to be opened for
each sequence slot.
   Through the use of the PLC user programs, the system is configured so
that each set of air injection valves opens in sequence, and the appropriate
vapor extraction valves open simultaneously to allow recovery of all injected
gas in the area of application. The PLC system also allows pulsing of the
biosparging system. Further, dewatering points located proximally to oper-
ating vapor extraction points may be operated synchronously with the air
sparging/SVE system.
   Valve opening configurations may by altered either during the initial con-
figuration or by modifying an existing configuration as indicated below:
        •  me number of sequence slots created (limited to a maximum of 17);
        •  the time duration for each sequence slot in minutes (limited to a
           maximum of 1,440);
        •  the AI set valves assigned to a sequence slot including one or
           multiple valves (values from 01 through 17);
        '»    ' "'i'   , ,i !•„ in < ,"„„ -:„  ':	 ,":;:'",   „ , •„i."!•,.', mi; ii ; *   .    •   •    '• ',  „ ^ j.itiF:1!11;,
        •;,  the SVE-point valves  assigned to a sequence slot including one
           or multiple valves (values from 01 through 25); and
        •  the DW-point valves assigned to a sequence slot including one or
           multiple valves (values from 01 through 25).
8.6.5  Implementation
                 •'       •      •        '     • 'iiili •, ii ,' "   •  '  I"     /   '""  i iif, ;
  Expedited remedial construction was implemented through the fol-
lowing steps:
        1. Initial excavation of shallow soils exhibiting elevated contami-
          nant levels permitted site redevelopment construction to proceed
          immediately. Excavated soils were stockpiled on an adjoining
          parcel for treatment in a soil cell.
        2. The site redevelopment contractor was selected as the remedial
          contractor to allow seamless transition between remedial and
          redevelopment civil construction.  Simultaneous and optimal task
          sequencing was achievable without contractor conflicts.

                  •  '  '   	   ;,.    8.76'";""..  .'i.:.^:'  ;   '.,    ''"   	1;,,
                                                                    ,
                                                                -	 	I JT "
                                                                . , .. 1 	
-------
                                                             Chapter 8
           Construction tasks, such as excavation of poor structural soils
           and replacement with acceptable fill, were conducted in conjunc-
           tion with the excavation of contaminated shallow soils, as well as
           the excavation and construction of subsurface treatment system
           components. Developer goals and environmental work plan
           goals were achieved concurrently.
         3. Preliminary design and construction of subsurface components
           were conducted prior to the completion of final mechanical plant
           design.  This was acceptable based on the known applicability of
           the remedial techniques selected. Documentation and designs
           for groundwater recovery systems, subsurface aeration systems,
           and impermeable liner systems are standard in the industry. Fi-
           nal blower, pump, treatment, and instrumentation system selec-
           tion and integrated engineering were completed during expe-
           dited subsurface system construction. The availability of
           off-the-shelf remediation equipment from reputable vendors for
           processes within previously-documented application ranges
           ensured that an appropriately-sized and constructed mechanical
           plant could be completed.
   A critical aspect of the construction effort was the remote access  system,
 which was needed due to operational cons traints imposed by the use of the
 commercial property overlying the target 2:ones.  System construction incor-
 porated the necessary plumbing, including 125 1.3-cm (1/2-in.) ID HDPE
 tubes leading from each air injection point and vapor monitoring point to the
 treatment facility, 74 5-cm (2-in.) diameter PVC pipes leading from water
 and vapor points to the treatment facility, as well as electrical and pneumatic
 conduit for the groundwater recovery pumps. All tube, pipe, and conduit
 installations were completed in the field arid demarcated and stubbed while
 treatment facility design was completed and construction was undertaken.
 Incorporation into the mechanical plant then proceeded, providing point
 access from the central location, with limited field access available for water
 system components only.
   Individual air injection, vapor extraction, groundwater recovery, and in-
jection well performance tests were conducted during startup. Manual flow
 control valves were adjusted to achieve the desired flow rates, and secondary
 automatic valves were operated by the system computer based on a variable
 operator input program.
   Operation of the in situ remediation system was divided into three phases.
During Phase I of operation, the following remedial actions were implemented:
                                 8.77
-------
lie IIIF	i; i, •	!'n	;iii	•.   • ''ill	itf i>	iiiWiiii'Biiii:1 ''liiiiri:11; ii'i'iii :.'\	• " ••„	 	!•  	   '   .  ' 'ffi	• >  " ",. "IB	't*	'|l|;l '•^••.'•tsa1	:• ("ini'iKini:'1.1"'!*'! i"~ tiirrM •	tn.ii.-t	v>m,i	EIIK	v JK i
                                 Case Studies
                                                                                 !
                                         • liquid-phase PHC recovery with <0.5 m (1 ft) drawdown at re-
                                           covery points;
                                         • SVE operation at SVE-point triads in which VOC concentrations
                                           measured by field PID measurements were >100 ppmv; and
                                         • clean water injection through injection wells, as much as possible
                                           using: (1) liquid-phase granular activated carbon treated process
                                           water, and (2) tap water.
                                 Phase I operation will continue until liquid-phase PHC recovery decreases
                                 by at least 50%.
                                    During Phase n operation, the following remedial actions will be imple-
                                 mented:
                                         • groundwater recovery will be increased through all RWs. The
                                           pumping drawdown will be progressively increased in 1-m (3-ft)
                                           increments at a minimum of 48-hour durations followed by incre-
                                           mental drawdown increases when the liquid-phase PHC recovery
                                           has decreased sufficiently;
                                         * injection of amended water into the injection wells at up to de-
                                           sign flow rates; and
                                         . operation of the subsurface aeration system (i.e., SVE system
                                           operation only) as a continuation of the Phase I operation.
                                 This operational period will continue until the VOC recovery rate decreases
                                 by approximately 80%.
                                    During Phase III operation, the following remedial actions will be
                                 implemented:
                                         • groundwater recovery system will be continued as in Phase H
                                           operations;
                                         • water injection system operation shall be a continued as in Phase
                                           n operations; and
                                         • initiation of full-scale subsurface aeration including SVE and air
                                           injection in accordance with engineering specifications  for auto-
                                           matic SVE and AI point activation.
                            , „           ,     '  n   " 	  "'  !      , ,i!  ,  .« i"1!i  " "iiijii 'iii1' 'mi,   j, ,r - i|, • si '   ,   ,• •' : „"   '"'" .   ' • : ii ' ""i .i',111:!1    , i
                                 This operational period shall continue until the following monitoring results
                                 are obtained:
                                         •  detectable liquid-phase PHC is not observed in monitoring or
                                            other wells', and
                                                                   8.78
-------
                                                            : Chapter 8
        •  groundwater quality compliance results indicate that hydrocarbon
           concentrations in groundwater have decreased to below estab-
           lished remediation standards.
                                       i

8.6.6 Operation Results
   Since startup, no LPH has been detected in site wells, and LPH has not
been recovered in recovery wells.  Hydrogeologic evaluations indicate that
the material is most likely present in its original locale, but has become
adsorbed or trapped within soil interstitial areas.
   Initial testing and operation of the SVE system indicates that operation is
within design parameters. The system has been recovering VOCs at a rate
somewhat below its design rate. Therefore, conversion of treatment opera-
tion from the catalytic incinerator (used during high-VOC loading) to
vapor-phase activated carbon (used during low-VOC loading) occurred
within the first several months of operatiom.
   Preliminary testing of the air injection system indicates that operational
performance deterioration of approximately 25% is to be expected. This
deterioration was adequately compensated for by performance allowances in
design and conservative estimates of performance based on pilot-test results.
   The first round of quarterly groundwater monitoring data were collected
approximately 3 months after remedial system startup. Results have not
been reported to date, but will be used during final system optimization..
                                       !
8.6.7  Conclusions
   The environmental project undertaken at the subject site was a coopera-
tive effort to meet the needs of all interested parties. This cooperative cli-
mate enabled the environmental specialists and constructors to create an
effective and practical remediation system within the broader scope of pro-
viding a usable commercial property in an expedited fashion.
   This project also showed the value of conducting treatability and pilot
tests, combined with a somewhat conservative design to allow system instal-
lation at a site where access subsequent to construction would be limited.
Additionally, the design incorporated synergistic technologies that are inte-
grated to improve performance of one technology based on the specific de-
sign of another.
                                 8.79
-------
Case Studies
8.7  Intrinsic Remediation of a
Hydrocarbon-Contaminatecl Aquifer
    "•  :'               '   ' '   '.  •''",, i'1' ' 's  '' :!" !'!'"   '; 'Ji!'?: '•'* " ''!   " •; "M"
   This case study describes the application of intrinsic remediation for fate
and transport assessment and plume management at Site 13/26 at Eielson Air
Force Base (AFB), Alaska, which was contaminated with fuel hydrocarbons.
This case study was developed from a 2-year field project conducted from
1993 to 1995 by the Utah Water Research Laboratory at Utah Stale Univer-
sity and funded through the U.S. Air Force Center for Environmental Excel-
lence, Brooks AFB, San Antonio, Texas (Dupont et al.1996). Characteris-
tics of the intrinsic remediation methodology described in Section 5.3 of this
monograph are highlighted as they apply to site assessment, contaminant
fate and transport evaluation, and long-term modeling with and without
implementation of source removal to develop a recommended remediation
strategy for the site.
                                                            'i i1!1,"!1!.. I fl"!'1!!!! " .i!lr '  ;!' ,
8.7.1 Site Description
   Eielson AFB is locatec! in the Tanana River Valley in Central Alaska, ap-
proximately 200 km (124 mi) south of the Arctic Circle.' Most of the base is
constructed on fill material underlain by an unconfined aquifer consisting of
60 to 90 m (200 to 300 ft) of alluvial sands and gravels overlying a
low-permeability bedrock formation (USAF 1994). The aquifer system
below the base is bounded to the northeast by the Yukon-Tanana uplands and
is approximately 70 to 80 km (43 to  50 mi) wide in the area of the base
(CH2M Hill 1982). The direction of groundwater flow throughout the base
is generally to the north, with groundwater encountered at 2.5 to 3.5 m (8 to
12 ft) below ground surface at various times of the year.
   The actual source(s) of contamination at Site 13/26 and the exact mass of
contaminant releasecl"to the environment are unknown. The release of prod-
uct from several JP-4 and diesel fuel storage tanks near the southeast end of
the main taxi-way and fuel bladder filling operations are believed to have
contributed to groundwater contamination in this portion of the site. Leak-
age of a large aboveground JP-4 storage tank and its associated piping is also
known to have occurred.
   Figure 8.28 is a conceptual model of pure product, residual saturation,
and dissolved hydrocarbon plume distribution throughout Site 13/26 based
on groundwater and free product samples collected from November 1993 to
July 1995. It is noted that it was originally thought that one large (>600-m
                               8.80
-------
                                                           Chapter 8
[>l,875-ft] long) hydrocarbon contaminant plume existed at the site moving
north from the area near Tank 300. This conceptual model was based on 15
relatively widely-spaced, large-diameter groundwater monitoring wells. As
indicated in Section 5.3, finely-spaced monitoring points are essential if the
boundaries of contaminant plumes are to be accurately delineated in the
intrinsic remediation plume management approach. This guidance is high-
lighted in Figure 8.28. Based on the improved site characterization possible
from the 45 closely-spaced monitoring points placed throughout Site 13/26,
four distinct contaminant plumes were identified, each much smaller than
the originally-defined plume. The length of groundwater plumes has a sig-
nificant impact on the outcome of an intrinsic remediation evaluation in
terms of verifying degradation rates and in modeling long-term plume be-
havior; it is critical to properly identify unique contaminant plumes before
further evaluation.

8.7.2  Implementation of Intrinsic Remediation Plume
Management Approach
   The intrinsic remediation plume management approach focuses on evalu-
ation of the potential or real risk posed by  contamination at a given site.
This evaluation is made using the protocol presented in Section 5.3 and in-
volves the following:
        • determining whether the plume is stable under existing site con-
          ditions;
        • verifying that contaminant attenuation is mediated through bio-
          logical action;
        • quantifying the contaminant degradation rates taking place under
          site conditions;
        •  evaluating the long-term behavior of the plume with and without
          active source removal; and
        • making decisions regarding implementation of long-term moni-
          toring and/or source removal based on: (1) regulatory and public
          acceptance of intrinsic remediation, and (2) the technical feasibil-
          ity and cost-effectiveness of source removal actions at the site.
This general approach was implemented at Site 13/26.  Details of the
intrinsic  remediation management efforts at this site are given in the
following sections.
                                8.81
-------
Case Studies

    I
    d>
    o

    g
    C/3
    o
    I

 co 15
 
-------
                                                           Chapter 8
8.7.2.1  Assessment of Steady-State Conditions
   Steady-state conditions were assessed at Site 13/26 by inspection of
plume centerline concentrations over time and analysis of integrated plume
mass data for the site. Steady-state conditions were evident from centerline
concentration data as indicated by TPH results shown in Figure 8.29.  Center
of Mass data also indicated a stable, continuous source of hydrocarbon con-
tamination at Site 13/26.

8.7.2.2 Estimation of Contaminant Degradation Rates
   Biological contaminant removal at Site 13/26 was verified through an
analysis of the concentration distribution of nondegradable plume-resident
tracer compounds (PRTs) in the plume relative to that of the more degrad-
able BTEX components. As indicated  in Section 5.3, accelerated loss of
B1EX compounds relative to the PRTs provides evidence of biodegradation
within a contaminant plume. Figure 8.30 shows the normalized concentra-
tion profile for BTEX and four tracer compounds (1,2,3-trimethylbenzene,
1,2,4-trimethylbenzene, 1,3,5-trimethylbenzene, and 2,4-dimethylpentane)
that were identified throughout the plume at Site 13/26. As indicated in
Figure 8.30, as the plume moves downgradient away from the source area at
TP19 toward PS05 (see Figure 8.28 for sampling point locations), even the
groundwater in equilibrium with residual-phase material becomes signifi-
cantly depleted in BTEX components and is highly enriched in the less de-
gradable tracer compounds, confirming a biologically-mediated pathway for
BTEX removal.
   Due to the extent of groundwater data available, the estimation of con-
taminant degradation rates for Site  13/26 was possible using the Domenico
(1987) model mentioned in Section 5.3 and described in detail by Gorder et
al.(1996).
   A distinct free product phase was observed in the source area at Site 13/
26 at various times throughout the study. The actual extent of mobile,
free-phase liquid was dependent on groundwater depths; product was ob-
served when the groundwater table was low, but not when high groundwater
levels submerged and occluded product below the groundwater table.
   Figure 8.28 shows residual- and dissolved-phase contamination identified
from groundwater samples collected from the site in July 1995. The distri-
bution of free-phase product and residual saturation throughout a. site must
be delineated to identify the point at which a true dissolved-phase plume
begins because fate and transport models provide estimates for degradation
rates only in the dissolved plume. At Sites 13/26, the edge of residual-phase
                                8.83
-------
i;
i>
ji

11
     00

     8
                                                                      Figure 8.29
                                            Centerline GroundwaterTPH Concentration Data Over Time
                                         100
                                                     200
                                                                 300          400          500

                                                                   Distance Downgradient of tp22 (m)
B614*
    600
      ,sp08
      Lsp08-
700         800
                                               o
                                               Q
                                               in
                                               (D
                                                                                                                                                  Q.
           « November 1993
           O May 1994
           • September 1994
           a July 1995

           Source: UWRL1997
-------
                                                                 Figure 8.30
                       Compound Concentration Data for BTEX and PRTs Normalized to Values in TP19—July, 1995
                                                          % of Initial % TPH Normalized to Well TP19
                         500%
                         450%
                         400%
GO
00
cn
                                     TP19
                                                 PS01
PS07
PP02
PS05
Groundwater movement is from left to right in this figure.
•B Benzene   I  I p-Xylene   H 1,2,4-TMB
E3 Toluene    CH BTEX     E3 1,3,5-TMB
H E-Benzene  C3 1,2,3-TMB  H 2,4-DMP

Source: UWRL1997
                                                                                                                                                o
                                                                                                                                                Q

                                                                                                                                                I
                                                                                                                                                oo
-------
Case Studies
contamination downgradient of the source area was identified based on the
known composition of product from the site and a comparison of individual
compound concentrations measured in the groundwater to those expected in
equilibrium with the product material based on Raoult's Law. The Raoult's
Law concentration (C uilib) is based on. the mole fraction XA, of a specific
contaminant in the product material as indicated in Equation 8.5:
                                                  	  •"  ' I'..'1    ..i. ,  'I1'1	l,'I!!', ,  |. „•'•;•..''

                                                                (8.5)

where:
        S   =   thepure compound solubility  of Compound A in the
                 aqueous phase.
   If the measured concentration of Compound A in the groundwater is
equal to or greater than C uilib  the groundwater can be considered to be in
equilibrium with mobile free product or residual-phase NAPL within the soil
pore space. Modeling of dissolved-phase contaminant migration must begin
at points downgradient from the identified residual material if degradation
rates are to be representative of field conditions observed at the site.
   Once the edge of a plume was identified, PRTs were used in this model-
ing approach to develop "dilution-conected" degradation rates for the reac-
tive BTEX components of the groundwater plume. These tracer compounds
were used to provide flow calibration for the Domenico fate and transport
model.  The model was calibrated by setting the  groundwater velocity,
source configuration, and* simulation time input data to values shown in
Table 8.7. Estimated retardation coefficient values, R, and measured dis-
solved plume source area concentrations, Co, for these compounds were then
used to fit the model to measured centerline concentration values. PRT deg-
radation rates and aquifer dispersion values were adjusted until maximum
model fit (minimum mean square error [M§E]) to measured field data was
achieved.  With the low degradation rates observed for these compounds,
aquifer dispersion properties had the greatest impact on model-fitting results,
facilitating model calibration to field-estimated aquifer dispersion values.
   This calibration effort resulted in the mean dispersion properties listed as
final model calibration values in Table"8.7," which were used to estimate the
degradation rates of the more reactive BTEX components.  BTEX degradation
rates were estimated by minimizing the MSB of the Domenico model fit to
measured plume centerline groundwater data by adjusting their biodegradation
rates.  Final model calibration results for two of me PRTs and benzene are
shown in Figure 8.31, and contaminant degradation rates estimated for all com-
pounds of interest using this modeling approach are listed in Table 8.8.
	       '      	      •'	    '' l     '"   '      '"	

        •.:.     •  '•/•.    .     '•:' 8.86  ;      '; l
-------
                                                              Chapter 8
                               Table 8.7
          Final Calibration Results Using! Plume-Resident Tracers
              Identified Throughout the Plume — July, 1995
Parameter
Longitudinal
Dispersion (m)
Tangential
Dispersion (m)
1,2,3-TMB
1.9
0.095
1,2,4-TMB
1.5 '!
0.075
1,3,5-TMB
1.9
0.095
2.3 DMP
2.1
0.105
Final Model
Calibration
Value
1.85
0.092
Vertical Dispersion (m)       0.001       0.001  ;     0.001       0.001       0.001

Degradation Rate (1/d)       0.00052     0.0002       0.0003       0.0002        *
R                      7.3        11.6        6.1         3.8         **

Groundwater Velocity        0.069       0.069       0.069       0.069       0.069
(m/d)
Simulation Time (d)
Co (mg/L)
Source Width (m)
Source Thickness (m)
4,650
0.17
50
2
4,650
|
038
50
2
4,650
034
so
2
4,650
026
50
2
4,650
**
50
2
  Parameter varied to obtain the minimum MSE value in BTEX calibration.
 * Specific to each compound.
8.7.2.3 Estimation of Source Mass
   The multiple source areas and individual contaminant plumes shown in
Figure 8.28 were delineated based on: (1) free product observed at the site
during field sampling, and (2) dissolved BTEX concentrations above equi-
librium values based on free product composition results. The residual satu-
ration areas included groundwater samples where BTEX levels equal to or
higher than those in equilibrium with pun; product were measured, indicat-
ing that a nondissolved, residual product material remained trapped in the
soil matrix in those areas.  The areas delineated as high-level contamination
areas in Figure 8.28 included sampling locations where BTEX and TPH
levels were below the levels indicative of equilibrium with residual satura-
tion, but that were above regulatory limits. Using these data, the amount of
residual mass at Site 13/26 was estimated assuming a residual saturation, Sr,
of 25% of the pore volume based on the sandy soil texture of the site (Mobil
Oil Corporation 1995). The results of the contaminant mass estimation  are
                                  8.87
-------
 Case Studies
 summarized in Table 8.9 and suggest that more than 760,000 kg (1,672,000
 Ib) of JP-4 equivalent product remained in iffie site soil and groundwater.
           ".   .       ";'  ', "  •  •  r figure	8.31	
            Domenicp Model Calibration to Groundwater Plume
           CenterHne Data for Two PRTsi and Benzene — July; V995
            Simulation Parameters: Degradation Rate - 0.00009/d, R» 3.9, t - 15.8 years

                 ^ •'   300   	 "
                 O """N     (
                 1
-------
                                                              Chapters
                           Table 3.8
        Contaminant Degradation Rates Estimated
            from Model Calibration to Field Data
Compound
Benzene
Toluene
Ethylbenzene
p-Xylene
2,4-Dimethylpentane
1 ,3,5-Trimethylbenzene
1 ,2,4-Trimethylbenzene
1 ,2,3-Trimethylbenzene
. Degradation rate (1/d)
-0.00145
-0.00149
-0.00087
1 -0.00036
-0.00042
I
-0.00012
-0.00015
1
-0.00021
                           Table 8.9
         Estimated Source Area  Mass and Lifetime
Soil Porosity (n)

Residual Saturation, Sr (%n)
                                             ;
Product Density


Residual-Phase Product Area

  Width


  Length

                                    j
  Depth


  Total Volume

   •
  Residual-Phase Product Volume



Residual-Phase Product Mass (g)

Source Lifetime Estimation

  Measured Mass Flux From Source Area


  Estimated Years to Deplete Source Based on Flux Rate
    03

    25

 801 kg/m3
 (49.9 lb/ft3)
    50 in
   (164ft)

   100m
   (328 ft)

    2m
   (6.6 ft)

  10,000 m3
 (352,876ft3)

   949m3
 (33,523 ft3)
(250,681  gal)

 759,758,497
99,492 mg/ft/d
  1,242 g/d

    1,676
                              8.89
-------
                    II I1|L	V :.'. "'	!': " .'•'(! ' lliil* If":  ""i, "" ~li:!'1!!:in	!*
	 ,	
                         a site, If the plume under consideration is at steady-state, its plume foot-
                         print should remain constant in position over time until the source of con-
                         tamination is depleted or physically removed. Following source depletion
                         or removal, the dissolved plume will begin to contract as the assimilation of
                         contaminants in the aquifer exceeds their release rate from the source area.
                     •	;  '•     •  '•  '•	•'  "  '- ' "	 * •  •  '     	 i  	    	: 	
                         The impact of source removal can be modeled using the approach described
                         by Gorder et al. (1996) which entails superimposing a plume with a source
                         area concentration, -Co, at the time of source removal  and modeling the
                         combined plume for time intervals t + T, where t = cumulative time from the
                         beginning of the release to when  source removal occurred, and T = time
                         since source removal occurred. This procedure allows the prediction of the
                         time required for the dissolved plume to degrade below the level of regula-
                         tory concern following source removal. Based on this information, a deci-
                         sion Can be made regarding the expected benefit from source removal in
                         terms of reducing the time required for management of the site to ensure
                         long-term risk reduction.
                 	        	        8.90
         	       	          	j.        ,       :   ..  :    „
":.t    i;*!!! 	   • . •• , A,:1  llijjll	,,''	• ill! ,1', -jri  ">•' '   . i.  .'!•.	.,.: I,1,	 -   , ,	 i ...  •' I	 '  • :,i	!!.  I'll,11!- ••  ,!   :'!-,„  -' ,. S, '.,..,. *j* I-B	I:..
                   	iiiiijiii.,;,!' .1	«:	i	aii^    , !•!	iii . i''*:"!!:" • •'	A :m	riiii;: w'iimi	i.. ..':,': -i."i,.	•	im	a; iiitii,	i>,,' »iiiii'.,;.' •• > ,ii!!ifiii	,} qiii -iii A	>,; ..iiiiiH •'«  A ,:• ;ni i *' '1.11. 'fcjii ait;  t"
-------
                                                           Chapter 8
   This analysis was conducted for Site 13/26 where the source term Was pro-
jected to have an extensive lifetime. Results for benzene plume centerline con-
centration predictions are shown in Figure 8.32. As indicated in this figure, due
to the rapid degradation rate observed for benzene at this site, plume concentra-
tions are projected to be below regulatory limits of 5 mg/L within 7 years fol-
lowing source removal. While 100% source removal is probably impossible,
results of this analysis indicate that source removal is important at the site due
to the large mass of residual-phase material that serves as a continuous,
long-term source of groundwater contamination and the rapid assimilation of
this plume once the flux of contaminant into the aquifer is halted.

8.7.3  Results of Intrinsic Remediation Assessment

8.7.3.1  Selection of Intrinsic Remediation Plume Management
Approach
   Based on decision logic presented in Figure 5.15, selection of intrinsic
remediation depends on the acceptability of the expected life of the plume,
which would be long at Site 13/26 without some source removal.  An evalua-
tion of the assimilative capacity of the TEA pool existing at Site 13/26 using
the stoichiometry presented in Table 5.11 is summarized in Table 8.10.
These calculations suggest that marginal levels of TEAs exist to metabolize
the dissolved BTEX components found at Site 13/26 and that nitrate and
sulfate are the two primary active TEAs there.
   Based on: (1) the result of the source mass and source lifetime calcula-
tions previously presented, (2) the prediction of a short plume lifetime with
source removal at the site, and (3) the finding that an insufficient supply of
TEAs exists throughout the site to provide continual attenuation of the hy-
drocarbon plume without source treatment, the final recommendation for
Site 13/26 was implementation of intrinsic remediation for dissolved plume
management, with active source removal to reduce the source and plume
lifetime and contaminant flux into the plume to acceptable levels.

8.7.3.2  Long-Term Monitoring Plan
   With implementation of intrinsic remediation, a long-term monitoring
network is required. For this network to serve multiple purposes, a combina-
tion of upgradient, downgradient, and within-plume monitoring locations is
necessary.  The recommended monitoring network for Site 13/26 is shown in
Figure 8.33.
                                 8.91
-------
                                          Case Studies
• J:JK ! <•,	i  t,	••     • it	"
                                                              '" -  ;	'  •'•    ;Figur© 8.32"
                                                          Benzene Centerline Concentration^ Predicted
                                                         1,5, and 7 Years Following jrjo%Source Removal
^
IJ>
•£?
§
'I
g
<3
1
1
a
1
u
B
1

'
1,200


1,000

800

600

400

200 (

1
•1
I
t

t
_ I
1
1
- \
\
\
V
tl


                                                           50
100     150     200     250     300      350
    Downgradient Distance from Source Area (ft)
                                                                                                                 400     450
                                                                 100     150     200     250     300     350    400     450
                                                                 1 '•      	   • '"   i""	'	•	" '	•• if1 	   ' ' 	'       '  	]	
                                                                    Downgradient Distance from Source Area (ft)
                                        —	• Predicted Steady-State Plume with No Source Removal
                                        —O— Predicted @ t = 1 yr after 100% Source Removal
                                          •  • Predicted @ t = 5 yr after 100% Source Removal
                                        •"•••:•• Predicted @t = 7 yr after 100% Source Removal
                                        Simulation Parameters: Degradation Rate = 0.00145/d, R = £.0, Steady-State Plume
                                        Source: UWRL1997
                                                                                8.92

-------
                                                          Chapter 8
                             Table 8.10
    Expressed Assimilative Capacity of the Aquifer System—July 1995
Electron Acceptor/Process
Dissolved oxygen
Nitrate-N
Iron/manganese reduction
Sulfate
Methanogenesis
Total Assimilative Capacity
Highest Observed Dissolved BTEX
Highest Observed Dissolved TPH
(Hg BTEX or TPH/L)
1,030
9,350
4,360
10,100
485
25,325
26,702
60,468
  Two sets of wells are used at Site 13/26 as part of the long-term monitor-
ing strategy. The first set, the long-term monitoring wells, consists of a
transect of plume centerline wells composed of one existing well located
upgradient of the JP-4 source area (Monitoring Well 13-1) and six additional
wells located within the observed hydrocarbon plume area. These wells are
used to verify the intrinsic remediation process and allow  the conceptual
model to be updated for plume and source area configuration over time. The
second set of monitoring wells consists of & transect of three existing wells
(26-4, 26-12,26-6) generally perpendicular to the direction of plume migra-
tion (approximately 100 m [305 ft]) downgradient from monitoring point
TP11) and an existing well, 26-15, approximately  100 m (305 ft)
downgradient of sampling point Sp32 to establish the point-of-compliance
(POC) for this site. The purpose of the POC wells is to verify that no BTEX
exceeding the federal MCL (5 |.ig/L) migrates beyond the area under institu-
tional control.
   A three- to five-year  sampling  frequency was recommended for the
site due to the projected lifetime of the source area and dissolved plume
with complete source removal. This interval provides sufficient data
over time to verify plume stability and source area depletion at a reason-
able frequency based on cost considerations, without compromising
human health or environmental quality. Samples should be collected at
the same time of the year to ensure comparable groundwater table eleva-
tions at each sampling event so that true changes in groundwater con-
centrations can be identified from historical data.
                                 8.93
-------
 Case Studies
  0)

  0
  o>
  D

  O


  O



  §!

  O
  o
  •  o
  -£ o
«> H—
2 o
  
 O
 o
                                                    .ig
                                                    i!
f»
II
                                                    ll£|  N
                                                    sill  §
                                                    TS K) d _  '"
                                                           E
                                                      ro
                                                    I;
                                                       **
                           8.94
-------
                                                            Chapter 8
8.7.3.3 Costs
   The field intrinsic remediation study conducted at this site cost approxi-
mately $450,000 over the 2 1/2-year project life.  Initial feasibility study
documentation (U.S. Air Force 1993) identified three treatment schemes for
remediation of petroleum hydrocarbon contamination at Site 13/26. These
included:  (1) a bioventing alternative with capping, bioventing, passive
product skimming, and hydraulic containment of the groundwater plume; (2)
an SVE alternative with soil excavation, SVE, active skimming, and ground-
water extraction and treatment; and (3) an extraction alternative that included
soil excavation and groundwater extraction and treatment. Total remediation
costs for each alternative are summarized in Table 8.11 (U.S. Air Force
1993), while the estimated costs  for an intrinsic remediation management
alternative with long-term monitoring and source control at the site are
shown in Table 8.12.  As indicated in the tables, intrinsic remediation which
does not require soil or groundwater removal and treatment reduces the esti-
mated cost of site remediation from $16,000,000 to $39,000,000  to less than
$6,500,000 while achieving the same remedial goals which are protective of
public health and the environment.

8.7.4  Summary and Conclusions
   This case study highlights the intrinsic remediation process  described in
detail in Section 5.3.  This process involves:  (1) the assessment of
steady-state plume conditions, (2) determination of degradation rates, (3)
estimation of the source quantity, (4) estimation of the source lifetime, (5)
prediction of the long-term behavior of the plume with and without source
removal, (6) assessment of aquifer assimilative capacity and the desirability
of source removal at the site, and (7) development of a long-term monitoring
strategy for verification of intrinsic remediation process performance  and
regulatory compliance purposes. These procedures were successfully ap-
plied at a hydrocarbon-contaminated site at Eielson AFB, Alaska.
   Approximately 950 m3 (250,000 gal) of fuel were thought to remain as
residual saturation in the source area of the largest plume found at this site.
With an assessment of the impact of source removal on the lifetime of the
dissolved plume, it was recommended that some active source  removal take
place to reduce the baseline source lifetime estimate (>1500 years) to a more
acceptable level. Long-term monitoring with upgradient, in-plume, and
downgradient POC wells was prescribed for the site at a sampling frequency
of once every three to five years to monitor the progress of intrinsic
remediation and allow for the continual refinement of the conceptual model
of fate and transport of hydrocarbon contaminants at the site.
                                       i
                                 8.95
-------
                                    Case Studies
i	i-,...(  :
                                                                     Table 8.11
                                                  Summary of Estimated Present-Worth Costs of
                                                     Proposed Remedial Action Alternatives*
                                                 Alternative
     ...
Bioventing
                                                                                         SVE
                                                                                                      ,,  .
                                                                                                      Extract
                                                                                                            I'.IS
Capital Costs
Soil Remediation
SVE/Bioventing
Passive Skimming
Groundwater Extraction
*,i i;l ' ' .„ :>' ' ,,11 ' ,i
Groundwater Monitoring
Groundwater Treatment
Mobilization
O&M Costs
Soil Remediation
SVE/Bioventing
Passive Skimming
Groundwater Extraction
Groundwater Monitoring
Groundwater Treatment
Contingencies, Administrative, Engineering
TOTAL

$980,500
$558,500
$115,660
$40,000

$5,000
$429^000
$319,200

$913,000
$440,000
$100,000
$1,826,000
$4,458,000
$3,943,000
$1,942,400
$16,070,000

$2,250,000
$1,666,875
::' '• ' " ' ' 	 	 '
$153,000
i 	 I 	 ,,i.-' - . 	 i i
$5,000
$434,000
$676,3(30

$173,000
$342,000

$1,742,000
$2,278,000
$4,166,000
$4,114,825
"i, || • 1 	 !"' • • .;' , i -, -
$18,001,000

$16,540,300

i. r ' 'i .. 	 Hi, i" '
"" $"185,060"
':., '. ,, 1 '3 ,'
$5,000
$434,000
$3,432,800
	
nl| 	 	


$1,246,000
$186,000
$504,066
$16,352,900
$38,886,000 ''
                                    * Unit costs and present-worth calculation used in this cost estimate were taken from those presented in US EPA
                                    1993. Present-worth calculations are made assuming an Interest rate of 5% for an operating period of 50 years for
                                    btoventing, 30 years for SVE, and 20 years for the Extract alternative.
                                      Implementation of the intrinsic remediation plume management approach
                                   at this hydrocarbon-contaminated site appears to be an effective alternative
                                   to conventional remedial options involving pump-and-treat, SVE,, and exca-
                                   vation and off-site treatment of the contaminated soil because it provides
                                   equivalent protection of public health and environmental quality at a signifi-
                                   cantly lower cost.  This is particularly true if the plume lifetime can be re-
                                           • IIIHl!      " '  „„  . . , " ,    	J ",,!I|| !	,i ) ' '!•' ' ,»..,!!",' »' , „". ,'l I'l I ,l!ll" ' . ''I' 	 ' 'In! ,  II    	    lh 	'  i i":.'ll'li ,'
                                   duced to 50 years or less with some form of active remediation within the
                                   source area at the site.
                                                                       8.96
-------
                                                                Chapter 8
                                Table 8.12
         Summary of Estimated Present-Worth Costs for Proposed
            Intrinsic Remediation with Limited Source Removal*
             Capital Costs
               Soil Remediation
               SVE/Bioventing
               Passive Skimming
               Groundwater Extraction
               Groundwater Monitoring
               Groundwater Treatment
               Mobilization
             O&M Costs
               Soil Remediation
               SVE/Bioventing
               Passive Skimming
               Groundwater Extraction
               Groundwater Monitoring
               Groundwater Treatment
             Contingencies, Administrative,
             Engineering
             TOTAL PRESENT WORTH
 $230,700
 $314,700
 $115,000

   $5,000
  $10,000
 $101,310

  $55,000
 $402,000
 $100,000

$4,458,000

 $616,708

$6,408,418
•Unit costs and present-worth calculation used In this cost estimate were taken from those presented in US EPA
1993. Present-worth calculations are made assuming an interest rate of 5% for an operating period of 30 years.
8.8  Land Treatment
   An integrated bioremediation program has been implemented at the
Champion International Superfund Site (Libby Site), a former wood-preserv-
ing facility in Libby, Montana.  The design consists of a land treatment sys-
tem to remediate contaminated soils, a groundwater extraction and
aboveground treatment system to treat heavily-contaminated groundwater,
and an oxygen injection system for in situ treatment of the contaminated
aquifer (Figure 8.34).  The land treatment system consists of two 4,050-m2
(1-acre) prepared-bed land treatment units (LTUs), which have been operat-
ing since 1989. This case study details the design and analysis of pilot-scale
and full-scale LTUs that have led to the successful biological treatment of
wood-preserving waste-contaminated soils sit this Superfund site.
                                   8.97
-------
 Case Studies
                                    Figure 8.34
         Bioremediation Process Technologies Used at the Libby Site
                                  (not to scale)
         3025^

      3031.
  3026.
                          In Situ Grpundwater
                          Bioremediation System
                                                                Groundwater Flow
                                                                N
                                                                A
                                                    Prepared-Bed
                                                    Land Treatment System
• Monitoring Wells
• Injection Wells

Source: SImsetal.1995a
                                      8.98
-------
                                                             Chapter 8
8.8.1  Site Description
   The Libby Site is an active lumber and plywood mill located southeast of
the town of Libby in northwestern Montana. Contamination of soils at the
Libby Site resulted from wood-preserving operations conducted from 1946
to 1969. Disposal of chemical wastes used in the wood-treating processes
contaminated the soils around the mill and the underlying groundwater. Re-
sidual contamination consists primarily of creosote and pentachlorophenol
(PCP) wood preservatives. Specific contaminants of most concern in soils at
the site include:                                    .
        •  polycyclic aromatic hydrocarbons (PAH compounds), which are
           the primary components of creoisote.  PAH compounds are asso-
           ciated primarily with the soil solid phase by adsorption;
        •  PCP which is somewhat volatile and, in the ionized form, soluble
           in water; and
        •  dioxins, an impure component in technical-grade PCP. Dioxins
           are nonvolatile, highly insoluble in water, and closely associated
           with the soil solid phase.
   Contaminated soils were located in three primary source areas at the
Libby Site: (1) an unlined waste pit, where wastewater and sludges had been
discharged during the operation of the pole-treating retort facility; (2) an
unlined pole butt-dip area, where treating fluids had been spilled during
operations; and (3) a former tank farm storage area, where treating fluids
were accidentally spilled into the surrounding bermed area (Figure 8.35). In
1989, contaminated soils from these three areas (approximately 57,400 m3
[75,000 yd3] of materials) were excavated down to the water table. Before
the tank farm and butt-dip areas were filled with clean soil, samples were
collected and analyzed to verify that contarrdnation had been removed.
   Because the major contaminants of concern were expected to be associ-
ated with finer-grained materials, the soils excavated from the tank farm and
butt-dip areas and the contaminated materials excavated from the waste pit
area (a totalof approximately 57,400 m3 [75,000 yd3] of soil and rock mate-
rials) were physically screened to remove rocks larger than 2.54 cm (1 in.) in
diameter (referred to as de-rocking). The screened soils from all three areas
(approximately 34,400 m3 [45,000 yd3] were placed in the excavated waste
pit area. The separated rocks were placed upgradient to the waste pit area to
construct sub-grade infiltration galleries.  This rock percolation bed  is used
for biological treatment of the contaminated rocks using effluent from the
above-grade, fixed-film bioreactor that is used to treat contaminated ground-
water at the site.

                                 8.99
-------
                                   Case Studies
  '	l!J  '   I
 «„,
                                                  "• '  '';''.,"'' '  ' ;/: Figure8.35	
                                                 Contamination Source Areas at the Libby Site
                                                                (not to scale)
                                   Source: Sims et at. 1995a
 ,,,'S"
                       Initial soil concentrations of contaminants of concern, expressed as a
                     geometric mean, in the contaminated soils from all three areas were deter-
                     mined as follows:
                                 ,,    '' '   ''   ,!   ,!•',"        , 	  , »'  ||;;	'!„	 • , I", ' ,, v , '  " •'   ' 	  I llii,!!!, ' 1
                             •  189 mg/kg total carcinogenic PAH compounds;
                             	    ,. •  ,   ,  ;,        	;;;	, , „;., I	  „      ,,	; ,	 :.,
                             •  29 mg/kg PCP; and
                             •  0.9 x 10;3 mg/kg tetrachioro-dibenzo-p-dioxm (2,3,7,8-tCDDi
                               equivalency.
                       However, the concentrations of contaminants in the soils varied from
                     sample to sample; the maximum concentrations for individual carcinogenic
                     PAH compounds, PCP, and 2,3,7,8-TCDD equivalency were greater by fac-
                     tors from 6 to 90 than the geometric mean  concentrations.

                                                     8.100
•• iir> .,i	1 i1 j] i l!
E ; L, i!
          :,,	i	i:11  i;
                                                                          . ill,:.
-------
                                        i                     Chapters


 8.8.2 Initial Remedial Goals/Regulatory Environment
   The contaminated surface soils at the Libby Site presented a potential public
 health threat via direct contact and ingestion. They were also of concern be-
 cause they served as source materials for ongoing releases of contaminants to
 the groundwater.  In 1979, Libby residents living near the mill began drilling
 wells for lawn irrigation after the public utility increased their water rates. Resi-
 dents adjacent to the mill discovered creosote and PCP chemicals that discol-
 ored and caused odor in the water from their newly drilled wells.
   In 1983,  the US EPA placed the Libby Site on the National Priorities List
 as a result of the residents' concerns and initial investigations.  In 1985,
 Champion International provided an alternative water supply to people
 whose wells were contaminated and conducted studies of the contamination
 problems, which included pilot-scale testing of remedial technologies. The
 City of Libby enacted a well permitting ordinance that prohibited new wells
 in the areas  of contamination.
   In 1988,  a pilot-scale land treatment demonstration unit (LTDU), 24 m by
 12 m (80  ft  by 40 ft), was constructed with berms adjacent to the waste pit
 area (Piotrowski 1991). Baseline sampling of PAH and PCP concentrations
 in the soil layers below the LTDU was concluded.  Approximately 46 m3 (60
 yd3) of screened contaminated soil  was placed on the unit and spread to a
 uniform depth of 15 cm (6 in.). Initial PAH' and PCP concentrations were
 determined  in this soil lift. The soil was periodically tilled and irrigated with
 a nutrient/microbial solution. Nutrients consisted of a dilute solution of
 inorganic nitrogen and phosphorus. The microorganisms used in the mix
 were isolated from the soils at the site and were grown as a mixed culture in
 26.5-m3 (7,000-gal) batches, with molasses as the organic growth substrate.
 Individual microbial species within the innoculum were not identified.
   During the treatment period, soil pH and moisture content were moni-
 tored.  Moisture adjustments were made as required; no pH adjustments
 were required. Triplicate, composited soil samples were collected monthly
 for three months to determine changes in contaminant levels in the soil lift
 during treatment. At the end of the study, soil layers below the LTDU were
 analyzed for PAH and PCP. Results were compared to the baseline data
 collected before a lift was placed on the LTDU to evaluate contaminant mi-
 gration into  the underlying soil during lift treatment.
   After 100 days of treatment, large reductiions in contaminant concentra-
tions were achieved (Table 8.13). In addition, little vertical migration was
observed below the LTDU, indicating that the contaminant reductions had
been primarily the result of biodegradation. Monthly results showed that most
of the biodegradation had occurred during the first 48 days of treatment.
                                 8.101
-------
Case Studies
          •	:   '       Table 8.13
        Summary of LTDU Results for Contaminant Concentrations
                      in, Treated Soil Over Time — 1988
- " ' '. 	 1 '
Concentrations' (mg/kg)
,, , , Date
July 1
August 1?
1 1"'1" i,1":: ' •
September 12
October 10
Percent Reduction (Overall)
PCP
i •. ,, • i,.!-':!!;.!!',''!' •::.' i s:".
750 ±292
90±13
" 	 	 •' , «„
24±8
22 ±0
97
Total PAH
Compounds2
785.6 ±133.9
17i.2±15.4
96.0 ±5.2
73.6 ± 1.9
91
Carcinogenic PAH
Compounds3
231.6 ± 47. &
137.1 ± 15.7
68^9 ±6.9
62.8 ±1.8
74
1 Concentrations expressed as mean ± standard error of analytical results from three composite replicates collected
  on each date,
2 Total PAH compounds = Sum of concentrations'of 16 priority pollutant PAH compounds.
3 Carcinogenic PAH compounds = Sum of concentrations of 12 of the 16 priority pollutant PAH compounds that are
  considered to ba or are potentially carcinogenic.
  	:	     	            I      '   '     '       	
Source: Woodward and Clyde 1990
   Based on the results of the pilot-scale studies, in December 1988, the
US EPA and the Montana Department of Health and Environmental Sciences
signed a Record of Decision (ROD) that designated biological treatment as
the remedial method for both soil and groundwater remediation.  Target
remediation levels (cleanup goals) for the contaminated soils (on a
dry-weight basis), as specified in the ROD (US EPA 1988c), were:
        •  88! mg/kg total carcinogenic PAHs [sum of fluoranthene, pyrene,
           benzo(a)anthracene, chrysene, benzo(b)fluoranthene,
           benzo(k)fluoranthene, benzo(a)pyrene, dibenzo(a,h)anthracene,
           benzp(g,h,i)perylene, and indeno(l,2,3-cd)pyrene];
        •  8 mg/kg naphthalene;
        •  8 mg/kg phenanthrene;
         ~   '  •  -	'  ' '   '  •''','.'..,  "" '  .:;.'   .'..''.' l  ... ,  '".'"   .   . '  	:. \'£.
        •  7.3 mg/kg pyrene;
     ••• 	;  •  ' " 	 ";,:i.i, '•,!! •.. IT,.': ••• r  -  .• '	, •  ,   .'IliS,• -' ;	' •! . I. :| "   • •   '     "  , /.., >< I--*'*  •
   •. '!:?  *  37 mgTkg PCP; and	
        •  <0.00i mg/kg dioxin equivalency (sum of
           2,3,7,8-TCDD-equivalent concentrations of polychlorinated
           dibenzo-p-dioxins and dibenzofurans).
                                   8,102
-------
                                                           Chapter 8
   During the summer of 1989, a full-scale land treatment demonstration was
conducted to collect information on contaminant degradation rates, evaluate the
potential for contaminants to migrate downward during treatment, and demon-
strate that biodegradation was the major mechanism of contaminant loss. The
first of two 4,050-m2 (1-acre) LTUs was constructed (see Section 8.8.3, Design
Approach).  The 1989 demonstration involved sequential application and treat-
ment of two 30.5-cm (12-in.) thick lifts of contaminated screened soil to the
LTU (Piotrowski 1991; Woodward-Clyde Consultants 1989c). Each lift con-
tained approximately 600 m3 (800 yd3) of soil.
   The first lift had been biologically pretreated with periodic tilling and
irrigation during storage in the former wastes pit area before it was applied to
the LTU. Results from treatment of this lift were anticipated to be represen-
tative of LTU performance under typical operating conditions. The second
lift was freshly excavated and contained recognizable fragments of pure
naphthalene crystals.  Results from treatmerit of this lift were expected to be
representative of LTU performance under "\yorst-case" conditions
(Piotrowski 1991).
   The first lift was applied in July  1989.  The lift was periodically tilled
and irrigated as necessary. No microbial solution was added.  Compos-
ite soil samples were collected from four quadrants in the LTU every
other week, and the samples were analyzed for PAH compounds and
PCP.  Leachate samples were also collected and analyzed when suffi-
cient leachate accumulated.  Groundwater samples were periodically
collected from monitoring wells located nip and downgradient from the
LTU and analyzed for PAH compounds and PCP. Treatment of the first
lift was continued for approximately 1 month, when analytical results
indicated that target remediation levels had been achieved for all con-
taminants except pyrene (Figures 8.36 and 8.37).
   The second lift was applied in August 1989.  Treatment continued until
early November (three months). Higher initial contaminant concentrations
and cooler ambient fall temperatures may have increased the time required
for treatment of the second lift over that required for the first lift (Figures
8.37 and 8.38).                                              ,
   Leachate and groundwater analyses indicated that little contamination
was migrating downward during land treatment. Similarly, qualitative inves-
tigations indicated that little contamination was volatilizing during tilling
operations.  In addition, appreciable numbers of total and viable microorgan-
isms were counted in soil samples.
                                8.103
-------
           Case Studies
                                   ;    ,  •	Figure 8.36
                         Mean Concentrations of Target Ccsntdminants
                                in Soils Treated (n trie LJU — T989
	-i, i!	t
   :i "" '!

.litiiil.  •'• :•
                          A. Total Carcinogenic PAHs '(Target remediation level: 88 mg/kg)

                               300
                          §
                                     0   20   40   60   80  100  120  140  160
                                               Day Number
                             1 "     .":     I i , "Vi .Jr.?'i .C	::»! • .' i  l-'j"!  | ,:, ;•••$ ,;,;,; . • | .!•
                            B. Pentachlorophenol (Target- remediation level: 37 mg/kg)
                                        20  40  60  80  100  120  140  160
                                               Day Number
Day Number 0 = June 31 st
Day Number 50 = August 8th
Day Number 150 = November 10th
Error bar = ±1 standard error
Source: Piotrowskl 1991
"Y  ,1.  ' +,,,';,.:   .  '   ••.  "
                                                                    ,	i'r I iii;
             During operation of the LTU, researchers from Texas A&M University
          evaluated toxicity of the treated soils using the Salmonella/miciosomG bioas-
          say (Ames test) (Donnelly et al. 1992). Soil samples from two lifts in Cell 1
          and background samples from the surrounding areas were screened. One lift
          had undergone land treatment for three months and contaminant concentra-
          tions had reached remedial goals.  The second lift had been treated for five
          weeks, but concentrations had not yet reached remedial goals. Results indi-
          cated that toxicity levels of the fully-treated lift samples were within the
          range of the values  for the  Libby background soil samples, while the toxicity
                                             8.104
                                                  ..I .      .1 j
-------
                2  ff   m
                c   3  ««•£
                i   H-
                TJ

                O
ZZZ
c c c
333
                        '-tcno
                        f?2.
                        " J»c
                        fls
                        m
CO

o
en
                                             Mean Concentration
                                               (±lSE)(mg/kg)
                                                                   3
                                                                   f
                                                                    B
                                                                    B
                                                                    S _
sr
a

i
€
              Mean Concentration
                (±lSE)(mg/kg)


           _IO*.O\OOOtOJxO\
           ooooooooo
     Mean Concentration
       (±lSE)(mg/kg)


            tf  6
                                                                                                        a
                                                                                                        i
                                                                                3
                                                                                t
                                                                                3
o


8

£


§
                                                                                                        I
                                                                                                        00
                                                                                                        3
                                                                                                                               (D
                                                                                                                               Q
                                                                                                                               3

                                                                                                                               O
                                                                                                                             .0
                                                                                                                            en  (D
                                                                                                                            -H D
                                                                                   en C

                                                                                   O fl»
                                                                                                                             I  Q
                                                                                                                                                    00 Q


                                                                                                                                                       5
                                                                                                                                                       Q

                                                                                                                                                       I
                                                                                                                                                                      O

                                                                                                                                                                      Q

                                                                                                                                                                      U
                                                                                                                                                                      S
                                                                                                                                                                      oo
-------
Case Studies
                               Figure 8.38
     Configuration of Treatment Cells Used in the LTU at the Libby Site
Source: Simsetal. 1995a
levels of the partially-treated lift samples were twice as high as background
toxicity levels.  The toxicity levels of the untreated waste pit soil, samples
were one to two orders of magnitude higher than toxicity levels in the treated
soils. Therefore, biological treatment of the soil did not form intermediate
breakdown products that were more toxic than the parent organic contami-
nants (Piotrowski et al. 1994).
     ;.   .   ..."    • ••; .  .•,•.,". '..,, ,."   '  - • '•;	     >• i - 	            •    i-'-': "
   Based on the results of the demonstration, a consent decree that required
Champion International to use bioremediation as the selected remedial
                                 8.106

-------
                                                             Chapter 8
 technology was entered in federal district court in October 1989. Although
 US EPA had formally approved the plan, land disposal restrictions promul-
 gated under the RCRA restricted application of the soils to land after August
 8,1990. Therefore, a "No-Migration Petition" (Woodward-Clyde Consult-
 ants 1989d, 1990) was filed with US EPA in February 1990, which included
 data from the demonstration conducted in 1989 that showed that no migra-
 tion of contaminants would occur during treatment. The US EPA formally
 approved the petition in October 1990, and full-scale soil remedial activities
 commenced at the site in 1991.

 8.8.3 Design Approach

   The soil remediation program was designisd to reduce organic contaminant
 concentrations to target levels that had been defined as acceptable from a regu-
 latory perspective and to minimize risks to public health and the environment.
   The location for construction of the two4,050-m2 (1-acre) LTU cells was
 selected based on previously existing site factors as well as possible future
 influences resulting from operation of the facility.  Previously existing site
 factors that influenced selection of the LTU location included: (1) proximity
 to the contaminated soils to be treated; (2) company on-site operational con-
 straints; and (3) soil and water quality analyses that indicated low level con-
 tamination of surface soil and subsurface environments.          ;
   Contaminated soils are treated in lifts (approximately 15 to 30 cm [6 to 12
 in.] in thickness) in the designated LTU cell until target soil contaminant levels
 are achieved for a given lift.  Degradation rates, amount (volume) of soil to be
 treated, initial contaminant concentration, duration of summer operational pe-
 riod, and LTU size determine the time required to remediate a given lift of con-
 taminated soil. Based on an estimated 45-day time frame for remediation of
 each applied lift of contaminated soil to acceptable contaminant levels, an esti-
 mated volume of 34,400 m3 (45,000 yd3) of contaminated soil requiring
 remediation, and a 8,100-m2 (2-aere) total LTU surface area, the time for
 completion of soil remediation was initially estimated to be 8 to 10 years.
  Design criteria for each LTU cell include provisions for total containment
 of contaminated soils, water, and leachate, with ultimate  treatment and dis-
posal of all contaminated soils within the LTU.

 8.8,3.1 Size
  The lined, prepared-bed LTU is compose*! of two cells with a final surface
area of 4,050m2 (1 acre) each (Figures 8.38 and 8.39). The first cell was
completed in 1989, while the second cell was constructed in 1991. The

                                8.107
-------
                                                                                         Jit
                    Case Studies
	          '.  •          .   '         ,       ,;•;       '        ;  ••      .1 ..   '  ,  ;	,} ' •"  ;  ,'•'
                    surface area required was based on the estimatedtotal quantity of material
                    excavated from the source areas minus the quantity of materials greater than
                    2.54 cm (1 in.) in diameter removed during de-rocking since contaminants
                    were expected to be associated with finer soil materials.
ill"!!"', '••     •    ' "'    i'l          '     ' •          i,       "  ',.•'''  I,  ' ilL" „«    '/I',, i,,,! I • '''I ''   " ':"  •','!'    ,   ' 	;> I Mi!!1
                       Each LTU cell is surrounded by a bermconstructed with low-permeability
                    soils that were compacted with a dozer, Incremental berm construction de-
                    sign allows modification of the LTU height to increase storage capacity as
                    needed. At least 0.6 m (2 ft) of elevation difference is maintained between
                    the top of the treatment zone and the top of the berms.  The use of berms
                    allows for containment, treatment, and ultimate disposal of additional con-
                    taminated soils if required. The berm. was designed to control run-on and
                    runoff associated with a 25-year, 24-hour storm event.
                       The LTU cells are each sloped to a central gravel drain (2% slope) to
                    control water within the unit. The gravel drain also is sloped to a gravel
                    sump (1% slope).
           ;,'	";i   '   T. •  ,;    ,       •.-.•    .,•••... ;•.,     !:; " i :„"  , vi  • i '     ;••• •'    •    ••  •<'"«•
                    8.8.3.2 Treatment Zone
                       The treatment zone (i.e., the zone in which the contaminated soil is tilled
                    and treated with moisture and nutrients) consists of a lower layer of sandy
                    material (46 cm [18 in.]) supporting a top layer (30.5 cm [12 in.]) of silty
                    material. The sandy material was collected on-site and  consists of uncon-
                    taminated material meeting the Unified Soil Classification System (USCS)
                    definition of SP-SM (poorly graded sands "or gravelly sands and silty sands)
                    or SM (silty sands). Maximum size is 1.3 cm (1/2 in.).  The silty material
                    consists of silts and/or clays collected from an on-site area and meets the
                    USCS definition of ML (silts and very fine sands, silty or clayey fine sands,
                    or clayey silts of low plasticity), ML-CL (silts and very fine sands, silty or
                    clayey fine sands, or clayey silts of low plasticity with.clays oflow to me-
                    dium plasticity or gravelly, sandy, or silty clays) or CL-ML (clays of low to
                    medium plasticity or gravelly, sandy, or silty clays with silts arid very fine
                    sands, silty or clayey fine sands, or clayey silts of low plasticity). Standard
                    filter criteria were used for material sizing between the treatment zone silty
                    layer and sandy layer to reduce clogging potential.
                       The contaminated soil is placed on top of the sandy/silty layer and is
                    actively treated through management activities. After reaching target
                    remediation levels, additional liftsof contaminated soil  are placed on previ-
                    ously-treated soils in the treatment zone.
                                                     8.108
-------
                                                           Figure 8.39
                                  Cross-Section of Treatment Cells Used in the LTU at the Ubby Site
                                                          (not to scale)
00


A
r 2,120
2,110
L 2,090
Elev. (ft)



,Berm
^^^El. 2.094 ft /Gravel Sump
2,094ft ?
Approx. Seasonal High
Groundwater Elevation
A-A'
Cross-Section

Liner & Leachate
/ Soil Liner / Collection System
In Situ Soils





/ Gravel D"^ B1.2;112ft





A'
2,120 -I
2,110-
2,100-
2,090-1
Elev. (ft)



B
r 2,120
2,110
•2,100
•2,090
B-B'
Cross-Section
yBerm „ .. w .
/ Liner & Leachate rra,minroin /s°»l Liner
^_ El. 2.112 ft /Collection System /Gravel Draln /E\. 2,112 ft
^•^Hbfe^ / .,2% > ^ < 2% / _^&^^
_?_ 12^r*T2% In Situ Soils


Elev (fA Approx. Seasonal High
Elcv-w Groundwater Elevation


B'
2,120 -I
2,110-
2,100-
2,090 J
Elev. (ft)
                                                                                                                                  9
                                                                                                                                  Q

                                                                                                                                 f
                                                                                                                                  CO
Source: Simsetal. 1995a
-------

Case Studies
8.8.3,3 Liner System
   A liner system was designed to minimize migration of leachate that may
bf generated from treatment operations and that might otherwise continue
downward through unsaturated zone soils into groundwater.
   The underlying liner system for each LTU cell consists of a 60-mil syn-
thetic flexible geomembrane liner placed on top of a compacted soil liner
(46 cm [18 in.] thick) constructed from low-permeability (5 • 10-s cm/sec)
soils (glacial lake sediments) collected on-site. The compacted soil beneath
the high density polyethylene (HDPE) liner was compacted to accomplish a
maximum permeability of 5 "• 10'7 cm/sec.  HDPE was chosen for the
geomembrane liner due to its documented compatibility with most common
wastes arid waste byproducts.  Leakage testing of the geomembrane liner
was performed using electrical resistivity. This method involves flooding the
lined facility and installing an electrical source in the water within the con-
tained area and an electrode outside the unit to complete the electrical cir-
cuit. The intact geomembrane liner acts as a resistance to the imposed cur-
rent, and any leaks can be detected using voltmeters to locate areas of
high-current flows. Additional analysis of leakage through the liner under
different scenarios of liner rupture was accomplished using the Hydrological
Evaluation of Leachate Performance (HELP) model
            '.,,,,  	,'	  ,        ,| ,'      „; '   . I i 1. '  '     ,      .    "

8.8.3.4 Leachate Collection System
   The purpose of the leachate collection system is to prevent leachate from
accumulating within the LTU and to monitor contaminant concentrations.
The leachate collection system, including surface water pumping, subsurface
drainage net, gravel drains, and collection pipes, is designed to collect
leachate generated from two sources of water: (1) water applied during
operation of the LTU, and (2) water from precipitation events. Leachate
collected in the bottom of each cell is removed to minimize buildup of
leachate (hydraulic head) on the liner system, thus reducing potential for
,,',,!|!    '„,  >  ^ | ! ,   ^	, , ,„   ,n|1  I , I,  | ,||, III,  ]"': I	|| «   , I ,| |f  I  "!'     	.11    '   ,1	  I III, I"
leakage, and to prevent free water buildup in the LTU that could eventually
lead to horizontal migration if the water levels were to exceed the top of the
flexible membrane liner.
   A drainage net (Tensar DN-3) covered by a geotextile filter fabric (Typar
3601) was placed over the geomembrane/soil liner system (Figures 8.40 and
8.41). Water filtration criteria were used to select the geotextile filter fabric
in order to minimize migration of treatment zone sands from above into the
drainage net, which' could clog the leachate collection system.
                                8.110
-------
                                                              Chapter 8
                               Figure 8.40
           Geomembrane/Soil Liner System Used for Leachate
                   Collection in the LTU at the Libby Site
                              (not to scale)
                                    6 in. Perforated
                                    HOPE Pipe
                                    D
                             	/- Treatment Zone	r- -
                                                             4 in. Perforated
                                                             HOPE Pipe
 Additional Pipe
 Cover on Berm
    12 in. Thick Silts
    18 in. Thick Sands
 £-Geotextile Filter
   Fabric
                 18 in. Thick Soil Liner
\
  Drainage Net
60-mile HDPE
Geomembrane
Source: Sims et at. 1995a
   A gravel drain (30.5-cm [12-in.] thickness) was constructed along the
entire length of the floor in each LTU cell.  River gravels, which contain
non-angular materials, were used in the drainage system to reduce puncture
potential and to maintain liner integrity. A collection sump and sloping riser
were constructed at the Ibwest point of the  gravel drain (at the north end of
each LTU cell, centered in the east-west direction). Two 10-cm (4-in.) diam-
eter slotted HDPE pipes  were wrapped in geotextile filter fabric and placed
in the gravel drain which was sloped to the: collection sump. The drain and
sump were backfilled with gravel and completely enclosed in geotextile filter
fabric. A single slotted 15-cm (6-in.) diameter pipe was located along the
base of each sump and was connected to a solid 15-cm (6-in.) diameter
HDPE pipe that rises upward along the interior slope of the north berm of
each cell. This pipe provides access to the  sump area for leachate removal.
                                  8.111
-------
 Case Studies
                                    Figure 8.41
                     Cross-Sections of leachate Collection
                    System Used in the ttU at the Lifciby Site
                                  (not to scale)
                    :.••':..     •    •' ';•"'  '»'- v".' '  • ::    . <•  !.!•,..•'
                                  Cross-Section F-F'
                              Typical Gravel Drain Section
                                4 in. Perforated HOPE Pipes   .Geotextile Filter Fabric
   12 im Thick Silts

   18 in. Thick Sands
                _ I	60-mile HOPE
          ~         -Drainage Net  /Geomembrane
- Treatment Zone	
   ,.,. „,  I?ft
   18 in. Thick Soil Liner VWeld (Typ.) /I
                                                               12%
                                                           In Situ Soil
                                             Gravel Drain  % Oeotextile Protective Fabric
                                  Cross-Section D-D
    6 in. Diameter Perforated HOPE Pipe
    12 in. Thick Silts
                               \
                 .Geotextile Protective Fabric
                                 Geotextile Filter
  18in.Thick.Sands _ _ &Z~Z
                                    •Treatment Zone
   Weld (Typ.)-

                             fj-vjvjyj"
     '•tf.f.fji&3jr
                             -^/Fabric
                             Mt—Drain
                                                                    Drainage Net
          18 in. Gravel Sump

In Situ Soil
                                 Geomembrane
                         18 in. Thick Soil Liner
                                  Cross-Section E-E'
                      •'.:—-Treatment Zone —3—
          6 in. Diameter Perforated HDPE Pipe —*^j
                                           12 in. Thick Silts
        18 in. Thick Sands
                                        Geotextile Filter
                                        Fabric
                    	Drainage Net
                     18 in. Thick Soil Liner *\60-mile HDPE
                     	   Geomembrane
                                    In Situ Soil
     11 ',; 'i1'1"
Source: Sims etal. 1995a
   Leachate is removed from the leachate collection system sump area using
an automated leachate collection pump and piping system (which can be
overridden for manual control).  The automated system ensures that a sig-
nificant level of water will not collect in the sump area.  Self-priming pumps
    „ »  ,   , 	        i  'i „,,'„''  ',       ,  'I li  i ;i   'flMii :•. '  • , ,i iklllll..  l| |, ,   ' : ,!   , i,     ' I';," „ I  '!•, ." ...  || nil*",
located in a heated pump house are used for leachate removal.  High- and
low-level automatic pump activation switches were installed near the base of
the HDPE pipe located in the sump for each cell.  The level controls are
                                      8.112
-------
                                                            Chapter 8
usually set so that 1 to 2 m3 (300 to 600 gal) are pumped when the system is
activated. The flow rate that is achieved is monitored using a flow meter
located within the pump house.  Discharge pipes installed below the frost
depth carry leachate to two 190-m3 (50,000-gal) storage tanks or to the
bioreactor. These pipes are insulated and heat-taped where exposed.
                                      i              •'.'"•'
   Daily inspections of the pump removal system are made on regular work
days. If failure should occur in the pumps or piping system, the system is re-
paired within one week to ensure continued removal of leachate from the LTU.
                                                           .......
   .Surface water is managed by daily monitoring of the LTU during regular
work days to see if significant amounts of water have collected on the sur-
face of the LTU. The surfaces of the LTU cells are sloped so that surface
water collects at the low point of the cells above the leachate collection
sumps. If water collects to a sufficient depth to be pumped by a submersible
pump, the water is promptly removed until the submersible pump is unable
to continue pumping.
   The recovered water is sprayed directly on the rock pad or injected into
closed or open trenches.  These disposal areas were selected because they
should be able to handle the maximum design storm event, which is the
24-hour, 25-year storm of 6 cm (2.4 in.). This maximum design storm event
could result in approximately 570 m3 (150,000 gal.) over the 24-hour event
being recovered from the 8,100-m2 (2-acre) LTUs and associated haul roads.
The trenches were tested at a 15.8-Lps (250-gpm) injection rate and were
able to handle water discharged at this rate.  The leachate collection pumps
have a pumping rate of only 3 Lps (50 gpm), so the infiltration trenches
should be able to handle the water being recovered from the LTU at this
lower rate.

8.8.3.5 Leachate Storage Unit
   Effluent in the two 190 m3 (50,000 gal) storage tanks can be directed to:  (1)
the LTU cells for irrigation in the summer months; (2) the infiltration galleries/
infiltration trench, where it can be amended with nutrients for use in; the in situ
aquifer bioremediation system; (3) to the rock pad; or (4) to open trenches. To
prevent freezing of collected liquids in the storage tanks, the design incorporates
a combination of electric immersion heating and sparging with warm, com-
pressed air. In addition, all exposed piping is insulated.

8.8.3.6 Passive Moisture Control System
   A passive moisture control system is installed within the LTU  adjacent to
the incremental berms to minimize the potential for soils to become

                                8.113
-------
Case Studies
    ,1,1 '!   .IIP i,  " v1  	.Illli  IP" "  I'I'llli	 „ Jill" ,!, ,11 |H', li '"  ', ,	„   ' In1 ill,	I,, ' II III, ,i, ,,  , '! MlTI, , l'|| ,,ll!lll  , i, I	I	    ,',   	I, ', ,1,,'!	Ill
saturated in this area. The system consists of interconnected perforated
HDPE pipes, wrapped in filter fabric and placed around the perimeter of the
LTU.  This drainage system drains water from areas adjacent to the berms
and carries it to the treatment zone above 'me'LW simp where it is removed
by the leachate collection system when it accumulates in excessive amounts.

8X4 Operations Description
  ii,i, ,   , „!', Lir'i'!,,  nil'1' t   	, i  ' i ' ,  ' , ' '.I ...Hi	,i, "i ' 	» , i, 111'1 iii i'1', '  ,  i I in, '!'l|	""i1 '     iV •  ,    " '!"' ihi; ,ir t inril
   The screened contaminated soils from the site that are stored in the waste
pit area undergo a two-step enhanced biodegradation treatment process.  The
first step involves stimulation of biodegradation within the waste pit area by:
(1) adding nutrients (approximately five times during a summer operational
season); (2) tilling twice weekly; and (3) adding bioreactor effluent, fire
pond water, or LTU leachate periodically to maintain a soil moisture level in
the tilling zone of 8.5% by weight. This pretreatment is used to reduce ini-
tial contaminant levels for subsequent treatment in the prepared-bed, lined
LTU.  Soil samples are collected and analyzed periodically to monitor mois-
ture levels hi the waste pit area  and to estimate moisture requirements for the
LTU.  No formal monitoring program is conducted to evaluate the effect of
the pretreatment process on contaminant biodegradation rates in the soils in
the waste pit area (Piotrowski et al. 1994).
   The second step in the treatment process involves placement and manage-
ment of the soils from the waste pit area in the  two LTU treatment cells,
which also serve as the final disposal location for the soils. Contaminated
soils are placed in the LTU cells in 15- to 30-cm (6- to 12-in.) lifts for treat-
ment during the summer. Moisture is applied to the LTU to maintain ad-
equate moisture levels (approximately 40% to 70% of field capacity) in the
treatment zone as well as for dust control.  Additional lifts are placed on the
LTUs when the total carcinogenic PAH and PCP concentrations in the treat-
ment zone for the preceding lift are at or below target remediation levels.
   yield capacity is measured in the field at least once per lift to assess the
moisture-holding capacity of the soils and to define the target moisture levels
corresponding to the desired 40% to 70% field capacity  range.  Field capac-
ity is determined by wetting a small area (approximately 1.5 m by 1.5 m (5 ft
by 5 ft) with fire pond water, covering the area with a plastic sheet, and wait-
ing 48 hours before a sample is collected. For example, if the capacity as
measured in the field is approximately 13% by weight, the soil moisture
content in me LTU should be maintained at approximately 5% to 9% by
weight.  Weekly laboratory moisture  measurements are made to calculate the
amount of water to be applied to the LTU to reach the target level. To esti-
mate moisture needs between weekly laboratory measurements, field

                                8.114
-------
                                                            Chapter 8
observations are made consisting of: (1) daily visual observations of the
moisture content of the surface soils; (2) the soil moisture profile with depth;
and (3) the amount of dust generated from the LTU operations.  Additional
moisture applied to the LTU is usually added immediately prior to tilling to
control the generation of dust during tilling. The LTU cells are actively
managed from approximately March to October each year.
   Water sources for irrigation include the fire pond, bioreactor effluent, and
LTU sump leachate. The water is applied manually to the LTU using a fire
hose connected to irrigation piping that is located around the perimeter of
the LTU. The application of water to the LTU depends on soil moisture
levels in the treatment zone. Water application is limited during high mois-
ture periods to minimize the volume of leachate produced in the LTU sump.
   Nutrients (inorganic forms of nitrogen and phosphorus, usually ammo-
nium sulfate and ammonium phosphate) are added to the LTU by dissolving
them in water applied to the LTU or by fertilizers applied directly to the
LTU. The amounts of nutrients added depends on nutrient requirements for
optimizing biodegradation and the amount of total organic carbon (TOC),
nitrogen, and phosphorus already existing in the soil.  The nutrient require-
ment used for bioremediation optimization was selected as a carbonrnitrogen
ratio in the soils of approximately 12-30:1 and a nitrogen:phosphorus ratio
of approximately 10:1. The soil in the treatment zone is monitored periodi-
cally for TOC, total Kjeldahl nitrogen (TKN), and total phosphorus to esti-
mate concentrations already existing in the soils. The amount of nitrogen
and phosphorus to be added is estimated by subtracting the amounts of nitro-
gen and phosphorus existing in the soil from the nutrient requirement, and
multiplying the remaining concentration by the estimated weight of the lift
being treated. Because a significant percent of carbon measured by the TOC
analysis may exist as ash or other unavailable forms of carbon, the use of
TOC to estimate the amount of carbon in the soil should result in a larger
amount of nutrients being applied to the LTU than that needed to enhance
degradation. Nutrients are added as frequently as every other day depending
on soil moisture and nutrient needs.
                                     i
   To enhance microbial activity by oxygenating the soils, the entire LTU is
tilled frequently (at least weekly, if possible, but dependent on weather con-
ditions) using a tractor-mounted rototiller or similar-type equipment. If the
LTU contains ponded water after storms, tilling is suspended until the soil
dries sufficiently for tilling.  Deep tilling to  promote biodegradation is used
occasionally in the LTU if deeper soils with contaminant concentrations
above the target remediation levels are detected. Deep tilling is continued
until the lower zone has contaminant levels  at or below the target

                                8.115
-------
                                Case Studies
'I'll, :iW

i. 3
                                                                               I
                                remediation levels specified in the ROD.  If contaminants consistently mi-
                                grate into underlying lifts, operation procedures can be modified by: (1)
                                applying smaller lifts; (2) increasing tilling frequency; or (3) reducing mois-
                                ture application.
                                  Operation of the LTU, including the application  of soil lifts", is discontin-
                                ued during winter months. New lifts are not loaded near the end of the treat-
                                ment year if contaminant levels are not expected to decrease substantially
                                before the operation of the LTU is discontinued for the winter.
                                  After all contaminated soils have been treated in the LTU, a protective
                                cover will be installed and maintained over the total 8,100-m2 (2-acre) treat-
                                ment unit to minimize surface infiltration, erosion, and direct contact.
 8.8.5 Routine Process Monitoring Procedures
   The US EPA-mandated monitoring program involves periodic collection
 and analyses of leachate, soil, groundwater, and air samples both outside and
 within the treatment cells during operation and closure periods (Woodward-
 Clyde Consultants 1992). Post-closure care will include monitoring and
 inspection following placement of the cap at the end of the closure period.
                      "          .     	   . "" ,i    .' '        '    ,. ,'	i
 8.8.5.1  Monitoring Outside the LTU
   Monitoring systems in the vicinity of the LTU include groundwater and
 air sample collection systems. Background samples were collected before
 operation of the system began and analyzed to establish a baseline for evalu-
 ating data collected during LTU operation.
   The groundwater monitoring system includes six wells (three
 downgradient, two upgradient, and one midway between the LTU and the
 waste pit area). The groundwater wells around the LTU are monitored
 semi-annually. Samples are analyzed at an off-site laboratory.
   Periodic ambient air monitoring is conducted to: (1) characterize emis-
 sions that may be released to the atmosphere from operations at the unit; and
 (2) quantify ambient concentrations of the compounds to protect the health
 of the workers at the LTU. Dust is expected to be the principal contaminant
 of concern for the workers at the LTU. Dust is controlled primarily by ap-
plying moisture to the LTU before tilling. If, during tilling operations, dust
generation becomes visible, additional moisture is applied to the LTU to
 suppress the dust and/or tilling operations are discontinued until weather
conditions change such that dust generation is reduced
                                                               8.116
             Jit;
-------
                                                           Chapter 8
   Air quality parameters measured include gaseous and particulate PAH and
PCP constituents. Air monitoring is performed shortly after loading during
the day that initial tilling occurs. This initial sampling event for a lift pro-
vides worst-case data for emissions from the LTU because the highest con-
centrations should be present in the soils at that time.  Two air samples are
collected at 1.5 m (5 ft) above ground level on the berms adjacent to the LTU
.to monitor for contaminant migration. One sample station is placed directly
upwind of the LTU, and a second sample station is placed directly downwind
of the LTU so that both downwind and upwind air quality data are obtained.
To place the samplers at the appropriate locations, the prevailing wind direc-
tion is monitored at an on-site meteorological monitoring station every two
hours .during the sampling period.  If the wind direction changes during any
of the 2-hour periods, the sampling stations are rotated so that they continue
to monitor the upwind and downwind air quality of the LTU. The total sam-
pling time of each collection period is approximately 6 hours.  One  sample is
collected for PAH and PCP analysis at two locations (upwind and down-
wind) around the LTU to provide four samples (two PAH and two PCP) per
collection period. A duplicate sample for PAH and PCP is collected from
the berm expected to be the downgradient berm prior to tilling. This dupli-
cate sample and one field blank, as well as the air samples collected from the
monitors during tilling, are analyzed by an off-site laboratory for PAH com-
pounds and PCP. Concurrent with ambient air sampling, 1-hour values for
wind speed, wind direction, sigma 0 (standard deviation of wind direction),
and temperature are recorded at the on-site meteorological monitoring sta-
tion.  These data are used to evaluate air dispersion characteristics (i.e., at-
mospheric stability) during air monitoring.
   Concentrations of PAH compounds and PCP measured in the air samples
are compared to concentrations used for modeling in the No-Migration Peti-
tion (Woodward-Clyde Consultants 1989d) to determine if acceptable con-
centrations are present. More frequent samples are collected if the concen-
trations are unexpectedly high.  If, after several years of monitoring, consis-
tently low concentrations of PCP and PAH compounds are measured from
the LTU, air monitoring will be discontinued upon approval by the regula-
tory agencies.
                                     I
8.8.5.2 Monitoring Within the LTU  ]
                                     i
   Monitoring systems within the LTU include systems to collect soil and
leachate samples. Soil monitoring involves collecting, compositing, and
analyzing of soil samples. Three types of soil  samples are periodically col-
lected from the LTU: (1) operational, (2) confirmation, and (3) compliance.
                                8.117
-------
            Case Studies
    T"!	  '" j.
!	Illhr /  III! Ill I
                                  Operational samples are collected periodically to monitor the degradation
                               of PCP and PAH compounds in the soils during the treatment of each con-
                               taminated soil lift.  These samples are analyzed by the on-site laboratory and
                               are lised primarily for making operational decisions for the LTU. The
                               samples consist of composite samples from each quadrant of the two LTU
                               cells. Each sample is composited from four randomly-selected individual
                               samples from within each quadrant. After the initial tilling of a
                               newly-placed lift of contaminated soil in the LTU cells and application of
                               soil moisture, composited soil samples from the uppermost lift of contami-
                               nated soil are collected from the four quadrants of each cell representing the
                               lift and are analyzed for PCP and MH compounds. Results from this sam-
                               pling are used to characterize baseline contaminant levels and identify degra-
                               dation rates in the LTU cells. After initial placement of a lift, samples are
                               usually collected every two to five weeks. The sampling frequency is in-
                               creased (up to one sample per week or less) when the soil concentrations in
                               the treatment zone are near target remediation levels.  Sample collection and
                               analyses continue until samples from each quadrant are at or below the target
                               remediation levels. After target levels are reached, a new lift may be loaded ,
1     I      '  ;•"' '         ' >        .'",." l!_J__,_   "-  "  . "!  '  "' • " '-"" .  '  .*  " 	-  •'• 	i'. '. •''' i ;' "I11        • .'       '  " i •	i	
                               oil the LTU.                                                 	
                                  After target remediation levels are achieved in the uppermost lift, an addi-.
                               tional sample is collected in the next lowest lift to evaluate the potential for
                               vertical migration of contaminants during treatment. The sample is collected
                               under the quadrant of the uppermost lift that contained the highest level of
                               contamination during treatment. The sample consists of a four-point com-
                               pdsite sample, collected in the same manner as the other operational
                               samples,  the sample is analyzed for PCP and PAH compounds by the
                               on-site laboratory. If PCP and PAH compounds exceed the target
                               remediation levels, deep tilling of the two top lifts is performed until target
                               remediation levels are achieved.
                                  Other operational samples collected and analyzed include: (1)
                               composited soil samples from the four soil quadrants to measure of TOC,
                               TKN, and total phosphorus immediately after the placement of a lift to as-
                               sess nutrient requirements; and (2) weekly soil moisture measurements to
                               determine water application rates.
                                •« | > ,,  M  „' ' i",,'"	      ,    ' ,     ,.,,,„ ,„ i'  i.|   .• | »  , „ i  i|  •.   "   •      ,r:   'M,," „ | '",]	•„,
                                  After the results of the operational samples have indicated that target
                               contaminant remediation levels have been reached, a confirmation sample is
                               collected from each quadrant from each lift treated and submitted to an
                              it  '  '".•  ••	i, ,1;   1	•• I  ".	, ,>	       ..   i , . 	,,,!  ,, ,. i ,, ..,.	„-	   f ,  i ..,
                               off-site laboratory for confirmation analyses.  These samples may be split
                               samples from the last operational sample analyzed or collected separately.
                               Each confirmation sample is a composite sample. Confirmation sampling
-------
                                                            Chapters
results are used to demonstrate that target remediation levels have been
achieved and to establish the validity of operational sampling data. These
sampling data are provided to the regulatory agency after receipt from the
off-site laboratory.
   Compliance samples are used to demonstrate that target remediation lev-
els have been met.  These samples consist of previously-collected confirma-
tion samples (if the concentrations are at or below target remediation levels),
or additional samples may be collected, if required. There are two types of
compliance samples. The first type is a single-lift compliance sample. A
minimum of four single-lift compliance samples (one from each quadrant) is
obtained from each treated soil lift to demonstrate that target remediation
levels established in the ROD for PCP, carcinogenic PAH compounds, naph-
thalene, phenanthrene, and pyrene have been met. Typically, these samples
consist of previously-collected confirmation samples. The second type of
compliance sample is a 3-lift sample for dioxin analysis.  Four 3-lift dioxin
compliance samples are collected from each LTU cell, one from each quad-
rant. The samples consist of soils collected from the full vertical interval of
the three lifts applied during that treatment interval. For quality assurance/
quality control (QA/QC) purposes, one duplicate split sample is collected for
every four samples collected. The dioxin analysis is performed by an
off-site laboratory.  If dioxin is detected above the target remediation level  in
the three-lift compliance samples, an evaluation will be made to determine
how to best address remedial goals. The results of the evaluation will be
presented to the regulatory agency for approval.
   The compliance samples analyzed by the off-site laboratory that meet the
quality requirements outlined in the Quality Assurance Project Plan (QAPP)
(Woodward-Clyde Consultants 1989a) are used to evaluate whether target
remediation levels are achieved.  Contaminant concentrations measured in
each of the four samples (one from each quadrant) for each lift are compared
to the target remediation levels specified in the ROD.  Only when the con-
taminant concentrations for all four samples (not just the mean of the four
samples) are at or below the target remediation levels for total carcinogenic
PAH compounds, naphthalene, phenanthrene, pyrene, and PCP is that lift
considered remediated.
   At the end of each treatment year, additional composite samples from
each lift placed in the unit during that year are collected and analyzed. If
target contamination remediation levels are  not met by the end of a given
year for the composite samples of all lifts placed in an LTU cell that year,
LTU operations are continued in the spring of the next year. No additional
lifts are placed in that LTU cell until such levels are achieved.

                                8.119
-------
                                Case Studies
                                   Leachate monitoring involves collection of samples from LTU sumps on a
                                quarterly basis and whenever leachate is produced during rainfall events.
                                Samples are analyzed for PCP and PAH compounds by the on-site labora-
                                tory. If a visible oil phase is present in the sample, the sample is sent to an
                                off-site laboratory for dioxin analysis. One duplicate sample is sent to the
                                off-site laboratory annually for PAH and PdP analyses as a QA/QC measure
                                for the on-site laboratory. If contaminant concentrations in the leachate are
                                below US EPA detection limits for a period of 1 year* leachate monitoring
                                wilibe reduced to twice annually.  Sump sampling may not be possible if
                                little or no leachate is recovered from the sump during a quarter. The sumps
                                (one from each cell) are monitored each work day during operation of the
                                LTU to evaluate if the leachate collection system is operating properly.
                                   A quality assurance program, QAPP (Woodwani-Clyde Consultants
                                1989a), was developed and is used to ensure appropriate analytical evalua-
                                tion of soil, leachate, groundwater, and air samples.
                                8.8.6  Results of Monitoring Activities
                                  Results of the US EPA-mandated monitoring program conducted by the
                                site operators have shown that lift treatment times vary.  Through 1992, the
                                time required for active treatment of the soil contaminants to reach target
                                remediation levels ranged from 32 to 163 days. The time requirement for a
                                specific lift depended on initial concentrations of the contaminants in the lift,
                                the time of year lift treatment was begun, rates of biodegradation achievable
                                for the organic contaminants of concern, and climatic characteristics (tem-
                                perature and precipitation) during lift treatment (Piotrowski et al. 1994).
                                  The length of the time required for active treatment has generally been
                                determined by the biodegradation rate of pyrene.  Pyrene has been the most
                                recalcitrant of the target contaminants with respect to reaching its remedial
                                goal of 7.3 mg/kg as defined in the ROD. In 1991, pyrene levels in. both
                                LTU cells had not reached target levels before the onset of winter, so treat-
1"' 11.!          ;  '     "      .   I.. I    ,    .:,  ,  •  •  ,i'"   ' „- !•',<	Ill,    °	 , 	 ,	    ,  . , 	
                                ment had to be continued during spring 1992 before additional lifts could be
                                added. From 1989 to 1992, pyrene required an average of 92 days of active
                                treatment (i.e., periods when tilling and irrigation are performed, but not
                                including periods when active treatment is suspended for the winter) to reach
                                target remediation levels; carcinogenic PAH compounds required an average
                                of 50 days; and PCP required an average of 43 days.  Due to these
                                longer-than-expected times required for active treatment, only one to two
                                lifts per year have been added to the LTU cells rather than one lift every 45
                                days as designed. Originally predicted to require 8 to 10 years for cleanup,
                                                                8.120
-------
                                                           Chapter 8
treatment of the total volume of contaminated soil may require as long as 30
to 40 years at the current rate of application (Hurst 1996).
   In cooperation with the US EPA Robert S. Kerr Environmental Research
Center and Champion International, Utah State University (USU) conducted
a comprehensive field evaluation of the prepared-bed land treatment system
at the Libby Site as part of the US EPA Bioremediation Field Initiative (BFI)
(Sims et al. 1993; Sims et al. 1994; Huling et al. 1995a, 1995b; Sims et al.
1995a, 1995b, 1995c). The BFI was established by the US EPA in 1990 to
aid in the development of bioremediation as an effective remediation tech-
nology. An objective of the BFI was to obtain and disseminate field-based
data and information from field experiences concerning the implementation
and performance of bioremediation techniques.
   Results of the USU study indicated that statistically significant decreases
(o=0.05) in PCP and PAH compounds occurred at field scale as determined
by both composite (routinely used for compliance monitoring) and discrete
(used in the field performance evaluation) soil sampling (Sims et al.  1995a,
1995b, 1995c; Huling et al. 1995a, 1995b). Detoxification of the contami-
nated soil, as measured by the Microtox™ assay, also occurred in the same
time frame as the degradation of the contaminants. No increase hi toxicity
hi lower lifts was observed when highly-contaminated soil was applied over
lifts that had previously undergone active itreatment. This indicated that any
vertical migration of water-soluble contaminants from the contaminated lifts
had no negative effect on microbial activity in the underlying treated soil.
   Additional studies at USU investigated potential design and management
criteria to accomplish faster remediation of the contaminated soils in the
prepared-bed land treatment system (Hurst et al. 1995a; Hurst 1996). The
studies focused on management of oxygen at depth in the LTU. If an oxy-
gen concentration of 2% by volume could be maintained in the soil atmo-
sphere, biodegradation of PAH compounds and PCP was shown to continue.
Therefore, continued treatment of soils in buried lifts at oxygen levels above
2% by volume may permit an increase in frequency of lift placement. With
adequate oxygen, a new lift of contaminated soil could be placed in the pre-
pared bed before concentrations of PAH compounds and PCP in the previous
lift reached target remediation levels.  Methods to maintain adequate oxygen
levels in the lower layers require additional investigation.

8.8.7 Future Closure and Post-Closure Activities
   Closure of the LTU cells will commence following treatment of all the
contaminated soil stored in the waste pit airea and completion of standard
                                8.121
-------
Case Studies
LTU operations for both cells.  Closure activities will be designed, to provide
long-term containment of disposed materials and protection of the environ-
ment. f he LTU will be closed in a manner that:  (1) minimizes the need for
further maintenance and (2) controls, minimizes and eliminates, to the extent
necessary to protect human health and the environment, the post-closure
esgape of hazardous waste, hazardous waste constituents, leachate, contami-
nated runoff, or waste decomposition products to the grburidwater, the sur-
face water, or the atmosphere.  The closed facility will be monitored to de-
tect any contaminant releases.
   Criteria that will be used to initiate closure activities include:  (1) little or no
evidence of movement of organics beneath the treatment zone, (2) achievement
of soil target remediation levels  for all constituents, and (3) no detection of
regulated contaminants in i leachate samples for at least the last 2 years of facility
operation. A regular program for tilling, watering, and maintaining the land
treatment area will be conducted until the above criteria are met
   Grpundwater monitoring will be continued through the closure period.
Upgradient wells (two wells) and downgraclient wells (four wells) will be
sampled semi-annually for target constituents.
   Random fill will be placed over the treated soil to prepare a minimum
grade from the crown of the LTU to the exterior berms (Figure 8.42).  A
30-cm (12-in.) thick compacted soil layer will be placed above the random
fill. A 3p-mil HDPE geomembrane liner will be placed over the soil layer,
and a 30-cm (12-in.) thick layer of cover sand in a single lift will be added to
protect the geomembrane. A 30-cm (1 2-in5 thick layer of topsoil will be
placed above the cover sand. The topsoil will be used to establish a vegeta-
tive cover of native plants to prevent erosion of the cover system.
   The side-slope of the cover system will consist of a drainage net (Tensar
DN-3 geonet) between two Typar 3501 filter fabrics placed above the  exte-
rior of the incremental and LTU-containmerit berm graded to a 3: 1 (horizon-
tal to vertical) slope.  A 30-cm (12-in.) thick layer or" cover sand will be
placed above the drainage net and filter fabrics; a 30-em (12-in.) thick layer
of topsoil will be placed above the cover sand. A vegetative cover will be
planted in the topsoil layer.
   It has been proposed that if Criterion (3) above is met at the end of the
LTU operational period, the following actions will be taken: (1) the cover
system will be redesigned to exclude the  flexible geomembrane and drainage
net; and (2) the HDPE liner at the bottom of the LTU cells will be punctured
in the sump areas, allowing the leachate to gravity drain to avoid any "bath-
tub" water accumulation in the system.
                                 8.122
-------
                                                                         Figure 8.42
                                       LTD Waste Placement and Final Cover Designs Used at the Libby Site
                                                                       (not to scale)
po

fo
Co
                                                                                         Drainage Net Sandwiched
                                                                                         Between Two Layers of
                                                                                         Geotextile Filter Fabric   f 1 ft Topsoil with Vegetative Cover
                                                                                                                    1ft Cover Sand
                                                                                                                      2 ft Min. Freeboard
                                                                                                                                 Incremental Berms
                          Liner and Leachate
                          Collection System
                                                                         Gravel Drain
                              Approximate Seasonal
                            High GfGUfiuwaicF BicvauOu
                                                      2112.5ft
 Berm Detail
                              Berm Fill

                      Anchor Trench Backfill
Treatment Zone	

                                                         18 in. Thick Soil Liner'
                                                                                     In Situ Soil
          30 in.

£-Geotextile Filter
 VFabric

    Drainage Net
  60-mile HDPE
  Geomembrane
                     \   Tlrai
                       \60-mili
                                      -2,140


                                      - 2,130


                                      - 2,120



                                      -  2,110


                                      -2,100


                                      -  2,090

                                      Elev. (ft)
                                                                            o
                                                                            Q
      Source: Simsetal. 1995a
                                                                                                                                                                  CD
-------
 Case Studies
   Final grading of the facility will maintain the berms surrounding the
 LTU at a sufficient height to control run-on, runoff, and wind dispersal.
 The berm will be raised with uncontaminated fill to contain the 25'year,
 24-hour storm event.
                                              i
   Post-closure care will continue for at least 5 years, to be terminated after
 the fifth year if target constituents in the soils, groundwater, and leachate are
 not detected above target remediation levels. The post-closure care period
 may be extended to 30 years if significant concentrations of target constitu-
 ents are detected.
                                              •j
   Primary post-closure activities include continued inspection and mainte-
 nance of the facility. The vegetative cover, run-on/runoff control system, and
 LTU sumps will be inspected on a monthly basis or after any major storm.
 The vegetative cover will be selected to adapt to the climate at the site; there-
 fore, after the first post-closure year, irrigation will be discontinued. Access
 to the facility will be restricted by the company security system and fencing
 around the site.
8.8.8 Costs
   Construction of the two LTUs at the Libby Site cost approximately
$400,000. In 1992, the annual operation and maintenance costs of the land
treatment system were estimated to be $117,000.
                                              i                     '
                                              \
8.8.9 Lessons Learned
             1 „  ,  I,1!'   '   ' ',  • „   , ,• .I'lffi1 ;Ui« ' „!"!,!" , :  « " 'i I	»• I • , "'•, M1 ,,'.  „ , • !   " ' .    .• Ml" ,i	!!.'
   The remedial action design at the Libby Site was based on known
biological principles and was demonstrated at the  site to be effective in
permanently destroying the contaminants of concern. The technology
was proven during a pilot study conducted at the site prior to completion
of the feasibility study. The success of trie pilot study, conducted prior
to the ROD (US EPA 1988c) provided federal and state regulators with
the confidence needed to approve the use of bioremediation. Full-scale
facilities were constructed and operated as LTDUs during the first 2
years after the ROD.  The demonstration provided information required
to develop reliable and cost-effective remedial designs.
                                8.124
-------
                             Appendix A
  CONTAMINANTS COMMONLY
    FOUND AT SUPERFUND SITES
A,l  List of Contaminants
A,2  Property Ratings of Chemical Classes
A.3  Properties
                 A.1
-------
                                                                                                                                                             !';  I
f-
.''-=«" : • ' - • '


Table A.1
Contaminants Commonly Found at

Halogenated Volatile Organics
Liquid Solvents
Carbon Tetrachloride
-• Chlorobenzene
Chloroform
Cis-l,2-dichloroethylene (d)
1,1-Dichloroethane (a)
1 ,2-Dichloroethane
1,1-Dichloroethylene
, 1,2-Dichloropropane (a)
Ethylene Dibromide (g)
Methylene Chloride
1,1,2,2-Tetrachloroethane
Tetrachloroethylene

Trans-l,2-dichloroethylene (d)
! 1,1,1-Trichloroethane

Non-Halogenated Volatile Organics
Kctoncs/Furans
Methyl Ethyl Ketone
4-Methy!-2-Pentanone
Tetrahydrofuran
Aromatics
Benzene (g)
Ethyl Benzene (g)
Styrene
Toluene (g)
m-Xylene (g)
o-Xylene (g)
p-Xylene (g)





Halogenated Semivolatile
KB* (b)
Aroclor 1242
Aroclor 1254
Aroclor 1260

Pesticides
Chlordane
DDD
DDE
DDT
Dieldrin

Chlorinated Benzenes

1,2-Dichlorobenzene
1,4-Dichlorobenzene

Superfund Sites

Non-Halogenated Semivolatile
Organics Organics Inorganics
PAHs (e) Arsenic (As)
Acenaphthene Cadmium (Cd)
Anthracene Chromium (Cr)
Benzo(a)anthracene Cyanide (CN) •;
Benzo(a)pyrene Lead (Pb)
Benzo(b)fluoranthene Mercury (Hg)

Benzo(ghi}perylene Selenium (Se)
Benzo(k)fluoranthene jron (ps) $
Chrysene
Dibenz(a,h)anthracene
Fluoranthene
Fluorene
Indeno{l ,2,3-cd)pyrene

2-Methyl naphthalene _
Naphthalene
TJ .'
T
Q.
5<" 4
-t
If
V
.----- -i
*
"i.
•\
•*" - "

^ I
C''- M
_rl
y : I
_ . ._ ^_
ij

j

3

",r. : :-; 8
-------
       1,1,2-Trichloroethane

       Trichloroethylene



       Gases

       Chloroethane

       Vinyl Chloride
Chlorinated Phenols

Pentachlorophenol (w)

2,3,4,6-Tetrachlorophenol
Phenanthrene

Pyrene



Non-Chlorinated Phenols

m-Cresol (e)

o-Cresol (e)

p-Cresol (e)

2,4-Dimethylphenol (e)

2,4-Dinitrophenol

Phenol
        (a) = may be component of antiknock fluids added to fuel oils
        (b) = constituent in some oils, greases, dielectric liquids, and thermostatic fluids
 J>     (d) = may be present in dye or lacquer solutions
 Co     (e) = constituent of crude oil fractions (including fuel and motor oils) and/or coal tar fractions (including creosote); creosote may be present as DNAPL
        (g) = constituent in fuel oils (e.g., gasoline)
	   (w) = combined with fuel oil #2 or kerosene when used as wood preservative
        NOTE:  Some contaminants listed may be present in subsurface as biological or chemical degradation products of others
                                                                                                                                                                                             TJ
                                                                                                                                                                                              (D
                                                                                                                                                                                              x
-------
I  ii



Property Ratings of


Chemical Gass
Halogenated Volatile
Liquid Solvents*

Gases


Melting
Point
Organics
low

low


Water
Solubility

moderate/
high.
high

Chemical


Vapor
Pressure

high
•
high

Classes

Henry's
Law
Constant

moderate/
high
high
Table A.2
Commonly Found


Dynamic
Density Viscosity

high t

low NA



at Superfund Sites (from


Kinematic
Viscosity

t

NA


£

toff/
moderate
low


Log

low/
moderate
low

Table A.1)

Aerobic
Biodegrad-
ability

t

ND



Potential
Subsurface
Mobility

moderate/
high
high
Nonhalogenated Volatile Organics
Ketones/furans
Aromatics
>
low
low

high
moderate/
high
high

high
moderate
high

low low
low moderate

moderate
moderate/
high
low
moderate

low
moderate

ND
high

high
moderate

** Halogenated Semivolatile Orgaaics*
PCBs
', ' Pesticides

Chlorinated Benzenes

" " Chlorinated Phenols
low
high

low/
moderate
moderate/
high
low
moderate

moderate

moderate
low
low

moderate

low51
moderate
low/
moderate
high

tow"
high M>
low/ NA.
high
high high

high NA
ND
NA.

high

NA
high
high

moderate

high
high
high

moderate

Mghp
low
low

high

Wghp
low
low

moderate

low
: , Non-Halogenated Semivolatile Organics
PAHs

Non-Chlorinated
Phenols
Inorganics
Se, As, CN Cr (VI)

\ ]' Hg, Pb, Cd, Cr (HI)
moderate/
high
moderate





low/
moderate
high



For detailed

moderate/
low
moderate/
low


t

low/
moderate


high NA

high high/
m.


NA

high/
m


high

low



high

low



moderate

high



low

high


high**
information on subsurface transport and fate behavior for these chemicals, see Table A.3.







low**
TJ
(D
Q.
x'
>

























-------
Qualitative Rating Key
Rating
Low
Moderate

High
Melting
Point (°C)
£13.00
>13.00
£100.00
>100.00
Water
Solubility
(mg/L)
£1.00
E+00
>1.00
E+00
<1.00
E+03
Sl.00
Vapor
Pressure
(mmHg)
£1.00
E-03
>1.00
E-03
<1.00
E+00
>1.00
E-KX)
Henry's
Law .
Constant
(atm- Density
nWmol) (g/cc)
£1.00 <1#
E-05
>1.00 =1
&05
£1.00
E-03
>1.00 >1*
&03
Dynamic
Viscosity
(centi-
poise)
<0.6
20.6
,1.0
>1.0
Kinematic
Viscosity
(centi- Log
stokes) K
7 ow
0.8 >35
Aerobic
Log Biodegrad-
KOC ability
£25 very slow
or
negligible
>12 moderate
<3.2
>3^ rapid
Potential
Subsurface
Mobility*
2.2
-------
r  :   s   i   ' f  < II  »;=
: ; ,: ;•:,;•..' ' ...; . / . .-,.•.. 	 :: ": . ''.• ''•. • _",:: . ,
"
Chemical
Halogenated Volatile Organics
Liquid Solvents
Carbon Tetrachloride
O> Chlorobenzene
Chloroform
— Cis-l,2-dichloroethylene(li)
1,1-Dichloroethane00
1 ,2-Dichloroethane
1 ,1-Dichloroethylene
1 ,2-DiehIoropropanew
Ethylene Dibromide
5.7E^)3*'"
1.1E-03*'"
154
E01*")
3.6E-03*'"
3.18 E-04m
257
(g/cc)


159471-11-
1.106'"
1.485'"
l^gdUJ
1.175'"
1^53'"
1.214'"
1.158'"
Z172'"
1.325'"
Kine-
Dynamic* matic*
Viscosity Viscosity
(cp) (cs)


0.965'" 0.605'c'
0.756 '" 0.683(c)
0563 '" 0.379^
0.467 m 0.364(c)
0377'" 032l(c>
OS4nl a67^c^
033"1 057 W
0.84111 0.72 (c)
l^e21*" 0.79ra
0.43™ 0.324^
Sites
Log


Z83"!
184"]
157'»
1.86'«
1.79"]
1.48"]
2.13"1
Z02"1
1.76"1
125"1

Log


2.&4"1
Z2'"
!.«"]
15'"
1.48"1
1.15 '"
1.81"1
1.71"1
1.45"1
054"1

Aerobic
Biodegrad- MCL(17)
ability (|lg/L)

'
DPi 5(0
W.A10" 100^ '_
Ara m
Bra ^ -
Ara nd
B" &
fJS -y(0
fJS j(p)
id o.05(!>)
pPl 5® «i i
Appendix A


; ».

- —

_- _



™
-------
1 , 1 ,2,2-TetrachIoroethane
Tetrachloroethylene
Trans-l,2-dichloroethylene(d)
1,1,1-Trichloroethane
1,1,2-TrichIoroethane
Trichloroethylene
Gases
Chloroethane (b.p. 12.5 C)
Vinyl Chloride (b.p. -13.9 C)
-43[7)
-22.7 m
-50m
.32 171
-36-
.87171

-1383 (7)
-157 m
2.9EM)3ra
1.5EH-0201
63E-K)3DI
9.5E+02111
4.5 £+03^
lEt03nl

5.7E+03111
1.1E+031'1
4.9E+00™
1.4E-t01nl
2.65E+02'35
1 E+02™
1.88E+01131
5.87 E+0111'

1E+031'1
2.3E4031'1
5E-04*'"
Z27
E-02*"'
6.6 &03*111
2.76
1.17
E-03*1"1
8.92

1.1 &021'1
6.95&01"1
1.600'1' 1.7?'" 110^°^ 239'1' 234 ''' N'2' nl
1.625111 0.8901 054^c^ 3-141" Z82U1 Ara 5*'
1^57 [1) 0.404 m 0.32l(c) Z091'1 1.77™ B^ T0(f}
1325 tl] 0.858 ll] 0.647^ Z49W 118™ Cra 200(f)
1.4436^ 0.119131 0.824(c) 2.17ra 1.75113' Cra nd
1.462m 0570[1) 0.390^ Z42ra 2.101'1 Ara 5®

0.9414 oct'1 na na 1.43 f'l 1.17 ni nd id
0.9121 15C[31 na m 0.69 w 031 tn nd 2(f)
Non-Halogenated Volatile Organics
Ketones/furans
Methyl Ethyl Ketone
4-Methyl-2-Pentanone



-86.4P1
-83<7'



2.68 E+05111'
1.9E404P)



7.12 E+01131
1.6E+01P)



Z74
E-04«131



0.805151 0.40P] 0.497^ 029 [17] 0.65 1"] nd id
0.8017 ra 0^848 P1 0.729(c) 12S 138[15] nd id














•o
a
x"
-------
*  :  8
                                                                        '-,
                                                                          ~   .  _'f -

:! ' " " : •• '•- * -' ,

Chemical
Tetrahydrofuran
! ; Annuities
! " Benzene00
i ; ; . >
! 03. Ethyl Benzene<8)
', t
i i
1 ! : Styrene
I \ '. Toluene te)
: ;: m-Xylene®
o-Xylene®
; p-Xylene(s)

Melting
Point
CO
-108^™

5JI"
.94.9717]
-30.6 P)
-95.1 n
-»'"
-25 m
*n
Properties
Water*
Solubility
(mg/L)
3E405*161

1.78E+0311'
1.52E+02111
3E+02ra
5.15E+02[1]
2E+0201
1.7E402P1
u.E«r»
Table A.3 (cont.)
of Contaminants Commonly Found at Superfund Sites
Vapor*
Pressure
(mmHg)
4.56E*01«<'

7.6E+01™
7E+00™
5E+OOra
2.2E+01™
9E+00™
7E+00™
9 E+00™
Henry's Law*
Constant
(atm-
m3/mol)
1.1

5.43
7.9E4)3*t"
2.28 E-03171
6.61
6S1
E-03*[1)
494
7.01
Density*
(g/cc)
0.8892 m

0.8765 l"
0.867 '"
0.9060 [I3]
0.8669™
0.8642*[1]
0.880*™
0.8610*"1
Dynamic*
Viscosity
(cp)
0551131

0.6468™
0.678 P)
0.751 »3)
058™
0.608 [1]
0.802 [1]
0.635 [1]
Kine-
matic*
Viscosity Log
(cs) K^
0.618 0.46'141

0.7379 (c) Z13"
0.782(c> 3.15™
0.829^ 3.16 [14]
0.669(c) 2.73™
0.717 13J 3^20™
0.932131 3.12™
0.753 P3 3.15™
Aerobic
Log Biodegrad- MCLII7]
K,,,. ability (jig/L)
nd nd nd

1J81«™ D*1 &
253™ DS^IO121 700*'
id nd nd
2.41 ™ Dra 2000*'
254™ nl lOOOO®
2.84™ nd lOOOO6*
254™ nd 10000*'
' ' " Halogenated Semivolatile Organics
!
: PCBs®)
rj; Aroclorl242





4.06

3.4E-04™

1385"'

id

nd s^gRl

^Ul 1^12} Ql
Appendix A
• . i



; ~ ~ ~~ ; ;
...!•= i



J ! 'I
-------
Aroclor 1254
Aroclor 1260
Pesticides
Chlordane
DDD
DDE
DDT
Dieldrin
Chlorinated Benzenes
1 ,2-Dichlorobenzene
1,4-Dichiorobenzene 	

10™ 1.2&02™
id 2.7E-031'1

106 tl] 5.6&02*"'
112171 1.60
E4JJ24CP]
88.4 l11 4.0E-02P)
108 m 3.1E-03"1
176.5 '" 1.86
E-01*141

-17 m 1E+021'1


7.71 2.8E-04111 1.538«9] nd nd 6.03 ^ nd NC) nd
E-051"
4.05 3.4E-041'1 1.4430cm nd id 7.15 P) id NI2) nd

1&05"1 2.2&04*"1 1.6*1'1 1.104 P1 0.69(c) 5.481'1 4581'1 Npl 2(p)
1 7.96 1.3851'71 na na 5.56™ 538ll] Mpl nd
E-0630C[17J E-06«U)
6.40E-06111 1.9E-04*111 nd na na 5.691'1 5.41™ Mra nd
1.5E-0701 2.8E-05*"1 0585 [n na na 636 m 5.48 m M(2) nd
1.78 E-07131 9.7E-06*181 1.75 PI na na 5.34^1 32311*1 NPI nd

9.6E-011" 1^8 1.3061'1 1.302m 0.997^ 338[" 3.061'1 T121 600(p>
6&OIM! JJ§ 1^475U! 1.258'" 1 008 W 339 [!i 3.07 !!i T'2' 750(0
E-03*™












Appendix A
-------


!"•; " : '
Chemical
, r Chlorinated Phenols
, - -_ ~ Pentachlorophenol(w)
= 2,3,4,6-Tetrachlorophenol
| >7 Non-Halogenated Semivoiaiile
! 0 PAHs™
i ; ;f >•;:
i " Acenaphthene
! i* ; - . Anthracene
i L =: - -
= '-'- - Benzowanthracene
1 Benzowpyrene
i r
i Benzo(b)fluoranthene
; Benzo(ghi)perylene
I jr
E" '~7 7 Benzo{k)fluoranthene
i ! v ' * ™
hs: - '- ;
-. : L

Melting
Point
CO

190 m
Grganics
925 PI
2163 pl
167 >4'
]79 m
167'43 .
278"2J
217ti2]
/ .- "
Properties
Water1
Solubility
(mg/L)

1.4E-+01m
l.OOE+03*11"
3.88E+00*181
1.4
E^)2*"2'
3.8E-03*'12'
MBO*™
16E-04*171
430
&03ttii]
-" •« -
"•?-• - : •
-; .- - '
^ . . _ *
Table A.3 (cont.) f
of Contaminants Commonly Found at Superfund Sites
Vaporf
Pressure
(mm Hg)

1.1 E-04D1
id
231
1.08
1.16
5.49
E^9*"8'
5.00
E-07«121
1E-10"4'
9J9
E-ll"41
Henry's Law1^
Constant
(atm- Density^
m3/mol) (g/cc)

2.8E-06"1 1.978 "'
id 1-839*|5)
120 1^25 "21
E-03*[c]
338 IJS'12'
EflS^l
4.5E-06"21 1.174"21
1^ nd
E-05«12)
1.19 nd
E-05*1"
534 nd
£-08*""
354 nd
&05ttii]
Kine-
Dynamict matict
Viscosity Viscosity Log
(cp) (cs) K^

na na 5.12 l1'
na na 41^^
na na 352tl2)
na na 4.45 "21
na na 5.61 (I2]
na na 6.061111
na na 657 "2!
na na 651 ""
na na 6.06""
Aerobic
Log Biodegrad- MCL1171 -
K,,,. ability (^g/L)

480111 A121 nd
2.0" n nd nd :
3.7'1" Dra nd
4.1"" Ara nd :
6.14"" N121 nd
6J4"11 nd o.2w ;
5.74"" id nd
62"" nd nd
5/74"" nd nd ^ [
Appendix A

- -



-------
Chrysene
Dibenz(a,h)anthracene
Fluoranthene
Fluoiene
Indeno(l>2t3-cd)pyiene
2-Methyl naphthalene
Naphthalene
Phenanthrene
Pyrene

254171 6&03*P> «3
2665 m 2J5E-03*1121 1E-10«12]
107 M 165 E-02[I01
E-01*"1 &06(l21
H6.7"2' 1.90 E400*"' 6-67
E04 P"
163112) 530 i E-10*"'1
&04«"1
34.58 m Z54E-tQl*W 6-8g
E-02181
802N 3.1E+OI*"2' 2.336
&oi^I2i
100'* 1.18 E+00*'8" 2.01
E04181
150 m 1.48 E-01*181 ^^
E-06 ID)
1.05 1274m na
733 1.252[i2] na
E-08am
65 1^Z52'12' na
&06*"4!
7-65 1.203 t121 na
E-05*^1
655 nd na
&08*"4'
5.06 Loosst'2' na
ErW™
127 1.162"21 na
£•03*"='
3^| 0.9800112' na
120 L271"21 na
Ar_*
E^J5M
« 5.61""
na 6.801"1
na 450n2]
na 4.18 |IZ1
m 65tul
na 3.86 1'2!
na 3301'2'
na 4.46'12'
na 458»21

531-"1 A5J<10121 nd
&52(nl nd nd
45811'1 A5.N1012' nd
35111' Ara nd
(SaH'l nd nd
353 [Mi nj ni
3.11 1'4' DPI id
41 on jjra nd
4.581"1 D5JM10121 nd










Appendix A
-------
,                                                                      i;
: -. _-"".-.-••-•-- --'''

Chemical
Non-Chlorinated Phenols
Phenol
2,4-Dimethylphenol(e)
2,4-Dinitrophenol
XtadW
<*W
p-Cresol(e)
Properties
Melting Water1
Point Solubility
CC) (mg/L)
41 m 8.4E+0401
25 f! 6.2E+03*p]
112171 6E+03*pj
12 m 2.35E-f04ra
31 m 3.1 E+W400171
34_gPl 2.40
Table A.3 (cont.)
of Contaminants Commonly Found at Superfund Sites
Vaport
Pressure
(mm Hg)
5.293
E-01!IJ
9.8E-02*p]
1.49
E05*"1
J-5!
Z45
&01*"2'
E-01*"2'
Henry's Law* Kine-
Constant Dynamic* matic* Aerobic
(aim- Density* Viscosity Viscosity Log Log Biodegrad- MCLfl71
m3/mol) (g/cc) (cp) (cs) K,,,, ^ ability ftig/L)
780E-07[cl 10S7641CI11 ^fQ50011' aar^ec I4fi01 115 1'1' Dra nd

2^B^J6*'^ 1.036^ na na 250^ 235"4' D^ nd
6.45 1.68^ na na 1.54^ 122'"' D^ ni
3-8 1.038 m 21 "21 28 -c' 156|12) 1.43[15] sd n!
4.7 102731121 na na ISS1121 1.23f151 nl nl
&05*121
3.5 1.0347 ^ na m 1.94 PI 1.28 tisi nj ^
&04*H)
,
	
Appendix A


; /;i, |


![; : f : i « i I
-------
      CHEMICAL
                                                                                                                                                               MCL
Co
        Inorganics
      arsenic (As)
      cadmium
      (Cd)
      chromium
      (Cr)


      cyanide (CN)


      iron (Fe)
      lead (Pb)
      mereuiy (Hg)
      selenium (Se)
May occur in more than one oxidation state in subsurface. Arsenate form (As043*) will dominate under oxidizing conditions. More toxic and mobile
arsemte form (AsO3") may dominate under increasingly reducing and acidic conditions. Volatile alkylated-As compounds may form under  reducing
conditions. Volatile arsine (AsH3) may form under highly reducing conditions. Adsorption of arsenate and arsenite forms will generally increase with
decreasing pH,

Occurs only in divalent form in aqueous solutions (e.g., Cd2*, CdCl* CdSO4°). Cd21- tends to be dominant species.  Adsorption behavior correlates with
cation exchange capacity (CEC) of soil and aquifer material. Adsorption/precipitation increases with increasing pH with most Cd precipitating out at
pH>6.

May occur in more than one oxidation state in subsurface. Trivalent form (Cr IE) is dominant under pH and redox conditions generally present in
subsurface. Cr ffl may be converted to highly mobile and toxic hexavalent form (Cr VI) under oxidizing conditions. Cr in is readily adsorbed in the
subsurface while Cr VI is not

Cyanide ion (CM') predominates in aqueous solution only at pH>9. Hydrogen cyanide (HCN) predominates at pH<9. HCN is volatile (v.p. 741 mm Hg
at 25C) and toxic. CM' behaves similar to halide ions and tends to complex with iron. Undissolved cyanide salts may be present in vadose zone.

May occur in more than one oxidation  state in the subsurface. Ferrous form (Fe2*) is most soluble and mobile, and dominates under reducing
conditions. Under oxidizing conditions, ferrous form is converted to ferric form (Fe3*). Ferric form is less soluble, less mobile, and will tend to
precipitate. Compounds and metals complexed to iron may be removed from solution through the precipitation process. Conversely, compounds and
meiais ausoroea to iron in the subsurface may be increasingly mobilized under increasingly reduced conditions.  Precipitated iron may hinder treatment
processes such as m-situ bioremediation and air stripping.

Dominant species in aqueous solution are  Pb2* under acidic conditions and Pb2*-- carbonate complexes under alkaline conditions. Adsorption behavior
correlates with cation exchange capacity (CEC) of soil and aquifer material. Adsorption/precipitation increases with increasing pH  with most Pb
precipitating out at pH>6.  Volatile alkylated-Pb compounds may be present or may form under reducing conditions.

May occur in more than one oxidation  state. May occur in subsurface in mercuric form (Hg21-), mercnrous form (Hg,4*), elemental form (Hg°), and in
alkylated form (e.g., methyl and ethyl mercury). Hg,2* and Hg2*  are more stable under oxidizing conditions and are strongly adsorbed by soils. Hg° and
alkylated forms are more stable under reducing conditions.  Conversion to alkylated forms may occur under reducing conditions. Hg° and alkylated -
Hg forms are volatile, toxic, and may not be as strongly adsorbed by soils.

May occur in more than one oxidation state in subsurface. Selenate form (Se042-) will dominate under oxidizing conditions. Selenite form (SeO  2) will
dominate under increasingly reducing conditions. Selenide form (Se2~) may dominate under highly reducing conditions. Selenate and selenite are more
soluble and mobile forms. Adsorption of selenate and selenite will generally increase with decreasing pH.  Volatile alkylated-Se compounds may
form under reducing conditions.
                                                                                                                                                             id
100*)
200(t)
300®
                                                                                                                                                             5

                                                                                                                                                                               !
                                                                                                                                                                                 •
-------
        !      !    --  . ,    sli
      j  i      ! !   !|J1  ! 3
                                                                                                                                                                                       (D

                                                                                                                                                                                       Q.
                                                                                                                                                                                       x"
                                 Table  A.3  (cont.)
Properties of Contaminants Commonly Found at Superfund Sites
    A = significant degradation with gradual adaption
    B = slow to moderate activity, concomitant with significant rate of volatilization
    C = very slow biodegradative activity, with long adaption period needed
  '  D = significant degradation with rapid adaption
    M = not significantly degraded under the conditions of the test method
    N = not significantly degraded under the conditions of test method and/or precluded by extensive rate of volatilization
    T = significant degradation with gradual adaption followed by deadaptive process in subsequent subcultures (toxicity)

  ;: (a) = may be component of antiknock fluids added to fuel oils; remedial treatment may require consideration of constituent in oil phase
    (b) = constituent in some oils, greases, dielectric liquids, and thermostatic fluids; remedial treatment may require consideration of constituent in oil phase
    (c) = calculated
  •' (d) = may be present in dye or lacquer solutions; remedial treatment may require consideration of constituent in oil phase
  :  (e) = constituent of crude oil fractions (including fuel oils and motor oils) and/or coal tar fractions (including creosote); creosote may be present as DNAPL; remedial treatment may require
  ','- consideration of constituent in oil phase
    (g)  = constituent in fuel oils (e.g. gasoline); remedial treatment may require consideration of constituent in oil phase
    IP) = proposed MCL
  ; ® = tentative MCL
—: (w) = combined with fuel oil #2 or kerosene when used as wood preservative; remedial treatment may require consideration of constituent in oil phase

    na = not applicable
    nd = no data found

     [ ]  Reference

    t = Values are given at 20'C unless otherwise specified
    *=Value !s at 25'C
    t = Value is at unknown temperature but is assumed to be at 20-30'C
-------
                                                                      Appendix B
                   LIST OF REFERENCES
Acomb, L.J., D. McKay, P. Currier, S.T. Berglund, T.V. Sherhart, and C.V. Benediktsson.  1995.
Neutron probe measurements of air saturation near an air sparging well. In Situ Aeration: Air
Sparging, Bioventing, and Related Remediation Processes. R.E. Hinchee, R.M. Miller, and P.C.
Johnson (eds.). Columbus, OH: Battelle Press.
Adler, T.  1996.  Botanical cleanup crews. Science News. 150:42-43.
AFCEE.  1994. Bioventing Performance and Cost Summary.  Prepared by Parsons Engineering
Science, Inc. for the Air Force Center for Environmental Excellence, San Antonio, TX.
Alexander, M. 1982. Most probable number of methods for microbial populations. Methods of
Soil Analysis, Part 2: Chemical and Microbiological Properties. 2nd edition. American Society
of Agronomy, Inc. Madison, WI:  Soil Science Society of America, Inc.
Alexander, M. 1994. Biodegradation and Bioremediation. San Diego, CA:  Academic Press.
Alexander, M. 1995. How toxic are toxic chemicals in soil? Environ. Sci. Technol 29:2713-2717.
Alexander, M., R.J. Bareye, J.T. Gannon, U. Mingelgirin, and V. Tan. 1991.  Movement of
Bacteria Through Soil and Aquifer Sand. EPA/600/2-91/010. Office of Research and Develop-
ment, Robert S. Kerr Environmental Research Laboratory, Ada, OK.
Allen, E.R. and S. Phatak.  1993. Control of organo-sulfur compound emissions using
biofiltration — methyl mercaptan. Presented at the Air and Waste Management Assoc. 86th
Meeting and Exhibition.  Denver, CO.  June 13-18.
Allen, E.R. and Y. Yang.  1992. Operational parameters for the control of hydrogen sulfide
emissions using biofiltration. Presented at the Air and Waste Management Association 85th
Meeting and Exhibition.  Kansas City, MO. June 21-26.
Anderson, T.A. and B.T. Walton.  1995. Comparative fate of [14C]trichloroethylene in the root
zone of plants from a former solvent disposal site. Environ. Toxicol. Chem.  14:2041-2047.

Anderson, T.A. and J.R. Coats (eds.). 1994. Bioremeidiation through rhizosphere technology.
ACS Symposium Series. Vol.563. Washington, DC: American Chemical Society.
Anderson, T.A. and J.R. Coats.  1995. Screening rhizosphere soil Samples for the ability to mineralize
elevated concentrations of atrazine and metolachlor. J. Environ. Sci. Health. 830:473-484.
Anderson, T.A., E.A. Guthrie, and B.T. Walton. 1993. Bioremediation in the rhizosphere.
Environ. Sci. Technol. 27:2630-2636.
APHA. 1989. Standard Methods for the Examination of Water and Wastewater. 19th edition.
Washington, DC: American Public Health Association.
Aprill, W. and R.C. Sims.  1990. Evaluation of the use of prairie grasses for stimulating
polycyclic aromatic hydrocarbon treatment in soil. Chemosphere. 20:253-265.
                                         B.I
-------
                                            List of References
* f.i*
I '!:}
t: LI I"
111" '<
in,: :«	
;: J*
 Aprill, W., R.C. Sims, J.L. Sims, and I.E. Matthews. 1990. Assessing detoxification and
 degradation of wood preserving and petroleum wastes in contaminated soil.  Waste Management
 and Research. 8:45,65.

 Baker, A., R. Brooks, and R, Reeves.  1988. Growing for gold...and copper... and zinc. New Sci.
 117:44-48.'              		  '	
 Baker, A.J.M. 1995. Metal hyperaccumulation by plants:  our present knowledge of the ecophysi-
 ological phenomenon. Witt Plants Have a Role in Bioremediation? Interdisciplinary Plant Group
 (ed.).  Proceedings/Abstracts of the 14th Annual Symposium, Current Topics in Plant Biochemistry,
 Physiology, and Molecular Biology. University of Missouri-Columbia. April 19-22.
 Baker, R.S., M.E. Hayes, and S.H. Frisbie. 1995? Evidence of preferential vapor flow during in
 situ air sparging. In Situ Aeration: Air Sparging, Bioventing, and Related Remediation Pro-
 cesses. R.EJ Hinchee, R.M. Miller, and P.C. Johnson (eds.).  Columbus, OH: Battelle Press.
  1     „',''",,•'"               '       '' V  ,. i' ' »i|!|i|	H'i i|ii!"'.,  	i|" nil	" !l     ..  „   i  '"   .... .  "	'"iiii	i 	K1	
       '',.',:         " "'      i   "   i        '• ''"'i1 "'!'' 4'''°      I "'I'1 ' I  ' • 	   ''    .,',..     .     "	' n.i  " •
 Banuelos, G.S., H.A. Ajwa, B. Mackey, L.L. Wu, C. Cook, S. Akohoue, andS. Zambrzuski.
 1997. Evaluation of different plant species used for phytoremediation of high soil selenium. J.
 Environ. Qual. 26:639-646.

 Barcelona, MJ. and T.R. Holm. 1991. Oxidation reduction capacities of aquifer solids. Environ.
 Sci.Tecknol.9(25):l565-l512.

 Barcelona, MJ., J.P. Gibb, J.A. Helfrich, and E.E. Garske. 1985. Practical Guide for Ground'^
 Water Sampling.  EPA/600/2 85/104. Office of Researchland Development, Robert s!Kerr
 Environmental Research Laboratory, Ada, OK.
 Barker, J J?., G.C. Patrick, and D. Major. 1987. Natural attentuation of aromatic hydrocarbons in
 a shallow sand aquifer.  Ground Water Monitor. Rev. Winter:64.7i.
 Earth, D.S. and B.J. Mason. 1984. Soil sampling quality assurance and the importance of an
 exploratory study. Environmental Sampling for HazardousWastes. ACS Symposium Series267.
 Washington, DC: American Chemical Society.

 Earth, D.S., B J. Mason, T.H. Starlcs, and K.W. Brown. J989. Soil Sampling Quality Assurance
 User's Guide. 2nd edition. EPA 600/8-89/046 US EPA, Environmental Monitoring Systems
 Laboratory, Las Vegas, NV.

 Bass, D.H. and R.A. Brown. 1995. Performance of air sparging systems— a review of case
 studies. Proc. Petroleum Hydrocarbons and Organic Chemicals in Ground Water: Prevention,
 Detection, and Remediation, Conference and Exposition. November 29-December 1.
                                                          I
 Baud-Grasset, P., S. Baud-Grasset, S.I. Safferman.  1993. Evaluation of the biorernediation of a
 contaminated soil .with phytotoxicity tests. Chemosphere. 26:1365-1374.
          11         ,'  '„       ,,'„,  "  ' MI ,'	     , ',,'	","L v    !r  j,  „,'„'  ,  "   	,„ji    " ,,„'  ''"' .i!,," r'rr  i,:i
 Beausoleil, YJ., J.S. Hunter, G.W. Barker, and S.N. Neumann. 1993.  SPE 26000, the use of air
 sparging in the remediation of a production gas processing facility. Presented at the SPE/EPA
 Exploration & Production Environmental Conference.  San Antonio, TX.  March 7-10.

 Bedient, P.B. and H.S. Rifai. 1993. Modeling in situ bioremediation in in situ bioremediation:
 when does it work?  Report of the National Academy of Sciences Committee on Bioremediation.
 B.E. Rittman et al. (eds.). Washington, DC: National Academy Press.

 Beeman, R.E., J.E. Howell, S.H. Shoemaker, E.A. Salazar, and. J.R. Buttram.  1993. A field
 evaluation of in situ microbial reductive dehalogenation by the biotransformation of chlorinated
 ethenes. Bioremediation of Chlorinated and Polycyclic Aromatic Compounds. R.E. Hinchee,
A. Leeson, L. Semprini, and S.K. Ong (eds.). Ann Arbor, MI: Lewis Publishers.
      "  .   ',i,            .   /   '„''''      ,  ""  , T  '"     i i.,.M i  iii.'   '              .       i  i
Bellin, C.A. and G.A. O'Connor. 1990. Plant uptake of pentachlorophenol from sludge-amended
soils. J. Environ.  Qual.  19:598-602.
                                                                                     EJ.2
     !, .»	ImU'iif.;	iJIW, L	:,|' 'ni J
-------
                                                                         Appendix B
Bergen, A. and M. Levandowsky. 1994. Restoration of intertidal salt marsh using Sporting
altemiflora seedlings and transplants: remediation results from a site heavily impacted by No. 2
heating oil. Abstract: Emerging Technologies in Hazardous Waste Management VI, ACS
Industrial & Engineering Chemistry Division Special Symposium, Volume I.  Atlanta, GA.
September 19-21.
Bianchi-Mosquera, G.C., R.M. Allen-King, and D.D. Mackay. 1994. Enhanced degradation of
dissolved benzene and toluene using a solid oxygen-releasing compound.  Ground Water Monitor
Rev. pp 120-128.
Biotreatment News.  1993,1994. In situ bioremediatiion of groundwater and subsoils at the French
Limited Site, TX. Part I (October, 1993), Part H (December, 1993), Part III (February, 1994).
Washington, DC: Devo Enterprises, Inc.
Bohn, H.  1992.  Consider biofiltration for decontaminating gases. Chemical Engineering
Progress. 4:34-40.
Borden, R.C., C.A. Gomez, and M.T. Becker. 1995.  Geochemical indicators of intrinsic
bioremediation.  Ground Water.  33(2):  180-189.
Boulding, J.R. 1995. Practical Handbook of Soil, Vadose Zone, and Ground Water Contamina-
tion Assessment,  Prevention and Remediation. Boca Raton, FL:  Lewis Publishers.

Bourquin, A.W.  1989. Bioremediation of hazardous ,waste.  Haz. Mat. Con. 2:16.
Bouwer, E.J.  1994.  Bioremediation of chlorinated solvents using alternate electron acceptors.
The Handbook of Bioremediation. R.D. Norris et al. (eds.). Boca Raton, FL: Lewis Publishers.
Bouwer, E.J. and P.L. McCarty.  1984.  Modeling of trace organics biotransformation in the
subsurface. Ground Water Monitor Review. 22:443-440.
Brady, N.C.  1990. The Nature and Property of Soils. 10th edition.  New York, NY:  MacMillan
Publishing Co.
Brooks, R.R., J. Lee, R.D. Reeves, and T. Jaffre. 1977. Detection of nickeliferous rocks by
analysis of herbarium specimens of indicator plants.  ,/. Geochem. Explor.  7:49-57.
Brown, R.A.  1993.  Air Sparging. EPA white paper. Prepared for Office of Solid Waste and
Emergency Response, Washington, DC.
Brown, R.A.  1989.  Oxygen sources for biotechnological application. Paper presented at
Biotechnology Work Group. Monterey,  CA. February 21-23.
Brown, R.A. and F. Jasiulewicz.  1992.  Air sparging: A new model for remediation.  Pollution
Engineering.  24(13):52-55.
Brown, R.A. and R. Fraxedas. 1991. Air sparging — extended volatilization to contaminated
aquifers.  Proc. Symposium on Soil Venting. Robert SI. KBIT Environmental Research Laboratory,
Houston, TX.
                                                 •
Brown, R.A. and R.D. Norris. 1988.  Nutrients for Stimulating Aerobic Bacteria.  U.S. Patent
4,727,031.                                      i
Brown, R.A., C. Herman, and E. Henry.  1991. The uise of aeration in environmental cleanups.
Haztech International Pittsburgh Waste  Conference, David L. Lawrence Convention Center.
Pittsburgh, PA. May 14-16.
Brown, R.A., R.D. Norris, and R.L. Raymond. 1984. Oxygen transport in contaminated aquifers.
Proc. Petroleum Hydrocarbon and Organic Chemicals in Groundwater: Prevention, Detection,
and Restoration.  Houston, TX: National Water Well Association.
                                          B.3
-------
                                         List of References                                 .
                ,  r .HI    »          ;    ,»!    	"    	1	!	'•	-—	;—j	:	--	"	—•—"
        ,   '.",      '	  •  '•    ',      .    :           """:   '    -,.:.•>•         ••   ',	 "i   <"}""  .  ', i  '  (,:•        '•  ••      i1   ,  ',,' ' I".'-.
                                         Brown, S.L., R.L. Chancy, J.S. Angle, and A.J.M. Baker. 1994. Phytoremediation potential of
                                         Thlaspi caeruiescens and Bladder Campion for zinc- and cadmium-contaminated soil. /. Environ.
                                         Qual.  23:1151-1157.
                                         Brubake^G.RMM.S.We^^^
                                         A Guide for Railroad Industry Use ofln-Situ B/orei«erfla//on. Report No. R-857. American
                                         Association of Railroads, Washington, DC.            .                          .
                                         Buscheck, T^E. and K.T. p'Reilly. 1995. Protocol for*Monitoring Intrinsic Sioremediation in
                                         Groundwater  Chevron Research and Technology Company, Health, Environment, and Safety
       i '          , f                ' „ N ,,|| '       ,     ,ii 	, ;„•   ,'     '' "  H"!!, .n1"1 ,       ,   •'•,,,  "',, " '  ' ,  •,'  •    ,!,"'!", "II ,h"        i      '    "   •'  • I "nil,if,
                                         Group. March.
                                         Buscheck, T.E... K.T. O'Reilly, and S.N. Nelson. 1993.  Evaluation of intrinsic bioremediation at
                                         field sites.  Proc. Conference on Petroleum Hydrocarbons and Organic Chemicals in Ground
                                         Water. Houston, TX: National GroundWater Association. November.
                                                                                                     i
                                         CH2M Hill. 1982. Installation Restoration Program Records Search.  Eielson APR, AK.
                                         Champion International Corporation. 1993. LTIJ199 f Annual Operational Report: Groundwa-
                                         ter Site, Libby, Montana. Libby, MT.
                                         Chaney, R.L. 1983. Plant uptake of inorganic waste constituents. Land Treatment of Hazardous
                                         Waste. JJF. Parr, P.B. Marsh, and J.M. Kla (eds.).  Park Ridge, NJ: Noyes Data Corporation.
                                         Chow, V.T., D.R. Maidment, and L.W. Mays. 1988. Applied Hydrology. New York: McGraw-Hill.
 • *     •  ",.      '•:       :".,  ,  ,;i  * "'h    ,,  •    '     .1";,     . '•	v v'i'    !,,   .,   ••••' '••"'  '  ,";'  'W-^,<  -'Vi'Sf1;"!-':.  • \   ••':'''''   "   •,.,   ' 'M1^.'
                                         Clayton, W.S., R.A. Brown, and^D.H. Bass. 1995. Air sparging and bioremediation:  the case for
                                         in situ mixing. In Situ Aeration: Air Sparging, Bioventing, and Related Remediation Processes.
                                         R.E. Hinchee, R.M. Miller, and P.C. Johnson (eds.). Columbus, OH: Battelle Press.
                                         Cookson.J.T. 1995. Bioremediation Engineering:  Design and Application. New York:
  .,   '            „         "    ';,   ';.:',    McGraw-Hill.            ' '        .    '...., 	,'	.'.''.',,.       ,'       ...      , ,  ',,
                  Hi1,,,         [   •  ,""!'   " r       "     '>      .  .          •   '":- -  „•' •<•	 	  ,• „"':• iiti •  ;  .   ',|	i       .  ,  . .    - '.;,    '  -, i
                                         Coover, M.P. and R.C. Sims. 1987. The rate of benzo[a]pyrene apparent loss in a natural and
                                         manure amended clay loam soil. Haz, Waste & Haz. Mat. 4(2):15l-158.
,'  ||,,;||  ,     , | i»      '          '  i i  ' |l    ,||i         ,i.«,   i 'i1    ,   '   f "•	• ,     :, " |  • ,  , ",,1^ i r   „ , „  ' ,,||,,, ,  | "I ;„ Vn ,   ' h '! "•" II ii, i        • ,,      •        i  ,
                                         Corapcioglu, M.Y. and M.A.  Hossain. 1991. Methanogenic  biotransformation of chlorinated
                                         hydrocarbons in ground water. J. Environ. Eng. 117(1).
                                         Cornish, J.E., W.C. Goldberg, R.S. Levine, and J.R. Benemann.  1995.  Phytoremediation of soils
                                         contaminatediwith toxicelements and radionuclides. Bioremediation of Inorganics. R.E.
                                         Hinchee, J.L. Means, and D.R. Bums (eds.).  Columbus, OH: Battelle Press.

                                         Cosgriff, D. 1996. Personal communication with Scott Ruling. Champion International.  Libby, MT.
                                         Cozzarelli,  I.MM R.P. Eganhouse, and MJ. Baedeck'er.  19510. Transformation of monoaromatic
                                         hydrocarbons to organic acids in anoxic groundwater environment. Environ. Geol. Water Sci.
                                         15(2):135-141.
                                         Cozzarelli,  I.M., M.J. Baedecker, R.P. Eganhouse, and D.G. Goerlitz. 1994. The geochemical
                                         evolution of low-molecular-weight organic acids derived from the degradation of petroleum
                                         contaminants in groundwater. GeoctiimicaetCo$mochimicaActa.5B(2):B63-B77.
                                         Cunningham, S.D., W.R. Berti, and J.W. Huang. I995a. Remediation of contaminated soils and
                                         sludges by green plants. Bioremediation of Inorganics. R.E. Hinchee, J.L. Means, and D.R.
                                         Bums (eds.). Columbus, OH: Battelle Press.
                                         Cunningham, S.D., W.R. Berti, and J.W. Huang. 19956! fhytoremediation of contaiminated soils.
                                         Trends in Biotechnology. 13:393-397.
                                         Daniel, J.A., B.S. Newell, and D.H. Kampbell. 1992. Use of dehydrogenase activity as an
                                         indicator of bioacclimation at field sites. Proceedings of the 3rd International Subsurface
                                         Restoration Conference on Ground Water Quality Research,  June 21-24, Dallas, TX.

                                                                                   B.4
-------
                                                                          Appendix B
 Darby, J.L. and D.F. Lawler. 1990. Ripening in depth filtration: effect of particle size on
 removal and head loss. Environ. Sci. Technol.  22(7): 1069-1078.
 Davis, J.W., N.J. Klier, and C.L. Carpenter. 1994. Natural biological attenuation of benzene in
 ground water beneath a manufacturing facility. Ground Water.  32(2):215-226.

 Davis, L.C., N. Muralidharan, V.P. Visser, C. Chaffin,, W.G. Fately, L.E. Erickson, and R.M.
 Hammaker. 1994. Alfalfa plants and associated microorganisms promote biodegradation rather
 than volatilization of organic substances from ground water. Bioremediation Through Rhizo-
 sphere Technology. T.A.. Anderson and J.R. Coats (edls.).  ACS  Symposium Series. Volume 563.
 Washington, DC: American Chemical Society.
 Day, M.J., J.A.M. Thomson, and A.P. O'Hayre. 1993,, In-situ bioremediation of DNAPL-
 impacted groundwater and subsoils using alternate electron acceptors at the French Limited
 Superfund Site, Texas.  Extended Abstract.  Presented at the I&EC Division Special Symposium,
 American Chemical Society Annual Meeting, Atlanta, Georgia.  September 27-29.
 de Bruin, W.P., J.J. Kotterman, M.A. Posthumas, G. Schraa, and A.J.B. Zehnder. 1992.
 Complete biological reductive transformation of tetrachloroethene to ethane. Appl. Environ.
 Micro. 58(6):1996-2000.
 Dec, J. and J.M. Bollag. 1994.  Use of plant material for the decontamination of water polluted
 with phenols. Biotecknol. Bioeng. 44:1132-1139.
 Dey, J.C., R.A. Brown, and W.E. McFarland.  1990.  Integrated site remediation combining
 groundwater treatment, soil vapor recovery, and bioremediation. Proc. llth National Conference,
 Superfund. Washington, DC. November.
 Dey, J.C., R.D. Norris, and P. Rosenwinkle. 1996. Design and implementation of a highly
 integrated and automated in situ bioremediation system for petroleum hydrocarbons.  Proc. 89th
Air and Waste Management Meeting.  Nashville, TN.  June.
 DiStefano, T.D., J.M. Gossett, and S.H. Zinder. 1991. Reductive dechlorination of high
 concentrations of tetrachloroethene to ethene by an anaerobic enrichment culture in the absence of
 methanogenesis. Appl. Environ. Micro. 57(8):2287-2292.
 DiStefano, T.D., J.M. Gossett, and S.H. Zinder. 1992. Hydrogen as an electron donor for
 dechlorination of tetrachloroethene by an anaerobic mixed culture. Appl. Environ. Micro.
 58(ll):3622-3629:
 Domenico, P.A. 1987.  An analytical model for multidimensional transport of decaying
contaminant species. J. Hydrology.  91:49-58.
Don, J.A. and L. Feenstra. 1984.  Odour abatement through biofiltration. Paper presented at
Symposium Louvain.  La Neuve, Belgium. April.
Donnelly, K.C., C.S. Anderson, J.C. Thomas, K.W. Brown, D.J.  Manek, and S.H. Safe. 1992.
Bacterial mutagenicity of soil extracts from a bioremediation facility treating wood preserving
waste.  J. Hazardous Materials. 30:71-81.
Downey, D.C., R.A. Frishmuth, S.R. Archabal, C.S. Pluhar, P.O. Blystone, and R.N. Miller.
 1995. Using in situ bioventing to minimize soil vapor extraction costs. In Situ Aeration: Air
Sparging, Bioventing, and Related Remediation Processes. R.E. Hinchee, R.M. Miller, and P.C.
Johnson (eds.).  Columbus, OH: Battelle Press.  3(2)::J47-266.
Dragan, D. 1993. Criteria for Feasibility of Biofiltration Technology in Air Pollution Control.
Report for AAS/EPA Environmental Science and Engineering Fellowship Program, US EPA.
Washington, DC.                                  !
                                          B.5
-------
                                          List of References
                                                                          	           	I	      	     	 :	
                                          Dragt, A.J. and S.P.P. Ottengraf.  1987.  Fundamental aspects of biofiltration and some practical
                                          experiences. Presented at the Conference on Advanced Technology and Biotechnology for
                                          Environmental Protection. Syracuse, Italy. March 19-21.
                                          Drake, L.R., S. Lin, G.D. Rayson, and P.J. Jackson. 1996. Chemical modification and metal
                                          binding studies of Datura innoxia. Environ. Sci. Technol. 30:110-114.
                                          Driscoll, F.G. 1986. Groundwater and Wells.  St. Paul: Johnson Division.

                                          Droste, E.X., B.Y. Su, and B.R. Weissman. 1994. The successful implementation of soil vapor
                                          extraction and air sparging remediation technologies at a Massachusetts site. In Proceedings of
                                          Focus on  Eastern Regional Ground Water Issues, NGWA, Burlington, VT, October,  p. 343-358.

                                          Dupont, R.R.  1986a.  Evaluation of air emission release rate model predictions of hazardous
                                          organics from land treatment facilities. Environ. Progress.  5(3): 197-206.

                                          Dupont, R.R.  1986b.  A flux chamber/solid sorbent sampling system for volatile organic air
                                          emission monitoring from hazardous waste land treatment systems: field results. EPA 600/9-86-
                                          013:662-673.  APCA Publication VIP-7. Proc 1986 EPA/APCA Symposium on Measurement of
                                          Toxic Air Pollutants.  Office of Research and Development, Research Triangle, NC.
                                          Dupont, R.R.  1987. Measurement of volatile hazardous organic emissions from land treatment
                                          facilities. J. Air Pollution Control Assoc.  37(2):168-176.
                                          Dupont, R.R.  1988. A sampling system for the detection of specific hazardous constituent
                                          emissions from soil systems. Hazardous Waste:  Detection Control Treatment. Amsterdam, The
                                          Netherlands: Elsevier Science Publisher, B.V.
                                          Dupont, R.R. 1990. Quantitation and Modeling of Volatile Hazardous Constituent Emissions From a
                                          Hazardous Waste Land Treatment Facility. Final Repprt submitted to the U.S. EPA, Office of
                                          Research and Development, RSKERL, Ada, OK. Cooperative Agreement CR-813211-01.
                                          Dupont, R.R.  1991.  Applications of treatability studies in management of fuels/petroleum waste
                                          impacted soils. Proc 84th Annual Meeting & Exhibition of the Air & Waste Management
                                          Association. Vancouver, British Columbia. June 16-21.
                                          Dupont, R.R. and J.A. Reineman. 1986. Evaluation of the Volatilization of Hazardous Constitu-
                                          ents at Hazardous Waste Land Treatment Sites. EP A/600/2-86/071. Office of Research and
                                          Development, Robert  S. Kerr Environmental Research Laboratory, Ada, OK.

                                          Dupont, R.R., W. Doucette, and R.E. Hinchee. 1991. Assessment of in situ bioremediation
                                          potential and the application of bioventing at a fuel-contaminated site.  In Situ and On-Site
                                          Bioreclamation. R.E. Hinchee and R.F. Olfenbuttel (eds.).  Stoneham, MA: Butterworth-
                                          Heinemann. pp 262-282.
                                          Dupont, R.R., R.D. Norris, and K. Gorder.  1996.  Selection of remedial strategies based on
                                          modeling of plume dynamics.  Poster Session Presented at the 7996" Conference and Exposition,
                                          Petroleum Hydrocarbons and Organic Chemicals in Ground Water: Prevention, Detection, and
                                          Remediation.  National Ground Water Association. November 12-15, Houston, TX.
                                                                      . •     ..•      :',  -"'  ', ,.    '• ':  I .    ';.  '   	'  ••	:.;, 'i.'..
                                          Dupont, R.R., D.L. Sorensen, M.W. Kemblowski, K. Gorder, and G. Ashby. 1996. An intrinsic
                                          remediation assessment methodology applied at two contaminated groundwater sites at Eielson
                                          AFB, Alaska. Proc First International Conference on Intrinsic Bioremediation.  IBC Technical
                                          Services, Ltd., London, England. March.
                                          Dupont, R.R., D.L. Sorensen, M.W Kemblowski, M.  Bertelson, D. McGinnis, L Kamil, and Y.
                                          Ma.  1997.  Monitoring and assessment of in-situ biocontainment of petroleum contaminated
                                          groundwater plumes.  Final Report Submitted to the U.S. EPA Analytical Sciences Branch,
                                          Characterization Research Division, National Exposure Research Laboratory, Office of Research
                                          and Development, Las Vegas, NV, and Research Triangle Park, NC. (In press).
                                                                                    B.6
iiini. iniiiiiiiw:,, 
-------
                                                                          Appendix B
 Dushenkov, V., P.B.A.N. Kumar, H. Motto, and I. Raskin.  1995.  Rhizofillration: the use of
 plants to remove heavy metals from aqueous streams. Environ. Sci, Technol. 29:1239-1245.
 Eckenfelder, W.W., Jr., and J.L. Musterman.  1995. Activated Sludge Treatment of Industrial
 Wastewater.  Technonomic Publishing Company Inc.

 Eitner, D.  1990-1991. Personal communication to R.F. Michelsen. G+E umwelt-technologie
 GmbH, Aachen, Germany.
 Eitner, D. and H.G. Gethke.  1987. Design, construction and operation of bio-filters for odour
 control in sewage treatment plants. Paper presented at the 80th Annual Meeting of the Air
 Pollution Control Association. New York, NY. June.
 Emery. D.D., and P.C. Faessler. 1997. First production-level bioremediation of explosives-
 contaminated soil in the United States. Annals of the New York Academy of Sciences Vol. 829.
 R.K. Bajpai and M.E. Zappi, Eds. Conference Proceedings: Bioremediation of Surface and
 Subsurface Contamination, Palm Coast, Florida, January 21-26,1996. P. 326.  ;
 Entry, J.A., L.S. Watrud, R.S. Manasse, and N.C. Vance. 1997. Phytoremediation and reclama-
 tion of soils contaminated with radionuclides. E.L. Kruger, T.A. Anderson, and J.R. Coats (eds.),
 Phytoremediation of Soil and Water Contaminants, ACS Symposium Series No. 664. American
 Chemical Society, Washington, DC.
 Enzien, M.V., F. Picardal, T.C. Hazen, R.G. Arnold, and C.B. Fliermans. 1994. Reductive
 dechlorination of trichloroethylene and tetrachloroelthylene under aerobic conditions in a sediment
 column. Appl. Environ. Micro. 60(6):2200-2204.
 ES&T,Inc.  1994.  BioTRANS. Areal Dissolved Phase Transport Model for Organic Chemical
 Spills with Natural or Enhanced Bioremediation,  Users Manual and Technical Guideline.
 Blacksburg, VA.
 Falatico, R.J. and R.D. Norris. 1990. The necessity of hydrogeologic analysis for  successful in
 situ bioremediation. Proc. Haztech International Pittsburgh Waste Conference. Pittsburgh, PA.
 October 2-4.
 Felsot, A.S. and E.K. Dzantor. 1997. Potential of biostimulation to enhance dissipation of aged
 herbicide residues in land-farmed waste.  E.L. Kruger, T.A.  Anderson, and J.R. Coats (eds.).
 Phytoremediation of Soil and Water Contaminants. ACS Symposium Series 664. American
 Chemical Society, Washington, DC.
 Ferro, A.M., L. Stacishin, W.J. Doucette, and B. Bugbee.  1994.  Crested wheatgrass accelerates
 the degradation of pyrene in soil. Abstract:  Emerging Technologies in Hazardous Waste
 Management  VI, ACS Industrial & Engineering Chemistry Division Special Symposium,  Volume I.
 Atlanta, GA.  September 19^21.
 Ferro, A.M., R.C. Sims, and B. Bugbee.  1994. Hycrest crested wheatgrass accelerates the
 degradation of pentachlorophenol in soil.  J. Environ. Qua!.  23:272-279.
Fiorenza, S., E.L. Hockman, S. Szojka, R.M. WoelleT, and J.W. Wigger. 1994. Natural anaerobic
degradation of chlorinated solvents at a Canadian miinufacturing plant. Bioremediation of
 Chlorinated and Polycyclic Aromatic Hydrocarbon Compounds. R.E. Hinchee, A. Leeson, L.
Semprini, and S.K. Ong (eds.). Boca Raton, FL: Lewis Publishers.
Flathman, P.E., K.A. Khan, D.M. Barnes, J.H. Carori, S.J. Whitehead, and J.S. Evans. 1991.
Laboratory evaluation of hydrogen peroxide for enhanced biological treatment of petroleum
hydrocarbon contaminated soil. In Situ Bioreclamation:  Application and Investigation for
Hydrocarbons and Contaminated Site Remediation. R.E. Hinchee and R.F. Olfenbuttel (eds.).
Stoneham, MA: Butterworth-Heinemann.
Fletcher, J.S., F.L. Johnson, and J.C. McFarlane. 1988. Database assessment of phytotoxicity
data published on terrestrial vascular plants.  Environ. Toxicol. Chem. 7:615-622..


                                          B.7
-------
I!  '•)	  ' 1      • '   IK1;: IW ».<''' - It" If'Biill	'"'  '"J	[  !     :"•.,.   "	   <•    •  •    ' !•	, •!  	"', •   •!       ' '.  •  •;•.  '  " TV ,||
";	•        ;":-i""      ';;.     :             '•  .'•'     •    •  ••"..'•. •  i«-    >•   '         •      'i.' t
                                         List of References
                               •I     i	
 Fletcher, J.S., P.K. Donnelly, and R.S. Hegde.  1995. Biostimulation of PCB-degrading bacteria
 by compounds released from plant roots. Bioremediation of Recalcitrant Organics. R.E.
 Hinchee, D.B. Anderson, and R.E. Hoeppel (eds.). Columbus, OH: Battelle Press.
     I	  , n ,'" JIN \: •  ,il'n, 	i  I,;!'11!!1 '' „, ''"	•; ' '] '  ' '|l''"'1 '»	      ,' li '' UN'!!1 i  J'PI1 „  '' •   	I1" I" '• I 	I	  i"    ' i           ' '!;  ',! 'I'1 ''I'1'1"
 Fletcher, M.  1977. The attachment of bacteria to surfaces in aquatic environments. Adhesion of
 Microorganisms to Surfaces. D.C. Ellwood (ed.). Academic Press.
 Poor, D.C., f'.c: Zwick, R.E. Hinchee, R.E. Hoeppel, Cl Kyburgi and L. Bowling. 1995. Passive
 bioventing driven by natural air exchange.  In Situ Aeration: Air Sparging, Bioventing, and
 Related Remediation Processes. R.E. Hinchee, R.M. Miller, and P.C. Johnson (eds.). Columbus,
 OH: Battelie Press.  3(2):369-376
 Forsythe, J.V., Y.M. Tsao, and R.D. Bleam.  1995. Bioremediation: when is augmentation
 needed? Bioaugmentation for Site Remediation. R.E. Hinchee, J. Fredrickson, and B.C. Alleman
 (eds.). Columbus, OH: Battelle Press.
 Frechen, F.B.  1993. Biological deodorization of odorous air from waste and wastewater
 treatment plants. Paper presented at the Air and Waste management Association 86th Meeting and
 Exhibition. Denver, CO. June 13-18.
 Freedman, D.L. and J.M. Gossett.  1989. Biological reductive dechlorination of tetrachloroethyl-
 ene and trichloroethylene to ethylene under methanogenic conditions. Appl. Environ. Micro.
 55(9):2144-51.
 Freeze, R.A. and J.A. Cherry. 1979.  Groundwater. Englewood Cliffs, NJ:  Prentice Hall.

 Fuller, WH. 1986. Site selection fundamentals for land treatment Land Treatment: A Hazardous
 Waste Land Treatment Alternative Water Resources Symposium No. 13. R.C. Loehrand J.F. Malina,
 Jr (eds.). Center for Research hi Water Resources, The University of Texas  at Austin.
"	  	     r '  ,,    ;s v   •• * ,   '   ;„•    ,.',' '  ,,  ;  -1,;11  i , • ,j	J  ;\-\	, .     ,,	',-	      f    >.   i. "'
 Gannon, LT., V.B'".' Manila!,' and M. Alexander.  1991. Relationship between cell surface
 properties and transport of bacteria through soil. Appl. Environ. Micro. 57(t):190-193.

 Gatliff, E.G.  1994.  Vegetative remediation process offers advantages over traditional pump-and-
 treat technologies. Remediation. 4:343-352.
 Gee, G.W. and J.W  Bauder. 1986. Particle-size analysis. Methods of Soil Analysis, Part /.
 Physical andMineralogical'Methods. A. Klute (ed.). 2nd edition.  Madison, WI: American
 Society of Agronomy.
 Ghiorse, W.C. and D.L. Balkwill.  1990. Enumeration and morphological characterization of
 bacteria indigenous to subsurface environments. Subsurface Microbiology:  Indigenous
 Enumeration and Characterization.

 Gibson, S.A*.'and G.W. Sewell.  1992.  Stimulation of reductive dechlorination of
 tetrachloroethene in anaerobic aquifer microcosms by addition of short-chain organic acids or
 alcohols. Appl. Environ. Micro. 58(4)1392-1393^

                                         Golueke, C.G. and L.F. Diaz.  19891 Biological treatment for hazardous wastes. BioCycle.  30:58.

                                         Gorder, K., R.R. Dupont, D.L. Sorensen, M.W. Kemblowski, and J.E. McLean. 1996. Applica-
                                         tion of a simple groundwater model to assess the potential for intrinsic remediation of contami-
                                         nated groundwater.  Presented to the First IBC International Conference on Intrinsic Remediation.
                                         London, England. March 18-19.
                                         Greenberg, A.E., L.s! Clesceri, and A.D. Eaton (eds.). 1992. Standard Methods for the
                                         Examination of Water and Wastewater.  Washington, DC: American Public Health Association.

                                         Griest, W.H., R.L. Tyndall, A.J. Stewart, H. Ho, K.S. Ironside, J.E. Caton, W.M. Caldwell, and E.
                                         Tan.  1991.  Characterization of Explosives Processing Waste Decomposition Due to Composting.
                                         ORNL/TM-12029. Prepared for U.S. Army Medical Research and Development Command.
                                         November.
                                                                                ,"",''„          i1                          ;
                                                                                   B.8
-------
                                                                         Appendix B
Griffin, C.J., J.M. Armstrong, and R.H. Douglass.  1992.  Engineering design aspects of an
in-situ soil vapor remediation system (sparging). In Situ and On-Site Bioreclamation.  R.E.
Hinchee and R.F. OifenbuttelI (eds.).  Stoneham, MA:  Butterworth-Heinemann.

Groundwater Services, Inc.  1996. BIOSCREEN, Intrinsic Remediation Decision Support System,
Users Manual. Version 1.1. Developed for the Air Force Center for Environmental Excellence,
Brooks AFB, San Antonio, TX.
Gudemann, H. and D. Killer. 1988. In-situ remediation of VOC contaminated soil and ground
water by vapor extraction and ground water aeration. Proc. Third Annual Haztech International
Conference. Cleveland, OH.
Gudin, C. and W.J. Syratt.  1975. Biological aspects of land rehabilitation following hydrocar-
bon contamination. Environ. Pollut.  8:107-112.
Hall, J., D.C. Downey, and W.H. Reed.  1995.  Use of air injection for site dewatering  and
enhanced bioremediation. Paper No. 58a.  Presented at the 7995 Summer Meeting of American
Institute of Chemical Engineers.                  \
Harvey, R.W., R.L. Smith, and L. George. 1984.  Effect of organic contaminants upon microbial
distribution and heterotrophic uptake in a Cape Cod, MA, aquifer. Appl.  Environ.,Micro.
48:1197-1984.                                  i
Hater, G.R., J.S. Stark, R.E. Feltz, and A.Y. Li. 1994. Advances in vacuum heap bioremediation at a
fixed-site facility: emissions and bioremediation rate. Hydrocarbon Bioremediation. R.E. Hinchee,
B.C. Alleman, R.E. Hoeppel, and R.N. Miller (eds.).  Boca Raton, FL: Lewis Publishers.

Hater, G.R., R.B. Green, T. Solsrud, D.A. Bower, Jr., and J.A. Barbush.  1995. Biological Unit
Processes for Hazardous Waste Treatment. R.E. Hinchee, G.D. Sayles, and R.S. Skeen (eds.).
Columbus, OH:  Battelle Press,  pp 171-190
Hatzinger, P.B. and M. Alexander. 1995.  Effect of aging of chemicals in soil on their biodegrad-
ability and extractability. Environ. Sci. Technol. 29:537-545.
Heron, G., J.C. Tjell, and T.H Christensen. 1992.  Oxidation capacity of aquifer sediments.
Environ. Sci. Technol. 26:153-158.               |
Heron, G., J.C. Tjell, and T.H Christensen. 1994.  Oxidation capacity of aquifer sediment,
hydrocarbon bioremediation. Environ. Sci. Technol, 28:278-284.
Herrling, B. and W. Buermann.  1990. A new method for in-situ remediation of volatile
contaminants in groundwater— numerical simulation of the flow regime. Proc. Eighth
International Conference on Computational Methods in Water Resources. Venice, Italy. June.

Herrling, B., J. Stamm, and W. Bnermann.  1991.  Hydraulic circulation system for in situ
bioreclamation and/or in situ remediation of strippable contamination.  In Situ and On-Site
Bioreclamation.  R.E. Hinchee and R.F. Oifenbuttel (eds.).  Stoneham, MA: Butterworth-
Heinemann.
Herrling, B., W. Buermann, and J. Stamm.  1991.  In-situ Remediation of Volatile Contaminants
in Groundwater by a New System of 'Vacuum-vaporizer-wells'. Institute of Hydromechanics,
University of Karlsruhe, Germany.
Herrling, B., J. Stamm, E.J. Alesi, P. Brinnel, F. Hirschberger, and M.R.  Sick.  1991. In situ
groundwater remediation of strippable contaminants; by vacuum vaporizer wells (UVB):
operation of the well and report about cleaned industrial sites. Third Forum on Innovative
Hazardous Waste Treatment Technologies: Domestic and International. Dallas, TX.  June.

Hinchee, R.E. and M.F. Arthur.  1990.  Bench-scale studies on the soil aeration process for
bioremediation of petroleum-contaminated soils.  J. Applied Biochem. Biotech.  28/29:901-906.
                                          B.9
-------
         List of References
         Hinchee, R.E. and S.K. Ong. 1992. A rapid in situ respiration test for measuring aerobic biodegrada-
         tion rates of hydrocarbons in soil. J. Air and Waste Management Assoc. 42(10): 1305-1312.

         Hinchee, R.E., S.K. Ong, R.N. Miller^ D.C. Downey, andR. Frendt.  1992. Test Plan and
         Technical Protocol for a Field Treatability Test for Bioventing. Final Report Prepared for Air
         Force Center for Environmental Excellence, Brooks AFB, TX.
         Hobbie, I.E., R.J. Daley, and S. Jasper.  1977. Use of nuclepore filters for counting bacteria by
         fluorescence microscopy. Appl. Environ. Micro. 33(5): 1225-1228.

         Hosier, K.R. 1990. Generic Protocol for Bench-Scale Testing ofBioremediation Technologies.
         Unpublished report prepared for SCS Engineers (Vancouver, BC) by the Wastewater Technology
         Centre (Environment Canada).
         Huddieston, R.L, C.A. Ble'ckman, and J.R. Wolfe.  1^86.  Land treatment biological degradation
         processes. Land Treatment:  A Hazardous Waste Land Treatment Alternative. Water Resources
         Symposium No. 13. R.C. Loehr and J.F. Malina, Jr. (eds.). Center for Research in Water
         Resources, The University of Texas at Austin.
         Ruling, S.G., B.E. Bledsoe, and M.V. White.  1990.  Enhanced Bioremediation Utilizing
         Hydrogen Peroxide as a Supplemental Source of Oxygen:  A Laboratory and Field Study. EPA/
         600/2 90/006. Robert S. Kerr Environmental Research Lab'oratoryrAda, OK.
         Huling, S.G., D.F. Pope, JJE. Matthews,  J.L. Sims, R.C. Sims, and D.L. Sorensen. 1995b. Wood
         preserving waste-contaminated soil: treatment and toxicity. Bioremediation of Recalcitrant
         Organics.  R.E. Hinchee, R.E. Hoppel, and D.B. Anderson (eds.). Columbus, OH: Battelie Press.
         Huling, S.G., D.F. Pope, J.E. Matthews,  J.L. Sims, R.C. Sims, D.L. Sorenson.  I995a.  Land
         treatment and the toxicity response of soil contaminated with wood preserving wastes. Journal of
         Remediation, pp 41-55.  Spring.
         Hurst, C.J.  1996. Prepared-bed bioremediation as affected by oxygen concentration in soil gas.
         Libby, Montana, Superfund site. Masters Thesis. Department of Civil and Environmental
         Engineering, Utah State University, Logan, UT.
             :   '" "	 l	;•• . • !. •' '•  '•"'  ', '  i   ,.;  ,"  .' ,/T ,:,' .   	!",,.,*i'.	' ! ,,:  "I ',.    ! •   .,'      . H ;  ' ' 0, !
         Hurst, J.C., R.C. Sims, J.L. Sims, D.L. Sorenson, J.E. McLean, and S.G. Huling. 1995a.
         Polycyclic aromatic hydrocarbon biodegradation as a function of oxygen tension in contaminated
         soil. Journal Haz. Waste Haz. Mat. 12:(3).
        Hurst, J.C.,R.C. Sims, J.L. Sims, D.L. Sorenson, J.E. McLean, and S.G. Huling.  1995b. Soil gas
        oxygen tension and pentachlorophenol biodegradation.  J. Environ. Eng. ASCE. Submitted.
        Hurst, J.C., R.C. Sims, J.L. Sims, D.L. Sorenson, I.E. McLean, and S.G. Huling.  1996.  Polycy-
        clic aromatic hydrocarbon biodegradation as a function of oxygen tension in contaminated soil.  7.
        Hazardous Material. 51:193-208.
 .    "..     '     . .  .    :            '.	  ,       "":,"     .:': i                 ,  ..
        Hurst, J.C.,R.C. Sims, J.L. Sims, D.L. Sorenson, J.E. McLean, and S.G. Huling. 1997. Soilgas
        oxygen tension and pentachlorophenol biodegradation. /. Environ. Eng., ASCE, Vol. 123(4), 364-370.
        Hutchins, S.R. 1992. Biotransformation and mineralization of alkylbenzenes under denitrifying
        conditions, Environmental Toxicology and Chemistry, (Ii):l4l3-i423.
                             	     lh  "     '   		 I                          i '
        Hutchins, S.R. and J.T. Wilson. 1991.  Laboratory and field studies on BTEX biodegradation in a
        fuel-contaminated aquifer under denitrifying conditions. In Situ and On-Site Bioreclamation.
        R.E. Hinchee and R.F. Olfenbuttel (eds.). Stoneham, MA: Butterworth-Heinemann.
        Hutchins, S.R., G.W. Sewell, D.A. Kovaks, and GA. Smith. 1991. Biodegradation of aromatic
        hydrocarbons by aquifer microorganisms under denitrifying conditions. Env/ron. 5a. TecAno/. 25(1).
••	I       •	,   ,.     •     ., -,:  •,      •   •  •	,   , ..  . vi, .   , • ,„	j  .  ,. ,r ;,,,-    :,.     	.   ,:•,,:!
        Hutchins, S JL, D.E. Miller, F.P. Beck, A. Thomas, S.E. Williams, and G.D. Willis. 1995. Nitrate-
        based bioremediation of JP-4 jetfuel: pilot-scale demonstration, Applied Bioremediation of Petroleum
        Hydrocarbons, Edited by R.E. Hinchee, J.A. Kittel, and HJ. Reisinger, Battelie Press; pp. 123-13L
      	                             "          	I            	" '
 ,:   ;:,        :                :..                '  B.IO     ;     '    ;     '     ,;,',;     "    ; •;
-------
                                                                         Appendix B
Imbrigiotta, T.E., T.A. Ehlke, and M. Martin. 1991. Chemical evidence of processes affecting the fate
and transport of chlorinated solvents in ground water at Picatinny Arsenal, New Jersey. Water-
Resources Investigations Report 91-4034. Proc, Technical Meeting U.S. Geological Survey Toxic
Substances Hydrology Program. G.E. Mallard and D.A. Aronson (eds.). Monterey, CA.

Interdisciplinary Plant Group.  1995. Will plants have a role in bioremediation? Proc. 14th
Annual Missouri Symposium of Current Topics in Plant Biochemistry, Physiology, and Molecular
Biology.  University of Missouri-Columbia.
IPS. 1990. Odor Control Completing the Composting Process. Glastonbury, CT.

Javanmardian, M., J.S. Huber, C.B. Olson, W.P. Schwartz, C.A. Masin, and VJ. Kremesec. 1995.
In situ biosparging at an Amoco site: subsurface air distribution and biostimulation. In Situ
Aeration: Air Sparging, Bioventing, and Related Remediation Processes. R.E. Hinchee, R.M.
Miller, and P.C. Johnson (eds.). Columbus, OH: Battelle Press.

Ji, W., A. Dahmani, D. Ahlfeld, J. Lin, and E. Hill, Jr. 1993.  Laboratory study of air sparging: air
flow visualization.  Ground Water Monitoring and Remediation.  13(4):115-125.

Johnson, P.C., R.L. Johnson, C. Neaville, E.E. Hansen, S.M. Stearns, and I.J. Dortch. 1995. Do
conventional monitoring practices indicate in situ air sparging performance? In Situ Aeration:
Air Sparging, Bioventing, and Related Remediation Processes. R.E. Hinchee, R.M. Miller, and
P.C. Johnson (eds.). Columbus, OH: Battelle Press.
Johnson, R.L., P.C. Johnson, D.B. McWhorter, R.E. Hinchee, and I. Goodman.  1993. An
overview of in situ air sparging. Ground Water Monitoring and Remediation.  13(4):127-135.

Kaback, D.S., B.B. Looney, C.A. Eddy, and T.C. Hazen.  1991. Innovative Groundwater and Soil
Remediation: In Situ Air Stripping Using Horizontal Wells.  Case Study at Westinghouse
Savannah River Company, Savannah River Site, Aiken, SC.
Kampbell, D.H. 1997. Personal Communication. US EiPA Robert S. Kerr Environmental Research
Center, Ada, OK.
Kampbell, D.H., J.T. Wilson, H.W. Read, and T.T. Stocksdale. 1987. Removal of volatile
aliphatic hydrocarbons in a soil bioreactor. J. Air Pollution Control Assoc. 37(10):1236-1240.

Kane, A.C., J.M. Fischer, and T.P. Wilson.  1996. Microcosm study to determine optimum levels
of oxygen and methane for cometabolic biodegradation of trichloroethylene in sediment and
ground water from Picatinny Arsenal, Morris County, New Jersey.  Presented at AGU Spring
Meeting, Baltimore, MD.  May 20-24.
Kao, C.M. and R.C. Borden. 1994. Enhanced aerobic: bioremediation of a gasoline contaminated
aquifer by oxygen-releasing barriers. Hydrocarbon Bioremediation. R.E. Hinchee, B.C. Alleman,
R.E. Hoeppel, and R.N. Miller (eds.). Boca Raton, FL: Lewis Publishers.

Keck, J., R.C. Sims, M. Coover, K. Park, B. Symons. 1989.  Evidence of co-oxidation of
polynuclear aromatic hydrocarbons in soil, Water Res: 23:1467.

Kerfoot, W.B. 1995. Slantwell sparging. Presented at the Tenth Annual Conference on
Contaminated Soils. University of Massachusetts at Amherst, Amherst, MA. October 23-26.
Kienbusch, M.R. 1986. Measurement of Gaseous Emission Rates From Land Surfaces Using an
Emission Isolation Flux Chamber: User's Guide. EP.A/600/8-86/008.  Environmental Monitoring
Systems Laboratory, Las Vegas, NV.
Klecka, G.M. and SJ, Gonsior.  1984.  Reductive dechlorination of chlorinated methanes and
ethanes by reduced iron (III) porphyrins. Chemosphere. 13:391-402.
Klecka, G.M., J.W. Davis, D.R. Gray, and S.S. Madsen.  1990. Natural bioremediation of organic
contaminants in ground water: Cliffs Dow Superfund site. Ground Water.  28:534-543.
                                         B.11
-------
                                         Ust of References
                                         Koganov, M.M., and D.B. Dawson. 1995. Bioremediation system based on FST-technology:
                                         History, status, outlook.  Poster Abstracts, In Situ and On-Site Bioreclamation, The Third
                                         International Symposium, April 24-27,1995, San Diego, CA. Battelle Memorial Institute.
                                         Konikow, L.F and J.D. Bredehoeft.  1978. Computer Model of two-dimensional Solute
                                         Transport and Dispersion in Ground Water. U.S. Geological Survey, Techniques of Water
                                         Resources Investigation. Bk. 7, Ch. C2.
,,;.  ' .           t i  ,       '  '  s;;  ,»•>;   ......  " \.  ;'":"'   '    %i';' :/ ' •    , - •     •  •,,"•  ; '  ^ir< ii ....... •.  .•  ;;: •   |ji.:"-   ..... :;•'.  •„:•:  , ,-(  ,,.»•••  F ;i ..... 11
                                         Korfiatis, G.P. and C. Christodoulatos. 1993.  Treatability studies as a remedial option screening
                                         tool for contaminated soils. Remediation: The Journal of Environmental Cleanup Costs,
                                         Technologies & Techniques.  3(4):395-4l2.
, ...... M     •  , ,>      ;l;:   !• «  •     "' "'i  ' '•!!  ,"  ,'  ., ' .  • »• ,; i :."Mi     ! t. "   1! i   .- •'      '. if. ••:",, , •   i ....... iHvli .  i ,.  i  ,  .;..>' ...... ,;,!„;: ...,.'•,    "   r"'..JF
                                         Rorom, S.F., M.J. McFarland, and R.C. Sims. 1996. Reduced sediments:  A factor in the design
                                         of subsurface oxidant delivery systems.  Ground Water, pp 100-105. Winter.
                                         Kumar, N.P.B.A., V. Dushenkov, H. Motto,, and I. Raskin. 1995.  Phytoextraction: the use of
                                         plants to. remove heavy metals from soils. Environ. Sci. fechnol.  29:1232-1238.
                                  -,";  ; Nost^t Reinhpld, New Ifprk, WewYorlc.  ""_" ..... '" .....  _  '  "  '       '  '       ........ i   '""__'
                                        Lawes, B.C. 1991. Soil-induced decomposition of hydrogen peroxide. In Situ Bioreclamation:
                                        Application and Investigation for Hydrocarbons and Contaminated Site Remediation.  R.E.
                                        Hinchee and R.F. Olfenbuttel (eds.). Stoneham, MA: Butterworth-HeiriernahnI
                                        Lee, M.D. and C.H. Ward.  1984 Microbial ecology of a hazardous waste disposal site: enhance-
                                        ment of biodegradation. Proc. of the 2nd International Conference on Ground Water Quality
                                        Research, Oklahoma State i University.
                                        Le'es'on! A. and R.E. Hinchee.  1995! PtincipiesandPracticeofBioventing. Final isport to US^f
                                        Environics Directorate, Armstrong Laboratory, TyridallAFB, FL.  US EPA, Washington, DC and
                                        USAP ^ Center for Environmental Excellence, Brooks AFB.TX.

                                        Leeson, A., R.E. Hinchee, D.C. Downey, CM. Vogel, G.D. Sayles, and R.N. Miller.  1995.
                                        Statistical analyses of the Air Force bioventing initiative results. In Situ Aeration: Air Sparging,
                                        Bioventing, and Related Remediation Processes. R.E. Hinchee, R.M. Miller, and P.C. Johnson
                                        (eds.). Columbus, OH: Battelle Press.

                                        Lehmicke, L.G., R.T. Williams, and R.L. Crawford. 1 979.  14C-Most probable number method
                                        for enumeration of active heterotrophic microorgamisms in natural waters. Appl. Environ.
                                        Microbiol. 38(4):644-649.
                                   S1  '  :'  '.,•.'  " j.   ,   ';.'...  ,'/  ..... ••.     I1,'';')'" ..... iSi'1  -S',' ' '    ,-i-j,'i ('''l11"'1'-'1"^1*11'1' '•'•. '  '"  !•''•'[."'   •'  ;   ! ......... <''
                                        Leson, G. and A.M. Winer. 1991. Biofiltration: an innovative air pollution control technology
                                        for VOC emissions. /. AirandWaste Management Assoc. 41(8):ld45-1054.
                                  ';•!!' 'i"1 '•!    •i:''iil ..... :,;   > ...... "   ,,,' ..... .......... • ....... .'";••.  ';  '.'•<•:,•',•    i£i] ••:•'.•  >"i  -.  K1' ....... •;'•'  > : ,1 , ilfli •  •  ;« ..... .,"  , , ; ; • i  "•'••. »   '! ;•*'
                                        Lewis, B.C., and M.M. MacDonell.  1990.  Release of radon-222 by vascular plants: effect of
                                        transpiration and leaf area. /. Environ. Qual. 19:93-98.
                                        Lira, D.V.  1989. Microbiology.  St. Paul, MN: West Publishing Co.
                                   iB'1!    •• ...... ii ........ • ..... •>'.•".     ;.n '.,; .......... ill it .i ........ .......... ;'•:., i ...... i.'i ...... I": .is :  ' ..... .""•   3»;/ .'  ..... I1"', I ,"''.•  .1 , ..... ;. ..... jr.. ,   . ..... i1 .• "  r   I'.jv.
                                        Loden, M.E. and C.Y. Fan.  1992. Air sparging technology evaluation. Proc. National R&D
                                        Conference on the Control of Hazardous Materials. Hazardous Material Control Research
                                        Institute, San Francisco, CA. February 4-6.
                                  fit/  '''] ..... ......   '  '     "   T  Illli  i  ., ,' II I ' i , , i , .....   , ,     •   ',"   .     I . II .      ,      i     .      i   ..... ,„
                                        Loehr.RlC.  1989.  Treatability Potential for EPA Listed Hazardous Wastes in 'Soil. EPA/600/2 -
                                        89/011. Office of Research and Development,  Robert S. Ken- Environmental Research Labpra-
                                  !':".  . toiy, Ada, OK.
                                      ..'  ...... ...... ,;   j   M"  ..    ,    ,.'..  ipii" ., ';.  r: "  ,  ....... \ ..... ,. ,„ . ' ...... .i'.1.!;,,".  "  ..  'ii1,,11 i      .     . !      . ..... ,  ,   j :,,i,i:,,
                                        Loehr, R.C. and J.F. Malina, Jr., Ed.  1986. Land treatment: a hazardous waste management
                                        alternative. Water Resources Symposium Number 13. Center for Research in Water Resources,
                                        Bureau of Engineering Research, College of Engineering, The University of Texas at Austin,
                                  • ..... '   Austin, Texas.      '   '     '     |    / "/"    i  '"' .......  " ^ ........ _   ".  '  ',    "


                                  •"-•     '   ..  '    '  ••  '   "     ,'  ••' '  '  •'- ':: .' B'.12 ............. '   "  ' ..... ' ...... '     '
-------
                                                                         Appendix B
Loehr, R.C., B.C. Erickson, L.A. Rogers, and D.M. Kelmar. 1990. Mobility and Degradation of
Residues at Hazardous Waste Land Treatment Sites at Closure. EPA/600/2 90/018. Office of
Research and Devleopment, Robert S. Kerr Environmental Research Laboratory, Ada, OK.
Loehr, R.C., W.J. Jewell, J.D. Novak, W.W. Clarkson, and G.S. Friedman. 1979. Land
Application of Wastes. Vol. II. Van Nostrand Reinhold Environmental Engineering Series, New
York: Van Nostrand Reinhold.
Logan, E.B., T.A. Hilbert, and R.G. Arnold. 1993. A method to screen cultures for properties
that affect microbial transport in porous media. Water Research. 27(6):955-962.

Lovley, D.R., and E.J.P. Phillips. 1988. Competitive mechanisms for inhibition of sulfate
reduction and methane production in the zone of ferric iron reduction in sediments, Appl.
Environ. Microbiol. (53): 2636-2641.
Lovley, D.R., F.H. Chapelle, and J.C. Woodward. 1994. Use of dissolved H2 concentrations to
determine distribution of microbially catalyzed redox reactions in anoxic ground water, Environ.
Sci. Technol., 28(7): 1205-1210.
Lundegard, P.D. and D. LaBrecque. 1995. Air sparging in a sandy aquifer (Florence, Oregon,
U.S.A.): actual and apparent radius of influence. J. Contaminant Hydrology.  19:1-27.
Lundegard, P.L. 1994.  Actual versus apparent radius of influence — air sparging in a sandy
aquifer. Proc. Petroleum Hydrocarbons and Organic Chemicals in Ground Water: Prevention,
Detection and Restoration. NGWA/API.  Houston, TX. November.
Lundegard, P.L. 1995.  Air sparging: much ado about mounding. In Situ Aeration: Air
Sparging, Bioventing, and Related Remediation Processes. R.E. Hinchee, R.M. Miller, and P.C.
Johnson (eds.). Columbus, OH: Battelle Press.
Lyngkilde, J. and T.H. Christensen. 1992.  Redox zones of a landfill leachate pollution plume. J.
Contam. Hydrol. 10:273-289.
Mackay, D.M. and J.A. Cherry. 1989. Groundwater contamination: pump-and-treat remediation.
Environ. Sci. Techno. 23(6):630-636.
Madsen, E.L. 1991. Determining in situ biodegradallon. Environ. ScLTechno. 25:1662-1673.
Major, M.A., J.M. Bollag, and J.C. Amos.  1994. Bioremediation Methods for Treatment of TNT
Contaminated Soils. Technical Report 9305. Prepared for U.S. Army Biomedical Research
Laboratory and Development Laboratory.
Marley, M.C. and C.J. Bruell.  1995. In Situ Air Sparging: Evaluation of Petroleum Industry
Sites and Considerations for Applicability, Design and Operation. API Publication 4609.
American Petroleum Institute, Washington, DC.
Marley, M.C., DJ. Hazebrouck, M.T. Walsh. 1992. The application of in situ air sparging as an
innovative soils and groundwater remediation technology. Ground Water Monitoring Review,  pp
137-145. Spring.
Marley, M.C., E.X. Droste, F. Li, and R.J. Cody. 1994.  Design of an in situ sparging trench.
Proc. Conference on Petroleum Hydrocarbons and Organic Chemicals in Groundwater:
Prevention, Detection and Restoration. NGWQ/APL Houston, TX.
Marlow, H., H.R. Muniz, and S. Koeningsberg. 1995. The use of oxygen releasing compound
(ORC) for groundwater bioremediation in Alaska. Presented at the American Chemical Society
l&EC Division Special Symposium, Atlanta, Georgia.  September 17-20.
Marlow, H.J., K.L. Duston, M.R. Wiesner, M.B. Tomson, J.T. Wilson, and C.H. Ward. 1990.
Microbial transport through porous media:  the effects of hydraulic conductivity and injection
velocity. J. Hazardous Mat. 28:6574.
                                         B.13
-------
                                        List of References
                                        Marshall, K.C., R. Stout, and R. Mitchell.  1971.  Mechanism of the initial events in the sorption
                                        of marine bacteria to surfaces. /. General Microbial.  68:337-348.
                                            	      '         "            .         "'   	!	I     	''      ''    	
                                        Martin, L.M., R.J. Samelli, and M.T. Walsh. 1992. Pilot-scale evaluation of groundwater air
                                        sparging: site-specific advantages and limitations. Proc. National R&D Conference on the
                                        Control of Hazardous Materials.  San Francisco, CA. Published by Hazardous Material Control
                                        Research Institute.  February.
                                        Mason, B.J. 1983.  Preparation of Soil Sampling Protocol: Techniques and Strategies. EPA,
                                        600/4-83-20. US EPA Environmental Monitoring Systems Laboratory. Las Vegas, NV.
                                        McAneny, C.C., P.O. Tucker, J.M. Morgan, C.R. Lee, M.F. Kelley, and R.C. Horz. 1985. Covers
                                        for Uncontrolled Hazardous Waste Sites. EPAMO/2-85/002. Office of Solid Waste and
                                        Emergency Response, Washington, DC.
^ i,""     !         '!              	iiln |  ,'iH;'i,         '                         „	  ,     :„        '   II, ' ,        IL      „    IM' |       '    ,|,  ; I ir;i
                                        McCarty, P.L.  1994. An overview of anaerobic transformation of chlorinated solvents. Presentedlat
                                        the U$ EPA Symposium on Intrinsic Bioremediation of Gro«wrf Water.  Denver, CO. August 30.
                                        McCaulou, D.R. 1993. The Effects of Temperature and Motility on the Advective Transport of a
                                        Deep Surface Bacteria Through Saturated Sediment. Masters Thesis. Department of Hydrology
                                        and Water Resources. The University of Arizona, Tucson.
                                        McGinnis, G.D., H. Borazjani, M.J. McFarland, D.F. Pope, and D.A. Strobel. 1988. Character-
                                        ization and Laboratory Treatability Studies for Creosote and Pentachlorophenol Sludges and
                                        Contaminated Soil. EPA/600/2 88/055.  Office of Research and Development, Robert S. Ken-
                                        Environmental Research Laboratory, Ada, OK.
                                        McLean, LE. and B.E. Bledsoe. 1992.  Behavior of Metals in Soils. EPA/540/S-92/018. Ground
                                        Water Issue Paper.  Office of Solid Waste and Emergency Response, Office of Research and
                                        Development, Washington, DC.
     i   i   'i      |'P     ,  ' '  ,„    ;|i', 3|   jfii*.    ,  	|| '' .*'  in..',i „»',;,  ,:»' a •'   .! ' ,  ."   „'  iinj   . ,     i. ,,,, ,  .i,1'", 	,' Hhi , „ i	', , n:  • • • <• ij:  || „   "''':,    i  •         '   ",!'"'
                                        Merck & Co., Inc.  1983.  The Merck Index.  10th edition. Rahway, NJ.
                                        Michelsen, R.F. 1992. Biofiltration— an emerging emission control technology.  Paper
                                        presented at the New England Environ. Expo. Boston, MA. April 28-30.
                                        Michelsen, R.F. 1993. Biofiltration. Full scale application.  Paper presented at the Air and Waste
                                        Management Assoc. 86th Meeting and Exhibition. Denver, CO. June.
                                        Michelsen, R.F. 1995. Biofiltration. Handbook of Air Pollution Control Engineering and
                                       •Technology. J.JD. Mycock, J. D. McKenna, andIL. Theodore (eds.). Boca Raton, FL: Lewis
                                        Publishers.
	 	"'"  •"•••      '"              ;  	        	'•        •''      - •   '     '•••   	   '• '   •'     !••('•	,   	      	
                                        Middleton, A.C. and D. Killer.  1990. In situ aeration of ground water:  A technology overview.
                                        Proc. 1990 Environment Canada Montreal Conference.  Montreal, Quebec, Canada.
                                        Miller, R.N. and R.E. Hinchee.  1990. A field scale investigation of enhanced petroleum
                                        hydrocarbon biodegradation in the vadose zone — Tyndall AFB, FLl  Proc. NWWA/API
                                        Conference on Petroleum Hydrocarbons and Organic Chemicals in Ground Water. Houston, TX.
                                        Miller, W.E., S.A. Peterson, J.C. Greene, and C.A. Callahan. 1985. Comparative toxicology of
                                        laboratory organisms for assessing hazardous waste sites. J. Environ. Qual.  14:569-574.
                                        Mobil Oil Corporation.  1995. A Practical'Approach to the Evaluation of Intrinsic Bioremedia-
                                   • ',..'. ."tion ofPetroleum Hydrocarbons in Groundwater. Prepared by the Groundwater Technology
                                        Group, Environmental Health and Safety Department and the Environmental Health Risk
                                        Assessment Group, Stonybrook Laboratories.
                                        Mohn, W.W. and J.M. Tiedje.  1992  Microbial reductive^ dehaiogenation. Micro. Reviews.
                                        56(3):482-507.
                                                                                 B.14
•if' •'	liii!!
                    ^^^^^ ....... v« ....... ..... : ......... f,. ..... iiiiiiiillb     -ii ..... „„, L t vi^, .......... >.b ...... ...... ,jr ....... „:„ ....... ..... in  ... i. '\\ \:s-a ,,,:ii .......... 'i:;;,;,!.!.!''! ....... i •.iii, ' „ jiiiiii' '.,t  ...'.ar. ........ i ..... iJiMnii'iHi.!.*:;!'!'!.!,.!.;. ...... .. .lii
-------
                                                                         Appendix B
Molina, M., R. Araujo, and J.R. Bond.  1995. Dynamics of oil degradation in coastal environ-
ments: effects of bioremediation products and some environmental parameters. EPA/600/R-95/
076. Abstract: Symposium on Bioremediation of Hazardous Wastes: Research, Development,
and Field Evaluations. Rye Brook, NY. August 10-12.
Montemagno, C.D. and R.L. Irvine. 1990. Feasibility ofBiodegrading TNT-Contaminated Soils
in a Slurry Reactor. CETHA-TE-CR-90062. Prepared for the U.S. Army Toxic and Hazardous
Materials Agency.
Montgomery, J.H. 1991. Groundwater Chemicals Desk Reference. Volume 2. Chelsea, MI:
Lewis Publishers.
Morgan, D.J., A. Battaglia, B.J. Hall, L.A. Vernieri, and M.A. Cushey.  1992.  The GRl Acceler-
ated Biotreatability Protocol for Assessing Conventional Biological Treatment of Soils:
Development and Evaluation Using Soils from Manufactured Gas Plant Sites.  Topical Report
(June 1988 - December 1992).  GRI92/0499. December.
Morgan, D.J., A. Battaglia, J.R. Smith, A.C. Middleton, D.V. Nakles, and D.G. Linz.  1990.
Evaluation of a biodegradation screening protocol for contaminated soil from manufactured gas
plant sites. Presented at the IGT Symposium on Gas, Oil, Coal and Environmental Biotechnology.
New Orleans, LA. December 3-5.
Mutch, R.D., R.D. Morris, and D.J. Wilson. 1997.  Groundwater Cleanup by In Situ,Sparging.
XL  An Improved Aeration Curtain Design. Manuscript in preparation.

Mycock, D., J D. McKenna, and L. Theodore (eds.).  1995. Handbook of Air Pollution Control
Engineering and Technology. Boca Raton, FL: Lewis Publishers.
Nakles, D.V. and J.R. Smith. 1989. Treatability protocol for screening biodegradation of heavy
hydrocarbons in soil.  Presented at the Hazardous Materials Management Conference and
Exhibition.  Atlantic City, NJ. June 13-15.
Nelson, D.W. and L.E. Sommers. 1982. Total carbon,! organic carbon,  and organic matter.
Methods of 'Soil Analysis, Part 2. Chemical and Microbiological Properties. A.L. Page (ed.).
2nd edition.  Madison, WI: American Society of Agronomy.
Newman, P.A.B., F.S. Peterson, P.E. Boersma, and R.E. Huddleston. 1994. Effects of sparging
on ground water hydraulics and ground water quality. Proc. 87th Annual Meeting Air and Waste
Management Association. Cincinnati, OH.
Nicholson, S.A. and N.M. Safaya. 1993. Restoring hazardous waste sites. Environ. ScLTechnol.
27:1022-1025.
Nofziger, D.L., J.R. Williams, and T.E. Short. 1988. Interaction of the Simulation of the Fate of
Hazardous Chemicals During Land Treatment of Oily Wastes: RITZ Users Guide. EPA/600/8-
88-001. Office of Research and Development, Robert S. Kerr Environmental Research Labora-
tory, Ada, OK.
Norris, R.D. 1992. Addressing problems associated with the distribution of oxygen and nutrients
in bioremediation processes.  Preprint 92-30-02. Presented at the 85th Annual Meeting and
Exhibition of the Air & Waste Management Association. Kansas City, MO.
Norris, R.D. and K.C. Harvey.  1994.  A comparison of the use of fungus and bacteria in
commercial bioremediation processes.  Presented at this 87th Annual Air and Waste Management
Meeting.  Cincinnati, OH. June.
Norris, R.D. and R.A. Brown.  1994. The evolution of a technology: the in situ bioremediation of
hydrocarbons. Presented at the 1994 Superfund Conference. Washington, DC.
                                         B.15
-------
         List of References
" 'V ,, ill

4 I	:
    i	I
 Norris, R.D., R.A. Brown, G.E. Hoag.  U.S. Patent Number 4,849,360. Apparatus and method for
 confining and decontaminating soil.  Date of Patent July 18,1989. Patent Assignee:  International
 Technology Corporation, forrahce, CA.  10 claims,1 drawing sheet
     I,'- '•; "' '. ;_ „',  r .,  •.   , •;   •     |    .  «,    . _,,'., • -;;•«,,_,   ,;  j ;;,  j ;  >  . ^  •  '•  ..   ,   V ' •  ;   |
 Norris, R.ECj.£t Dey, and D.P. Shine. 1993. The advantages of concerted bioremediation of
 lightly contaminated sites compared to intrinsic bioremediation.  Presented at the American
 Chemical Society I&E Special Symposium. Atlanta, GA.
 Norris, R.D., K. Dowd, and C. Maudlin.  1993. The use of multiple oxygen sources and nutrient
 delivery systems to effect in situ bioemediation of saturated and  unsaturated soils  in hydrocarbon
 bioremediation.  R.E. Hinchee, B.C. Allen, R.E. koeppeJl, and R.N. Miller (eds.). Columbus,
 OH: Battelle Press.
 Norris, R.D., S.S. Suthersam, and f J. Callmeyerr 1990. Integrating different technologies to
 accelerate remediation of multiphase contamination.  Presented at the NWWA Focus Eastern
 Regional Groundwater Conference. Springfield, MA. October.
 Norris, R.D., M.M. Megehee, D.L. Lowe, C.C. Wolf, and P.T. McGee. 1996.  Composting of
 munition contaminated soil: treatability test and a comparison with alternatives. Presented at the
 89th Annual Air and Waste Management Association Meeting, Nashville, TN, June 23-28.
 Norris, R.D., R.E. Hinchee, R. Brown, P.L. McCarty, L. Semprini, J.T. Wilson, D.H.  Kampbell,
 M. Reinhard, E.J. Bouwer, R.C. Borden, T.M. Vogel, J.M. Thomas, and C.H. Ward.  1993. In
 Situ Bioremediation of Ground Water and Geological Material:  A Review of Technologies.  EPA/
 600/R-93/124. Office of Research and Development, Robert S. Kerr Environmental Research
 Laboratory,Ada, OK.
 Norris, R.D., R.E. Hinchee, R. Brown, P.L. McCarty, L. Semprini, J.T. Wilson, D.H.  Kampbell,
 M. Reinhard, E.J. Bouwer, R.C. Borden, T.M. Vogel, J.M. Thomas, and C.H. Ward (eds.).  1994.
 In-situ bioremediation of  soils and groundwater contaminated with petroleum hydrocarbons.  The
 Handbook of Bioremediation.  Boca Raton, PL: Lewis Publishers.
 Nyer, E.K.  1992. Groundwater treatment technology.  Van Nostrand Reinhold, New York,
 New  York.
 Nyer, E.K.  1993. Aquifer restoration:  pump and treat and the alternatives. Ground Water
~ Monitoring Review, pp 89-92. Winter!
 Nyer, E.K., and S.S. Suthersan. 1993.  Air sparging:  savior of groundwater remediation or just
 blowing bubbles in the bath tub? Ground Water Monitoring Remediation.  l3(4):87-91.
 O'Hayre, A.P., M.J. Day,  and J.A.M. Thomson.  1993. Hydrologic characterization of a
 Superfund site: remedial  investigation through remedial action.  Proceedings of the 1993 Georgia
 Water Resources Conference, K. J. Hatcher, Ed., Institute of Natural Resources,  University of
 Georgia, Athens, Georgia. April 20-21. pp. 180-184.

 Olsen, S.R. and L.E. Sommers. 1982. Phosphorus. Methods of Soil Analysis, Part 2. Chemical
 and Microbiological Properties.  A.L. Page (ed.). 2nd edition. Madison, WI: American Society
 of Agronomy.
 Otte,  M.L., C.C. Kearns, and M.O. Doyle. 1995. Accumulation  of arsenic and zinc in the
 rhizosphere of wetland plants.  Bull. Environ. Contam. Toxicol. 55:154-161.

 Ottengraf, S.P.P.  1986. Exhaust gas purification. Biotechnology.  Vol.8: Microbial Degrada-
 tions.  W. Schonbora (Vol. Ed.j, H.J. Rehm, and G. Reed (eds.j. Weinheim, Germany: VCH
 Publishers.

 Ottengraf, S.P.P. and A.H.C. Van Den Oever.  1983.  Kinetics of organic compound removal from
 waste gases with a biological filter. Biotech. Bioengin.  15:3089-3102.
                                                  B.I 6
-------
                                                                          Appendix B
Overcash, M.R., W.L. Nutter, R.L. Kendall, and J.R. Wallace. 1985. Field and Laboratory
Evaluation of Petroleum Land Treatment System Closure. EPA/600/2-85/134.  Office of Research
and Development, Robert S. Kerr Environmental Research Laboratory, Ada, OK.
Page, A.L., R.H. Miller, and D.R. Keeney (eds.).  1982. Methods of Soil Analysis, Pan 2,
Chemical and Microbiological Properties. 2nd edition. American Society of Agronomy, Inc.
Madison, WI:  Soil Science Society of America, Inc.,
Pankow, J.F., R.L. Johnson, and J.A. Cherry. 1993. Air siparging in gate wells, cutoff walls and
trenches for control of plumes of volatile organic compounds (VOCs). Ground Water. 31(4):654-663.

Park, H.S. and R.C. Sims. 1994. Biodegradation of polycyclic aromatic hydrocarbon depending
upon oxygen tension in unsaturated soil. Hazardous Waste Management Handbook.  Englewood
Cliffs, NJ: Prentice Hall.
Parker, J.C., RJ. Lenhard, and T. Kuppusamy.  1987. A parametric model for constitutive
properties governing multiphase flow in porous media.  Water Resources Res. 23:618-624.

Parsons Engineering Science. 1991. Bioventing Pilot Test Results Report. Prepared for
Burlington Northern Railroad. Denver, CO. June.
Paterson, S., D. Mackay, D. Tarn, and W.Y. Shiu. 1990. Uptake of organic chemicals by plants: a
review of processes, correlations and models. Chemosphere. 21:297-331.

Pavlostathis, S.G. and P. Zhuang. 1993. Reductive dechlorination of chloroalkenes in micro-
cosms developed with a field contaminated soil. Chemosphere.  27(5):585-95.
                                                 i
Peal, R.G. and A. Benedek.  1980. Attainment of equilibrium in activated carbon isotherm
studies.  Environ. Sci. Technol.  14(1):66-7I.
Peterson, R.G. and L.D. Calvin.  1986.  Sampling. Methods of Soil Analysis,  Part 1:  Physical
and Mineralogical Properties. A. Klute (ed). 2nd edition. American Society of Agronomy.
Madison, WI:  Soil Science Society of America, Inc.
Phelps, M.B., F.J. Stanin, and D.C. Downey.  1995.  Long-term bioventing performance in low
permeability soils.  In Situ Aeration: Air Sparging, Bioventing, and Related Remediation
Processes. R.E. Hinchee, R.M. Miller, and P.C. Johnson (eds.).  Columbus, OH:  Battelle Press.
3(2):272-282.
Pignatello, J.J.  1986. Ethylene dibromide mineralization  in soils under aerobic conditions.  Appl.
Environ. Microbiol. 51:58-592.
Pignatello, J.J., C.R. Frink, P.A. Marin, and E.X. Drostb.  1990.  Field-observed ethylene
dibromide in an aquifer after two decades. /.  Contam. Hydrol. 5:195-214.

Piotrowski, M.R. 1991. Full-scale, in situ bioremediation at a Superfund site: a progress report.
Hydrocarbon contaminated soils and groundwater: analysis,  fate, environmental and public health
effects, and remediation. Second Annual West Coast Conference. EPACH Corporation, Newport
Beach, CA. March 6.
Piotrowski, M.R., J.R. Doyle, D. Cosgriff, and M.C. Parsons. 1994. Bioremedial progress at the
Libby, Montana, Superfund site.  Applied Technology for Site Remediation. R.E. Hinchee, D.B.
Anderson, F.B. Metting, Jr., and G.D. Sayles (eds.).  Boca Raton, FL: Lewis Publishers.
Piwoni, M.D. and J.W. Keeley.  1990.  Basic Concepts of Contaminant Sorption at Hazardous
Waste Sites.  Ground Water Issue Paper. EPA/540/4-90/053. Office of Research and Develop-
ment, Robert S. Kerr Environmental Research Laboratory, Ada, OK.

Pope, D.F. and J.E. Matthews. 1993. Bioremediation Using the Land Treatment Concept.  EPA/
600/R-93/164.  Office of Research and Development, Robert S. Kerr Environmental Research
Laboratory, Ada, OK.                              '
                                         B.17
-------
              List of References
              Prokop, W.H. .and H.L. Bohn. 1985. Soil bed system for control of rendering plant odors. /. Air
              Pollution Control Assoc.  35(12).
              Proseri, BJ., W.M. Korreck, and J.M. Armstrong.  1992. Design and preliminary results of a full-
              scale bioremediation utilizing an on-site oxygen generator system.  In Situ Bioredamation:
              Application and Investigation for Hydrocarbons and Contaminated Site Remediation.  R.E.
              Hinchce and R.F. Olfenbuttle (eds.).  Stoneham, MA: Butterworth-Heinemann.

              Qiu,X. 1995a.  Field Scale Pilot Study: Vegetation Enhanced Soil Bioremediation 1993-1994
              Progress Report. Health Safety and Environmental Technology Center, Union Carbide Corporation.

              Qiu, X. 1995b. Grass-Enhanced Biodegradation of' Benzo(a)pyrene and Hexachlorobiphenyl in
              Soil"—A Laboratory Treatability Study for SoifRemediationat Bound Brook Plant. Health Safety
              And Environmental Technology Center, Union Carbide Corporation.

            '  Qiu, X., S.I. Shah, E.W. Kendall, D.L. Sorensen, R.C. Sims, and M.C. Engelke.  1994. Grass-
              enhanced bioremediation for clay soils contaminated with polynuclear aromatic hydrocarbons.
              Bioremediation Through Rhizosphere Technology. T.A. Anderson and J.R. Coats (eds.). ACS
              Symposium Series.  Volume 563. Washington, DC: American Chemical Society.

              Raskin, I., N.P.B.A. Kumar, S. Dushenkov, M.J. Blaylock, and D. Salt. 1994. Phytoremediation
              — Using plants to clean up soils and waters contaminated with toxic metals. Abstract: Emerging
              Technologies in Hazardous Waste Management VI. American Chemical Society Industrial &
              "Engineering Chemistry Division Special Symposium. Atlanta, GA. September 19-21.

              Ratz, J.W., P.R. Guest, and D.C. Downey. 1994.  Results of a two-year in situ bioventing
              demonstration.  Proc. 18th Annual Army Environmental R&D Symposium. Williamsburg, VA.

              Raymond, R.L. 1974. Reclamation of Hydrocarbon Contaminated Ground-waters. U.S. Patent
              3,846,290.
              Raymond, R.L., V.W. Jamison, and J.O. Hudson.  1976. AIChE Symposium Series. 73:390-404.

              Raymond, R.L..R.A. Brown, R.D. Norris, and E.T. O'Neill.  1986.  Stimulation of Biooxidation
              Processes in Subterranean Formations. U.S. Patent 4,588,506.

              Raymond, R.L., V.W. Jamison, J.O. Hudson, R.E. Mitchell, and V.E. Farmer.  1978. Field
              Application of Subsurface Biodegradation of Hydrocarbon  in Sand Formation. Project
              No. 307-77. American Petroleum Institute, Washington, DC.
              Reddy, K.R. and W.F. De Busk. .1985.  Nutrient removal potential  of selected aquatic macro-
              phytcs. J. Environ. Qual.  14:459-462.
              Reeves, R.D. and R.R. Brooks.  1983. Hyperaccumulation  of lead and zinc by two  metallophytes
              from mining areas of central Europe.  Environ. Pollut. Ser. A.  31:277-285.
              Reinhard, M. 1994. In situ bioremediation technologies for petroleum derived hydrocarbons
              based on alternative electron acceptors (other than molecular oxygen). The Handbook of
              Bioremediation. Norns, R.D., R.E. Hinchee, R. Bro^
              D.H! Kampbeii, M. Reinhard, E.J. Bouwef, R.C7 Borden, T.M. Vogei, J.W. Thomas, and C.H.
              Ward (eds.). Boca Raton,  FL: Lewis Publishers.
              Reisinger, H.J., S.A. Mountain, P.A. Montney, A.S. Hoilman, and A.W. Darnell.  1995. Pressure
              dewatering:  an extension of bioventing technology. In Situ Aeration: Air Sparging, Bioventing,
              and Related Remediation Processes. R.E. Hinchee, R.M. Miller, arid P.C. Johnson  (eds.).
           -   Columbus, OH: Battelle Press.  3(2):267-276.
             ReTec Corporation.  (No date).  Closure of creosote and pentachlorophenol impounds. Company
             literature. Cited in U.S. EPA 1990c.
             Rhodes, I.A.L., E.M. Hinojosa, D.A. Barker, and R.A. Poole.  1996^ Conventional TPH pitfalls.
             Environmental Lab.  December/January.
                                                                        I

..   -   '...:                                            B.18
•I'll!.
-------
                                                                         Appendix B
Rifai, H.S., P.B. Bedient, R.C. Borden, and J.F Haasbeek.  1989. Bioplume II.  Computer Model
of Two-Dimensional Contaminant Transport Under the Influence of Oxygen Limited Biodegrada-
tion in Ground Water. Users manual version 1.0, preprocessor source code version l.O.source
code version 1.0. EPA/600/S8-88/093. Office of Research and Development, Robert S. Kerr
Environmental Research Laboratory, Ada, OK.
Rifai, H.S., R.C. Borden, J.T. Wilson, and C.H. Ward.  1995. Intrinsic Bioattenuation for
Subsurface Restoration. In Hinchee, R.E., J.T. Wilson, and D.C. Downey (eds.) Intrinsic
Bioremediation. Battelle Press, Columbus, OH. Pp. 1-30.
Riser-Roberts, E.  1992. Bioremediation of Petroleum. Contaminated Sites. Boca Raton, PL:
C.K. Smoley.
Rittman, B.E., A.J. Valocchi, E. Seagren, C. Ray, and B. Wrenn. 1992.  Critical Review of In
Situ Bioremediation.  GRI-92-0322. Gas Research Institute, Chicago, IL. 185 pp.

Rogers, J.A., D.T. Tedaldi, and M.C. Kavanaugh. 19513. A screening protocol for
bioremediation of contaminated soil. Environ. Progress. 12(2): 146-156.

Roszak, D.B. and R.R. Colwell.  1987. Survival strategies  of bacteria in the natural environment.
Microbiol. Rev. 51{3):365-379.
Russell, H.H., I.E. Matthews, and G.W. Sewell.  1992.  TCE Removal from Contaminated Soil
and Ground Water. EPA 540/S-92/002.  Office of Research and Development, Robert S. Kerr
Environmental Research Laboratory, Ada, OK.

Rutherford, K. and P.C. Johnson.  1995.  Interfacial mass transfer rates during in situ air sparging
— effects of process control changes and lithology. Proc. Petroleum Hydrocarbons and Organic
Chemicals in Ground Water: Prevention, Detection and Restoration. API/NGWA/STEP.
Houston, TX. November.                         ;
Salt, D.E., M. Blaylock, N.P.B.A. Kumar, V. Dushenkov, B.D. Ensley, I. Chet, and I. Raskin.
1995. Phytoremediation:  a novel strategy for the removal of toxic metals from the environment
using plants. Biotechnology. 13:468-474.
Samson, R., C.W. Greer, and J. Hawari.  1992.  Demonstration of a New Biotreatability Protocol
to Monitor a Bioprocess for the Treatment of Contaminated Soils. National Research Council
final report of the collaborative project among the Biotechnology Research Institute, Shell
Products Canada Limited, Groundwater Technology Canada Limited, and the St. Lawrence
Centre (Environment Canada).
Sandhu, S.S., B.S. Gill, B.C. Casto, and J.W. Rice. 1991. Application of Tradescantia
micronucleus assay for in situ evaluation of potential genetic hazards from exposure to chemicals
at a wood-preserving site.  Haz. Waste Haz. Mat. 8:257-262.
Sayles, G.D., A. Leeson, R.E. Hinchee, C.M. Vogel, R.C. Brenner, and R.N Miller. 1995.  Cold
climate bioventing with soil warming in Alaska. In Situ Aeration:  Air Sparging, Bioventing, and
Related Remediation Processes. R.E. Hinchee, R.M. Miller, and P.C. Johnson (eds.).  Columbus,
OH: Battelle Press.
                                                         .
Sayles, G.D., R.C. Brenner, R.E. Hinchee, C.M. Vogel, and R.N. Miller. 1992. Optimizing
Bioventing in Deep Vadose Zones and Moderate Climates:  Hill AFB Bioremediation of a JP-4
Spill.  Bioremediation of Hazardous Wastes. EPA/600/R-92/126. Office of Research and
Development, Washington, DC.
SBP Technologies, Inc. 1994. Final Report, S-l Soils Bioremediation Demonstration Pre-
Construction Laboratory Testing Results, Kopper's Feather River Site, Oroville, California.
                                         B.19
-------
                                          List of References
I, ;  [,         ,      > '  i       '   i",  ' 'it    , ,;:;,  •<>.  -     -            .             i   •  ,     , j	(,"
                                          Sctieunert, I., Z. Qiao, and F. Korte.  1986. Comparative studies of the fate of atrazine-14C and
                                          peritachlorophenol"14C in various laboratory and outdoor soil-plant systems. J. Environ. Science
                                          Health. 621:457-485.
                               » Iliiirll  II ,,!1' i
                               ' . .  t:	)l
                                           Schirna, S. D.J. LaBrecque, and P. Lundegard. 1994. Using resistivity tomography to monitor air
                                           sparging! Proc. Symposium, on Applications of Geophysics to Engineering land Environmental
                                           Problems (SAGEEP). Boston, MA.  Englewood, CO:  Environmental Engineering and Geophys-
                                          'ics Society.
                                            ,1,  ,.'. i"'!;.:,,!	 	  	  i  ,,;.   , .  	 r, ,   |  ., ..;,.  ; | _, .,,,_, ' I ;, J • I        .    '     . ,	  ; . j ';';,'"
                                           Schmidt, E.L. and E.A. Paul.  1982.  Microscopic methods for soil microorganisms. Methods of
                                           Soil Analysis, Part 2: Chemical and Microbiological Properties. 2nd edition. American Society
                                           of Agronomy, Inc. Madison, WI:  Soil Science Society of America, Inc.

If,:"  ;  . '    i  "'   '>•	"•       "'	"•;	I'l1'  .''}\.     Schnoor, J.L., L.A^cht, S.C. Mc^
                                           Phytoremediation of organic and nutrient contaminants. Environ. Sci. Technol. 29:318A-323A.
'" i.      '          •>	'      i    "•'.'•   •'    .          '.	 .	i  .	,'.r-.-ii	 ..  r-.	•	 i i .••)••'.',.	i	  ••  ">!-•"":,•„,	-!v	
                                           Schnurer, J. and J. Rosswall. 1982.  Fluorescein diacetate as a measure of total microbial activity
                                           in soil and litter. Appl Environ. Micro. 43:1256-1261.
                                           Scholl, M.A. and R.W. Harvey.  1992. Laboratory investigations on the role of sediment surface
                                           and groundwater chemistry in transport of bacteria through a contaminated sandy aquifer.
                                           Environ. Sci. Technol. 26(7):1410-1417.
                                           Scholl, M.A., A.L. Mills, J.S. Herman, and G.M. Homberger. 1990. The Influence of Mineralogy
                                           and Solution Chemistry on the Attachment of Bacteria to Representative Aqutfer Materials.
                                           Department of Environmental Sciences, Charlottesville, VA.
                                           Schwab, A.P and M.K.Banics.  1994.  Biologically mediated dissipation of polyaromatic
                                           hydrocarbons in the root zone. Bioremediation Through Rhizosphere Technology.  T.A. Anderson
                                           and J.R. Coats (eds.). ACS Symposium Series.  Volume 563. Washington, DC:  American
                                           Chemical  Society.
11  ' :    '           	         ""      '•       "	"'             "       •          	     ' •  ']  "  '	'	
                                           Scribner, S.L., T.R. Benzing, S.  Sun, and S.A. Boyd. 1992.  Desorption and bioavailability  of
                                           aged simazine residues in soil from a continuous corn field.  J. Environ. Qual.  21:115-120.
                                           Semple, K.T. and T.R. Fermor.  1995.  Composting systems for the bioremediation of chlorophe-
                                           nol-cpntaminated land. R.E. Hinchee, R.E. Hoeppel, and D.B. Andlerson (Eds.).  Bioremediation
                                           of Recalcitrant Organics, Columbus, OH:  Battelle Press,  p.93-100.
                                           Semprini,  L., D. Grbic' Galic', P.L. McCarty, and P.V. Roberts.  1992. Methodologies for
                                           Evaluating In Situ Bioremediation of Chlorinated Solvents. EPA/600/R 92/042.  Robert S. Ken-
                                           Environmental Research Laboratory, Ada, OK.

                                           Semprini,  L., P.V. Roberts, G.D. Hopkins, and P.L. McSarty. 1990. A field evaluation of in-situ
                                          biodegradation of chlorinated ethenes:  part 2, results of biostimula'tion and biotransformation
                                          experiments. Ground Water. 28(5):715-727.
                                          Semprini,  L., G.D. Hopkins, P.V. Roberts, D. Grbic'-Galic', and P.L. McCarty. 1991. A field
                                          evaluation of in-situ biodegradation of chlorinated ethenes: part 3, studies of competitive
                                          inhibition. Ground Water. 29(2):239-249.
                                          Shelton, D.R.  1990.  Development and Validation of aLaboratory Screening Protocol for
                                          Assessing  the Efficacy of Aerobic Biodegradation. Final report for Project #i27b,12130,0{J2,03R.
                                          United States Department of Agriculture, ARS, NRI, Pesticide Degradation Laboratory, BARC
,;«!;•    '          ;               '    ;   '  West, Beltsyille, MD.   '	" ^    	'	^    ,  '  	'"^'  '',"   	"J    _  „",/'',,,^1 "„,'",	
.(•"     ; i          it         "    .,.           ;      	j  ,   /i  .'. ;..  „	 ;  ii,       •           ;. 'I? ., ' !,;' r •    i '• I       -':., i MM ,,     . •    ., • I MJ,
                                          Sheppard,  S.C. and W.G. Evenden.  1985. Mobility and uptake by plants of elements placed near
                                          a shallow water table interface. /. Environ. Qual.  14:554-560.
                                                                                   B.20
-------
                                                                         Appendix B
Shimp, J.F., J.C. Tracy, L.C. Davis, E. Lee, W. Huang,! L.E. Erickson, and J.L. Schnoor. 1993.
Beneficial effects of plants in the remediation of soil and groundwater contaminated with organic
materials. Crit. Rev. Environ. Sci.Technol. 23:41-77.
Simpkin, T.  1996.  Personal communication with Scott Ruling.  CH2M Hill, Denver, CO.

Sims, J.L., J.M. Suflita, and H.H. Russell.  1992.  In-Situ Bioremediation of Contaminated Ground
Water. EPA/540/S-92/003. Office of Research and Development, Robert S.  Kerr Environmental
Research Laboratory, Ada, OK.
Sims, J.L., R.C. Sims, and I.E.  Matthews.  1989.  Bioremediation of Contaminated Surface Soils.
EPA/600/9-89/073. Office of Research and Development, Robert S. Kerr Environmental
Research Laboratory, Ada, OK.
Sims, R.C.  1986.  Loading rate .and frequencies for land treatment systems. Land Treatment: A
Hazardous Waste Land Treatment Alternative. R.C. Lpehr and J.F.  Malina, Jr. (eds.). Water
Resources Symposium No. 13.  Center for Research in Water Resources, The  University of Texas
at Austin.
Sims, R.C.  1990.  Soil remediation techniques at uncontrolled hazardous  waste sites: a critical
review. /. Air Waste Management Assoc.  40(5):703-732.
Sims, R.C., D.L. Sorensen, WJ. Doucette, and L.L. Hastings.  1986a.  Waste/Soil Treatability
Studies: Methodologies and Results. Waste Loading Impacts on Soil Degradation. Transforma-
tion, and Immobilization. Volume 2. EPA/600/6,86/003b. Robert S. Kerr Environmental
Research Laboratory, Ada, OK.
Sims, R.C., J.L, Sims, D.L. Sorensen, and L.L. Hastings.  1986b. Waste/Soil Treatability Studies:
Methodologies and Results. Volume 1. EPA/600/6,86/003a. Roberts. Kerr Environmental
Research Laboratory, Ada, OK.
Sims, R.C., WJ. Doucette, I.E. McLean, W.J. Grenney, and R.R. Dupont. 1988. Treatment
Potential for 56 EPA Listed Hazardous Chemicals in Soil.  EPA/600/6 88/001. Robert S. Kerr
Environmental Research Laboratory, Ada, OK.
Sims, R.C., J.L. Sims, and S.G. Hansen. 1991. Soil Transport and Fate Database 2.0 and Model
Management System.  Office of Research and Development, Robert  S. Kerr Environmental
Research Laboratory, Ada, OK.
Sims, R.C., J.E. Matthews, S.G. Huling, B.E. Bledsoe, M.E. Randolph, and D. Pope.  1993.
Evaluation of full-scale in situ and ex situ bioremediatiion of creosote wastes in soils and ground
water.  Annual Symposium on Bioremediation of Hazardous Wastes: Research, Development, and
Field Evaluations. US EPA, Dallas, TX.   May 4-6.
Sims, R.C., J.L. Sims, D.L. Sorensen, O.K. Stevens, S.G. Huling, B.E. Bledsoe, J.E. Matthews,
and D. Pope. 1994. Performance evaluation of in situ and ex situ bioremediation of creosote
wastes in ground water and soils. Annual Symposium on Bioremediation of Hazardous  Wastes:
Research, Development, and Field Evaluations.  US EPA, San Francisco,  CA. June 28-30.

Sims, R.C., J.L. Sims, D.L. Sorenson, and J.E. McLean.  1995a.  Champion International
Superfund Site, Libby, Montana: Bioremediation Field Performance Evaluation of the Prepared
Bed Land Treatment System. Volume 1.  EPA-600/R-95/156a. National Risk Management
Laboratory, Subsurface Protection and Remediation Diivision, Ada, OK.

Sims, R.C., D.L. Sorensen, J.L. Sims, and J.E. McLean. 1995b.  Champion International
Superfund Site, Libby, Montana: Bioremediation Field Performance Evaluation of the Prepared
Bed Land Treatment System. Volume II - Figures and Tables. EPA-600/R-95/156b. National
Risk Management Laboratory,  Subsurface Protection and Remediation Division, Ada, OK.
                                         B.21
-------
                        List of References
                        Sims, R.C., J.L. Sims, D.L. Sorensen, J.E. McLean, and S.G. Huling. 1995c.  Prepared bed reactor for
                        full-scale remediation of soil contaminated with wood preserving wastes: field bioremediation
                        evaluation. Proc. 21st Annual Risk Reduction Engineering Laboratory Research Symposium. EPA/
                        600/R-95/012.  Office of Research and Development, Cincinnati, OH.  April 4-6.

                        Song, H.G. and R. Bartha.  1990. Effects of jet fuel spills on the microbial community of soil.
                        Applied Environ. Micro.  56:646-651.
                       '	      ,   i ' •"   '	,   ;,, , .  ''i,'1 '•  •' .'i,1'1'11''" IP '";'  ly':';1!""'!!, r'/vL'I1 T ,i,'|'mi	' i',.1'"" ",i!J	 > jiVi! :»i • i. ,,'„»„  I;,*" '/ill pillS111.
                        Srinivasan, P. and J.W. Mercer.  1987.  BIOJD:  One Dimensional Model for Comparison of
                        Biodegradation and Adsorption Processes in Contaminant Transport. Users manual and
                        documentation for version  1.0. HerndonsyA:GeoTrans, Inc.

                        Srinivasan, P. and J.W. Mercer.  1988.  Simulation of biodegradation and sorption processes in
                        ground water.  Ground Water. 26:475-487.
       Stanler, R.Y., J.L. Ingraharn, M.L. Wheelis, and P.R. Painter. 1986. The Microbial World. 5th
       edition. Englewood ciiffs.NJ: Prentice Hall.
v'd ,	   ,.  I1,  ';;.'/ •  *  '  "•':<.', >''*••-. • ' *;.w 	^ •:.;,•,..	K':;  ,  ?':, 1 (•:*(>< ""  i1"1   , *:,•; •  ,  ';::  ..•••rifft.:
       Steinberg, S.M., JJ. Pignatello, and B.J. Sawhney. 1987. Persistence of 1,2-dibrornoethane in
       soils: entrapment in intraparticle micropores. Environ. Sci.  Technol. 21:1201-1208.

       Steward, K.J. and H.D. Laford. 1995. Case study: bioremediation in the Aleutian Islands.
       Biological Unit Processes for Hazardous Waste Treatment. R.E. Hinchee, G.D. Sayles, and R.S.
       Skeen(eds.).  Columbus, OH:  Battelle Press.
       Suflita, J.M., A. Horowitz, D.R. Shelton, and J.M. Teidje.  1982. Dehalogenation: A novel
       pathway for'the anaerobic biodegradation of haloaromatic compounds. Science. 218:1115-1116.
       Symons, B.D. and R.C. Sims.  1988. Assessing detoxification of a complex hazardous waste,
       using the Microtox Assay. Archives Environ. Contam. Toxicology.  17:497-505.

       Tabatabai, M.A. 1982.  Soil enzymes. Methods of'Soil'Analysis, Pan(2: Chemical and Microbio-
       logical Properties.  2nd edition. American Society of Agronomy, Inc.  Madison, WI:  Soil
       Science Society of America, Inc.
       Taylor, J.K. 1987. Quality Assurance of Chemical Measurements, Lewis Publishers Inc., Chelsea, MI.

       Thibodeaux, L.J. 1979. Chemodynamics:  Environmental Movement of Chemicals in Air Water
       and Soil. New York: John Wiley & Sons.

       Thomson, J.A.M., M.J. Day, R.L. Sloan, and M.L. Collins.  1995. In situ aquifer bioremediation
       at the French Limited Superfund Site. Applied Bioremediation of Petroleum Hydrocarbons. R.E.
       Hinchee, J.A. Kittel, and H.J. Reisinger (eds.). Columbus, OH:  Battelle Press.
       Traverse Group. 1993. Air Sparging Interdiction Field for Containment of Contaminated
       Grpundwater Plumes. Final Report for Contract Fo86.35-92-C-b064, submitted to Departmentof
       Air Force, Elgin AFB, Florida.

       University of Cincinnati. 1993. Screening Protocol for Evaluating Biotreatability Potential of
       Contaminated Soils. US EPA Contract 68-C9. Department of Civil & Environmental Engineer-
       ing, Center Hill Facility, Cincinnati, OH.

       U.S. Department of Energy (DOE).  1991.  Cleanup o/'VOCs in Non-Arid Soils — The Savannah
       River Integrated Demonstration. WSRC-MS-91-290, Rev. 1. Washington, D.C.
,.''':;[       ,  -'  i1""''  ' , I- •  .;i '•',:  '  • ""'*;  i (','                 mi i     i i  In      i:,  iti't  ' .."-.i  '!i';'«!,K*f	
       U.S. Departmentof Energy (DOE).  1994. Summary Report ofa Workshop on Phytoremediation
       Research Needs. DOE/EM-Q224.  Washington, D.C.

       US. EPA. 1980. Interim Guidelines and Specifications for'Preparing Quality Assurance'Project
       Plans. QAMS-005/80. Office of Solid Waste and Emergency Response! Washington, DC.
                                                                                                                  ,y i "M!
                                                                                                         ,,.'"•.ill", ' 1 ! It
	
      US EPA.  1982.  Test Methods for Evaluating Solid Waste, Physical, Chemical Methods.  2nd
      edition. SW-846. Office of Solid Waste and Emergency Response, Washington, DC.
                                                                 B.22
                                                                                   1
-------
                                                                        Appendix B
US EPA.  1983. Hazardous Waste Land Treatment. EPA/53Q-SW-874. Municipal Environmen-
tal Research Laboratory, Cincinnati, OH.
US EPA.  1984a. Permit Applicants Guidance Manual for Hazardous Waste Land Treatment,
Storage and Disposal Facilities, Final Draft. EPA/530/SW-84-004. Office of Solid Waste and
Emergency Response, Washington, DC.
US EPA. 1984b. Review of In-Place Treatment Techniques for Contaminated Surface Soils. Volumes
1 and 2. EPA/540/2-84-003 A and B. Office of Research smd Development, Washington, DC.
US EPA.  1984c. Methods for Chemical Analysis of Water and Waste.  EPA-600/4-84-017.
Office of Research and Development, U.S. Environmental Protection agency, Cincinnati, Ohio.
US EPA.  1985a. Covers for Uncontrolled Hazardous Waste Sites. EPA/540/2-85/002. Office of
Research and Development, Washington, DC.
US EPA.  1985b. Remedial Action at Waste Disposal Sites.  EPA/625/685/006. Office of
Research and Development, Washington, DC.
                                                 I
US EPA.  1986a. RCRA Ground-Water Monitoring: Technical Enforcement Guidance Document
(TEGD). OSWER-9950.1.  Office of Solid Waste and Emergency Response, Washington, DC.
US EPA.  1986b. Permit Guidance Manual on Unsaturated Zone Monitoring for Hazardous
Waste Land Treatment Units.  EPA/530-SW-86-040. Office  of Solid Waste and Emergency
Response, Washington, DC.
US EPA.  1986c. Test Methods For Evaluating Solid Waste,  Vol. 1A. 3rd edition.  SW-846.
Office of Solid Waste and Emergency Response, Washington, DC.
US EPA.  1986d. Test Methods For Evaluating Solid Waste, Vol. IB. 3rd edition . SW-846.
Office of Solid Waste and Emergency Response, Washington, DC.
US EPA.  1986e. Permit Guidance Manual for Hazardous Waste Land Treatment Demonstra-
tions. EPA/530/SW-86-032. Office of Solid Waste and Emergency Response, Washington, DC.
US EPA. 1987. Guidance Manual on Hazardous Waste Land Treatment Closure/Post-Closure. EPA
OSWER Directive 9476.00-9. Office of Solid Waste and Emergency Response, Washington, DC.
US EPA.  1988a. Interaction Simulation of the Fate of Hazardous Chemicals During Land
Treatment of ystes: RITZ Users Guide. EPA/600/8-88-001.  Office of Research and Develop-
ment, Robert S. Kerr Environmental Research Laboratory, Ada, OK.
US EPA. 1988b. Guide to Technical Resources for the Design of Land Disposal Facilities. Risk
Reduction Engineering Laboratory, Center for Environmental Research Information, Cincinnati,
Ohio. EPA/625/6-88/018. 63pp.                     |
US EPA. 1988c.  Interim Protocol for Determining the Aerobic Degradation Potential of
Hazardous Organic Constituents in Soil. US EPA Scientific  Steering Committee, Biosystems
Technology Development Program, and Soil Treatment Processes Committee.  Robert S. Kerr
Environmental Research Laboratory, Ada, OK.
US EPA. 1988d. Record of Decision:  Libby Ground Water  Superfund Site, Lincoln County,
Montana.  Region VIII, Montana Operations Office, Helena,  MT.
US EPA. 1989a.  Corrective Action: Technologies and Applications. Seminar publication. EPA/
625/4-89/020. Center for Environmental Research Information.  Cincinnati, OH.
US EPA, 1989b. Methods for Evaluating the Attainment ofCleanup Standards Volume 1: Soils and
Solid Media. EPA 230/02-89-042. Office of Policy, Planning, and Evaluation, Washington, DC.
US EPA. I989c.  Test Method. The Determination or Inorganic Anions in Water by Ion
Chromatography—Method 300.0. Office of Research and Development, Municipal Environmen-
tal Research Laboratory, Cincinnati, OH.

                                        B.23
-------
           List of References
           US EPA  1990a. Measurement of Hydraulic Conductivity Distributions: A Manual of Practice.
           EPA/600/8-90/046. Robert S. Kerr Environmental Research Laboratory, Ada, OK.
           US EPA. I990b. Guide for Conducting Treatability Studies Under CERCLA Aerobic Biodegra-
           dalion'lMboratory Screening.  Risk Reduction Engineering Laboratory, Cincinnati, OH, and
           dffice of Emergency and Remedial Response, Office of Solid Waste and Emergency Response,
           Washington, DC.
1     I    ,  ; ,,  " ''  , •  ">!  ^.'"4 ......... .: ..... ., "  . i ...... '•»    •• •• ,-i; ....... „: iff |i.  ..a ..... ,; ..... iftv \ 'fls  ":• i   •. ..... i«t ..... i"1"   •"» IT' i- ', i 'Si1' «"
           US EPA. l99Qc^ Engineering bulletin. Slurry blade gradation. EPA/540/2-907016. US EPA,
           Office of Emergency and Remedial Response, Washington, D.C., and Office of Research and
           1 - ',i!'!,!!!,,,,!" ' .,:  I Ji'l ..... ..... .," ......... r „,..,,„,   ,  .....   ................. , ........... , , ............... , ,   ..... ,      .........      „    ........
           Development, Cincinnati, Ohio.
           US EPA. 1992a. Methods for Evaluating the Attainment of Cleanup Standards, Volume 2:
           Ground Water. EPA/230/R-92-014.  Office of Policy, Planning, and Evaluation, Washington, DC.
          "|. '•  ..... I11!,,!1 „ " i,;, ii!!':: ,  ,'1"'1    ,     " :•„ ii •'' , '     • ",„,.' <"' ."''"  !"  '" ,  .''".i,"11!!,,::"1 '" '  ':.Ji i 'it!. ' '"jf ' 4i!', •    '"„!'•  :•''• '» ,n', •  ' •l"'i|' J!' 'JiVI'1 'j \ ' f U f'j '
           US EPA* i_992b;; A Citizen's Guide to Air Sparging. EPA/542/F-92/010.  Office of Solid Waste
           and Emergency Response, Washington, DC.
           US EPA. 1993a. Data Quality Objectives Process for Superfund: Interim Final Guidance,  EPA
           540-R-93-071.
           US EPA. 1993b. Guide for Conducting Treatability Studies Under CERCLA: Biodegradation
           Remedy Selection, Interim Guidance, Superfund Innovative Technology Evaluation. EPA/540/R
           93/519a.  Risk Reduction Engineering Laboratory, Cincinnati, OH, and Office of Solid Waste and
           Emergency Response, Washington, DC.
           USEPAl 1993c. Remeliatlon Technologies Screening Matrix arid Guide.  EPA 542-B-93-06^.
           U.S. EPA, Solid Waste and Emergency Response (OS-1 10W), Technology Innovation Office,
           Washington, D.C.
           US EPA- 1994a.  Data Quality Objectives Decision Errors Feasibility Trials (DEFT) Software, V.
           4.0, EPA/600/R-96/056, September 1994.
           US EPA. 1994b. Remediation Technologies Screening Matrix and Reference Guide. 2nd edition.
           EPA/542/B-94/013. D.D Environmental Technology Transfer Committee, Washington, DC.

           US EPA. 1995a. '"Guidance for' the Data Quality Objectives P^
 ,;,"   ':,:     September, 1995.  '   '""'"   '"    ....... '"" '  ''   ""  "'"''"  !  '" !!  "!'  ' "     .......  " ...... ........

           US EPA. .19951). Guidance for the Preparation of Operating Procedures for Quality- Related
           US'EPX.' 1995c. How to Evaluate Alternative Cleanup Technologies for Underground Storage
           Tank Si,tes, a Guide for Corrective Action Plan Reviewers.  EPA510-B-95-007. Office of Solid
           WaSte and Emergency Response, 5403W, Washington, DC.
           US EPA. 1996a. A Citizen's Guide to Phytoremediation. EPA 5^I2-F-96-014. Technology
           Innovation Office, Office of Solid Waste and ^Emergency Response, Washington, DC.
           US EPA. 1996b. Guidance for the Data Quality Assessment Process: Practical Methods for Data
           Analysis, EPA/600/R-96/084, July, 1996.
           US EPA. 1997a. EPA Requirements for Quality Assurance Project Plans, QA/R-5, Draft,
           October, 1997.
           US EPA. 1997b. Guidance on Quality Assurance Project Plans, QA/G-5, Draft, September, 1997.

           USAF.  1993. OU-2 Remediation Investigation/Feasibility Study, Feasibility Study DRaft Report.
           Eielson AFB, Alaska.  March
           USAF.  1994. OUs 3, 4, 5 RI Report - Vol. I. Eielson AFB, AK.
                                                   B.24
-------
                                                                          Appendix B
 USGS. 1977. National Handbook of Recommended Methods for Water Data Acquisition. Office
 of Water Data Coordination, Reston, VA.
 Utah Water Research Laboratory. 1995. Large-Scale Field Demonstration Project: Use of
 Bioventing Technology to Remediate Petroleum Hydrocarbons at Fire Protection Training Area
 1, F.E. Warren AFB, Cheyenne, WY. Final report submitted to Battelle Memorial Institute,
 Columbus, OH.
 Utah Water Research Laboratory. 1997. Intrinsic Remediation Engineering Evaluation/Cost
 Analysis For Site 13/26, Eielson Air Force Base, Alaska.  Technical Report Submitted to the Air
 Force Center for Environmental Excellence, Brooks Air Force Base, San Antonio, TX.

 Vajpayee, P., U.N. Rai, S. Sinha, R.D. Tripathi, and P. Chandra.  1995.  Bioremediation of
 tannery effluent by aquatic macrophytes. Bull. Environ. Contam. Toxicol. 55:546-553.

 van Loosdrecht, W.N., J. Lyklema, and AJ.B. Zehnder,, 1990. Hydrophobic and electrostatic
 parameters in bacterial sdhesion. Aquatic Sciences. 52(1):103-114.

 van Loosdrecht, M.C.M., J. Lyklema, W. Norde, G. Schraa, and AJ.B. Zehnder.  1987a.
 Electrophoretic mobility and hydrophobicity as a measure to predict the initial steps of bacterial
 sdhesion.  Applied Environ. Micro.  53(8):1898-190l.
 van Loosdrecht, M.C.M., J. Lyklema, W. Norde, G. Schraa, and AJ.B. Zehnder.  1987b. The role
 of bacterial cell wall hydrophobicity in adhesion. Applied Environ. Micro.  53(8):1893-1897.

 VDI. 1991. Richtlinien 3477. Biological waste gas/waste air purification, biofilters, VDI-
 Handbuch Reinhalten der Luft, Band 6.
 Vidic, R.D., M.T. Suidan, U.K. Traegner, and G.F. Nakhla. 1990. Adsorption isotherms:  illusive
 capacity and role of oxygen. Water Resources. 24(10): 1187-1195.
 Vogel, T.M., C.S. Griddle, and P.L.  McCarty.  1987. Transformation of halogenated aliphatic
 compounds. Environ. Sci. Technol.  21(8):722-736.
 Vroblesky, D.A.  and T.M. Yanosky. 1990.  Use of tree-ring chemistry to document historical
 ground-water contamination events. Ground Water. 28:677-684.
 Wackett, L.P. and D. L. Allen. 1995.  Comment on "bioremediation in the rhizosphere." Environ.
 Sci. Technol. 29:551-552.
 Walton, B.T. and T.A. Anderson.  1990. Microbial degradation of trichlorethylene in the
 rhizosphere: potential application to biological remediation of waste sites. Appl. Environ.
 Microbio. 56:1012-1016.
 Watanabe, M.E.  1997.  Phytoremediation on the brink of commercialization. Environ. Sci. Tech.
 31:182-186.
 Ward, Calvin H., Raymond C. Loehr, Evan K. Nyer, Michael R. Piotrowski, J. Michele Thomas,
 James C. Spain, John T. Wilson, and Robert D. Norris. 1995.  Innovative Site Remediation
 Technology — Bioremediation. Annapolis, MD: American Academy of Environmental Engineers.

 Wiedemeier, T.H., M.A. Swanson, J.T. Wilson, D.H. Kambell, R-N. Miller, and J.E. Hansen. 1995.
 Patterns of intrinsic bioremediation at two U.S. Air Force bases. In Hinchee, R.E.", J.T. Wilson, and
 B.C. Downey (eds.) Intrinsic Bioremediation. Battelle Press, Columbus, OH. Pp31-52.
 Wiedemeier, T.H., J.T. Wilson, D.H. Kampbell, R.N. Miller, and J.E. Hansen. 1996. Technical
 Protocol for Implementing Intrinsic Remediation with Long Term Monitoring for Natural
Attenuation of Fuel Contamination Dissolved in Groundwater. Draft (Revision 0).  Air Force
 Center for Environmental  Excellence, Brooks AFB, TX.
                                         B.25
-------
              List of References
if I    ,	'',
	til   i"i	
Vftedemeier, T.H., D.C. Downey, J.T. Wilson, D.H7Kampbell, R.N. Miller, and I.E. Hansen.
1994. Technical Protocol for Implementing Intrinsic Remediation with Long-Term Monitoring
Option for Natural Attenuation of Dissolved-Phase Fuel Contamination in Groundwater.  Draft.
Air Force Center for Environmental Excellence, Brooks AFB, TX.

Wilding, LP. 1985. Spatial variability: its documentation, accommodation and implication to soil surveys.
Soil Spatial Variability. D.R. Nielsen and J. Bouma (eds). Wageningen, Netherlands.
Williams, R.T., P.S. Ziegenfuss, and W.E. Sisk. 1992. Composting of explosives and propellant
contaminated soils under thermophilic and mesophilic conditions.  J. Indust. Micro. 9:137.

Wilson, B.H, BE Bledsoe, D.H. Kampbell, and J.T. Wilson.  1986. Biological fate of hydrocar-
bons at an aviation gasoline spill site. Proc. Petroleum Hydrocarbons and Organic Chemicals in
Ground Water: Prevention, Detection, and Restoration.  Conference and Exposition, National
Water Well Association and The American Petroleum Institute, Houston, TX. November.
   "   /  ,•:•"•   ,  i",11.. V"','••• :.•",•.   "  •.-."• -" !,:''.>  .'SIM,,! '"   !h,, |.M.i'	'i	I, .;i	.("'in,:1  i1.,.1*1  ,«..! ;,,i. I"illi;;<'•':„ 1,
Wilson, D.J., R.D. Norris, and A.N. Clarke.  1996b. Groundwater cleanup by in-situ air sparging.
Volume X. Air channeling model for biosparging of NAPL. Separ.  Sci. Technol.  31:1357-1376.

Wilson, D J., R.D. Norris, and A.N. Clarke.  1996a. Groundwater cleanup by in situ sparging.
Volume K. Air channeling model for NAPL removal. Separ. Sci. Technol. 31:915-939.

Wilson, D.J., S. Kayano, R.D. Mutch, and A.M. Clarkei  1992.  Groundwater cleanup by in situ
sparging.  Volume I. Mathematical modeling. Separ. Sci. fechnot. 27:1023.
Wilson, J.T. and B.Hi Wilson.  1985. Biotransfbrmation of trichloroethylene in soil. Appt.
Environ. Micro.  49(0:242-243.
Wilson, J.T., G.D. Miller, W.C. Ghiorse, and F.R. Leach.  1986. Relationship between the ATP
content of subsurface material and the rate of biodegradation of alkylbenzenes and chlorobenzehe.
J. Contam. Hydrology.  1(1986):163-170.
    	                                              I
Wilson, J.L., L. Leach, J. Michalowski, S. Vandegrift, and R. Callaway.  1989. In situ bioreme-
diation of spills from underground storage tanks: new approaches for site characterization, project
design, and evaluation and performance. EPA/600/2-89/042. Robert S. Kerr Environmental
Research Laboratory, U.S. EPA, Ada, Oklahoma.
Wilson, J.T., J.Fl McNabb, J. Cpchran, T.H. Wang, M.B. Tomspri, and P.B.Bedient.  1985.
Influencei  of micrpbiai adaptation on the fate or organic" pollutants in groundwater.  Environ.
Toxicol. Chem. 4:721-726.
Wilson, J.T., F.M. Pfeffer, LW.' WeaveK D.H. kampbeli, f .H. Wiedemeier, I.E. Hansen, and R.N.
Miller.  1994.  Intrinsic bioremediation of JP-4 jet fuel. Proc. EPA Symposium on Intrinsic
Bioremediation of Ground Water. EPA/540/R-94/515.

Wilson, S.A.  1992.  The Preparation and Analysis of Soil Compost Material for Inorganic and
Explosive  Constituents: Final Report. U.S. Geological Survey Report for the U.S. Army Toxic
and Hazardous Materials Agency. October.
                                                         I
Wolf, D.C., T.H. Dap, H.D. Scott, and T.L. Lavy. 1989. Influence of sterilization methods on
selected soil microbiological, physical, and chemical properties! J. of Environmental Quality,
(18): 39-44. '	        "   '" '	""""   	  " '"
Wpilum, AG.  1982. Cultural methods for soil microorganisms. Methods of Soil Analysis, Part
2: Chemical and Microbiological Properties. 2nd edition. American Society of Agronomy, Inc.
Madison, WI:  Soil Science Society of America, Inc.
Woodward-Clyde Consultants.  1988. Feasibility Study for Site Remediation,  Vols. land 111,
Libby, MT. Denver, CO.   '           	'""	
Woodward-Clyde Consultants.  1989a. Quality Assurance Project Plan, RI/FS/RD/RA Activities,
Ground Water Site, Libby, Montana. Denver, CO.
                                                       B.26
-------
                                                                          Appendix B
Woodward-Clyde Consultants.  1989b. Health and Safety Plan, LTD and Soil Demonstration,
Ground Water Site, Libby, Montana.  Denver, CO.
Woodward-Clyde Consultants.  1989c. Land Treatment Demonstration Unit: One Acre LTD,
Ground Water Site, Libby, Montana.  Denver, CO.
Woodward-Clyde Consultants.  1989d No Migration Petition for Land Treatment Unit, Libby
Ground Water Site, Libby, Montana.  Denver, CO.

Woodward-Clyde Consultants. 1990. Pre-Final Design Report: Land Treatment Unit, Ground
Water Site, Libby, Montana. Denver, CO.

Woodward-Clyde Consultants.  1991.  Final Design Report Volume 1, Land Treatment Unit,
Ground Water Site, Libby, Montana.  Denver, CO.

Woodward-Clyde Consultants.  1992.  LTU Operations and Monitoring,  Ground Water Site,
Libby Montana. Denver, CO.

Wurdemann, H., M. Wittmaier, U. Rinkel, and H.H. Hanert. 1994. A simple method for determining
deficiency of oxygen during soil remediation. Hydrocarbons Bioremediation.  R.E. Hinchee, B.C.
Alleman, R.E. Hoeppel, and R.N. Miller (eds.). Boca Raton, FL: Lewis Publishers.

Yang, X., C.Y. Jeng, V. Kremesec, B. Fisher, and L. Curran.  1995. Natural Attenuation as a
Remedial Alternative. Technical Guidance. Amoco Corporation. July.

Yang, Y. and D. Alibeckoff. 1995. Biofiltration for control of carbon disulfide and hydrogen sulfide
vapors. Presented at the 7995 Conference on Biofiltration. Los Angeles, CA.  October 5-6.

Yanosky, T.M. and D.A. Vroblesky.  1992. Relation of nickel concentrations in tree rings to
groundwater contamination. Water Resour. Res. 28:2077-2083.

Yavorsky, J. 1993. Biofiltration for control of gas streams containing low  concentrations of
volatile organic compounds. Paper presented at the Air and Waste Management Association 86th
Meeting and Exhibition. Denver, CO. June 13-18.
Ziminski, R.W. and G. Ferrara.  1993. The control of odors and VOCs from flavor manufacturing
facilities by a state-of-the-art commercial biofilter. Paper presented at the Air and Waste
Management Association 86th Meeting and Exhibition. Denver, CO.  June  13-18.

Zwick, T.C., A. Leeson, R.E. Hinchee, R.E. Hoeppel, and L. Bowling. 1995. Soil moisture
effects during bioventing in fuel contaminated arid soils.  In Situ Aeration:  Air Sparging,
Bioventing, and Related Remediation Processes.  R.E. Hinchee, R.M. Miller, and P.C. Johnson
(eds.).  Columbus, OH:  Battelle Press. 3(2):333-340.
                                         B.27
-------
                                                                                                                         it ,:
      	i   li
                                                                                                 :	(   ••
                                                                                                 MI,   •
I'll r  , ' /  „ "I
                                    -1	  t , i.1 1    '  	


                                    	V  :'l,ii 'fir i''I1     ' , "  '",i,  *
I'     !   ,ii!  ""!  t)
-------
                                                         Appendix C
               LIST OF ACRONYMS
AAEE       American Academy of Environmental Engineers
ATP        Adenosine Triphosphate                             ,  .
B(a)P       Benzo(a)pyrene
BTEX       Benzene, Toluene, Ethylbenzene, and Xylenes
BOD       Biological Oxygen Demand
CEC        Cation Exchange Capacity
CERCLA    Comprehensive Environmental Response, Compensation, and
            Liability Act
COD        Chemical Oxygen Demand
CoM        Center of Mass
DCA        Dichloroethane
DCE        Dichloroethene(Dichloroethylene)
DNAPL     Dense Nonaqueous-Phase Liquid
DO         Dissolved Oxygen
DRE        Degradation/Removal Efficiency
DRO        Diesel Range Organics
FAME       Fatty Acid Methyl Ester
FID         Flame lonization Detector
FLTG       French Limited Task Group
FS          Feasibility Study
GC         Gas Chromatograph(y)         :
GRO        Gasoline Range Organics
HDPE       High-Density Polyethylene
HMX       High Melting Explosive
HPLC       High-Performance Liquid Chromatography
HSP        Health and Safety Plan
HSWA       Hazardous and Solid Waste Amendments
IAS         In Situ Air Sparging
KTPP       Potassium Tripolyphosphate
LEL        Lower Explosion Limit
LNAPL      Light Nonaqueous-Phase Liquid  ;
LTU        Land Treatment Unit
MMO       Methane Monooxygenase
MPN        Most Probable Number
MS         Mass Spectrometry            <
MSB        Mean Square Error
NAPL       Nonaqueous-Phase Liquid
ORC        Oxygen Release Compound
PAH        Polycyclic Aromatic Hydrocarbon
                                C.I
-------
List of Acronyms
PCB        Polychlorinated Biphenyl
PCE        Perchloroethene (Perchloroethylene or Tetrachloroethylene)
PCP        Pentachlorophenol
PCR        Polymerase Chain Reaction
PH)        Photoionization Detector
PLC        Process Logic Control(ler)
POC        Point of Compliance
PRP        Potentially Responsible Party
PRTs       Plume Resident Tracers
PVC        Poly Vinyl Chloride
RAP        Remedial Action Plan
RCRA      Resource Conservation and Recovery Act
RDX       Royal Demolition Explosive
RI          Remedial Investigation
RITZ       Regulatory and Interactive Treatment Zone (Model)
ROD       Record 9^ Decision
ROI        Radius of Influence
SARA      Superfund Amendments and Reauthorization Act
SIB         $oil Injection Bed
STF        Soil Transport and Fate (Database)
SVE        Soil Vapor Extraction
TCE        Trichloroethene(Trichloroethylene)
TEA        Terminal Electron Acceptor
TKN       Total Kjeldahl Nitrogen
TNT        Trinitrotoluene
TOG        Total Organic Carbon
TCP AH     Total Carcinogenic PAH
TPH        Total Petroleum Hydrocarbon
US EPA     U.S. Environmental Protection Agency
VC         Vinyl Chloride
VOC       Volatile Organic Compound
                                   C.2
-------
    THE WASTECH® MONOGRAPH SERIES (PHASE  II) ON
       INNOVATIVE SITE REMEDIATION TECHNOLOGY:
                       DESIGN AND APPLICATION
 This seven-book series focusing on the design and application of innovative site remediation
 technologies follows an earlier series (Phase I, 1994-1995) which cover the process descriptions,
 evaluations, and limitations of these same technologies.  The success of that series of publications
 suggested that this Phase II series be developed for practitioners in need of design information
 and applications, including case studies.

 WASTECH® is a multiorganization effort which joins! in partnership the Air and Waste Manage-
 ment Association, the American Institute of Chemical Engineers, the American Society of Civil
 Engineers, the American Society of Mechanical Engineers, the Hazardous Waste Action
 Coalition, the Society for Industrial Microbiology, the Soil Science Society of America, and
 the Water Environment Federation, together with the American Academy of Environmental
 Engineers, the U.S. Environmental Protection Agency, the U.S. Department of Defense, and the
 U.S. Department of Energy.

 A Steering Committee composed of highly respected members of each participating organization
 with expertise in remediation technology formulated and guided both phases, with project
 management and support provided by the Academy.  Each monograph was prepared by a Task
 Group of recognized experts. The manuscripts were subjected to extensive peer reviews prior to
 publication. This Design and Application Series includes:
Vol 1 - Bioremediation
Principal authors: R. Ryan Dupont, Ph.D., Chair,
Utah State University; Clifford J. Bruell, Ph.D.,
University of Massachusetts; Douglas C. Downey,
P.E., Parsons Engineering Science; Scott G. Huling,
Ph.D., P.E., USEPA; Michael C. Marley, Ph.D., Xpert
Design and Diagnostics, Inc.; Robert D. Norris, Ph.D.,
ECKENFELDER, INC.; Bruce Pivetz, PhJX,
Manfech Environmental Research Services Corp.

Vol 2 - Chemical Treatment
Principal authors: Leo Weitzman, Ph.D., LVW
Associates, Chair; Irvin A. Jefcoat, Ph.D., University
of Alabama; Byung R. Kim, Ph.D., Ford Research
Laboratory.

Vol 3 - Liquid Extraction Technologies:
Soil Washing/Soil Flushing/Solvent Chemical
Principal authors: Michael J. Mann, P.E., DEE,
ARCADIS Geraghty & Miller, Inc., Chair, Richard J.
Ayen, Ph.D., Waste Management Inc.; Lome G. Everett,
Ph.D., Geraghty & Miller, Inc.; Dirk Gombert U, P.E.,
LIFCO; Mark Meckes, USEPA; Chester R. McKee,
Ph.D., In-Situ, Inc.; Richard P. Traver, P.E., Bergmann
USA; Phillip D. Walling, Jr., P.E., E. I. DuPont Co. Inc.;
Shao-Chih Way, Ph.D., In-Situ, Inc.

Vol 4 - Stabilization/Solidification
Principal authors: Paul D. Kalb, Brookhaven National
Laboratory, Chair, Jesse R. Conner, Conner Technolo-
gies, Inc.; John L. Mayberry, P.E., SAIC; Bhavesh R.
Patel, U.S. Department of Energy; Joseph M. Perez, Jr.,
Battelle Pacific Northwest; Russell L. Treat, MACTEC
Voll 5 - Thermal Desorptlon
Principal authors: William L. Troxler, P.E., Focus
Environmental Inc., Chair, Edward S. Alperin, IT
Corporation; Paul R. de Percin, USEPA; Joseph H.
Button,.P.E., Canonie Environmental Services, Inc.;
JoAnn S. LIghty, Ph.D., University of Utah; Carl R.
Palmer, P.E., Rust Remedial Services, Inc.

Vol 6 - Thermal Destruction
Principal authors: Francis W. Holm, Ph.D., SAIC, Chair,
Carl R. Cooley, Department of Energy; James J.
Cuclahy, P.E., Focus Environmental Inc.; Clyde R.
Demipsey, P.E., USEPA; John P. Longwell, Sc.D.,
Massachusetts Institute of Technology; Richard S.
Magee, Sc.D., P.E., DEE, New Jersey Institute of
Technology; Walter G. May, Sc.D., University of Illinois.

Vol 7 - Vapor Extraction and Air Sparging
Principal authors: Timothy B. Holbrook, P.E., Camp
Dreisser & McKee, Inc., Chair, David H. Bass, Sc.D.,
Groundwater Technology, Inc.; Paul M. Boersma,
CH2M Hill; Dominic C. DiGiulio, University of
Arizona; John J. Eisenbeis, Ph.D., Camp Dresser &
McKee, Inc.; Neil J. Hutzler, Ph.D,, Michigan
Technological University; Eric P.  Roberts, P.E., ICF
Kaiser Engineers, Inc.
 The monographs for both the Phase I and Phase II
 series may be purchased from the American Academy
 of Environmental Engineers'", 130 Holiday Court, Suite
 100, Annapolis, MD, 21401; Phone: 410-266-3390,
 Fax: 410-266-7653, E-mail: aaee@ea.net
                  •it U.S. GOVERNMENT PRINTING OFFICE: 1998 - 821 - 370 / 93295
-------
                                                                                                                                                                    r:      jfj
	.1
                                                                                                                                                                 111!  if      Jli"!
                                                                           r  .   •  • .   fii	iu
                                                                                                                                                                 .11	'       -	:,
-------
Prepared by the American
Academy of Environmental
Engineers under a
cooperative agreement with
the U.S. Environmental
Protection Agency

EPA 542-B-97-004
-------
                                                                                                                                                                                                                                                                                                                                                                                                                        	14
 I  Bit1  '>li   i       i
                                                                                          Mi:'      MS!
                                                                                                                                                                                                                                                                                                                                                                                                                         '•  1
nil iiiini ill I    i  ..... 'i'liiiiiiiMii
                                                         i jn;1 ,  i,,;/     ,    „  i.  ,  iu ....... M|IIIIIIJIIIIIIIIIIIK  jJii
-------