ABSTRACT
Results are presented from a laboratory biotreatability study
which evaluated effects of nitrate salt amendment on organic solvent
biodegradation in groundwater. High concentrations of dissolved-phase
volatile organic compounds (VOCs) were detected at the subject
site, including several feet of light non-aqueous phase liquid
(LNAPL). Formation materials consist of fine-grained soils of
low permeability. LNAPL recovery, groundwater extraction/treatment,
and in-situ bioremediation were considered for site remediation.
An in-situ (anaerobic) bioremediation strategy was proposed based
on low formation permeability and high VOC levels. Enhanced microbial
denitrification was selected due to the high aqueous solubility
and low cost of nitrate salts. A biotreatability study consisting
of groundwater screening for biodegradation indicator parameters,
an enumeration study for indigenous denitrifying bacteria, and
a bench-scale respirometry/mineralization study evaluated the
potential for enhanced microbial denitrification. Study results
indicated: (1) on-site groundwater was acidic, did not contain
detectable nitrate, and contained high VOC levels that were potentially
cytotoxic; (2) population densities of indigenous denitrifying
bacteria were low; and (3) attempted biostimulation did not result
in significantly reduced nitrate concentration or Chemical Oxygen
Demand (COD) during a 120-day bench-scale study. Results suggest
that denitrifying bacteria were inhibited and enhanced microbial
denitrification was not viable. Consequently, remediation was
limited to LNAPL recovery and groundwater extraction/treatment.
INTRODUCTION
Releases of VOCs threaten groundwater resources because contaminant
aqueous solubilities often significantly exceed toxicity thresholds.
Many remediation technologies are costly and may not mitigate
groundwater contamination expeditiously. In-situ bioremediation
is an alternative remediation technology involving injection of
a terminal electron acceptor (oxidant) and inorganic nutrients
to enhance intrinsic (natural) biodegradation of target compounds
by indigenous microbes. During VOC biodegradation, microbes use
oxidants to collect electrons released during metabolism, and
inorganic nutrients in combination with organic carbon from VOCs
to sustain cellular biomass. Although certain microbes can respire
different oxidants, depending on subsurface conditions, O2
is the most energetic and therefore is preferred for most in-situ
bioremediation projects.
A serious limitation of in-situ (aerobic) bioremediation is its
reduced effectiveness for treating highly-contaminated groundwater
systems of low hydraulic conductivity. Low permeability restricts
dissolved oxygen (DO) transport and high concentrations potentially
exert a large biochemical oxygen demand on groundwater. In addition,
low O2 aqueous solubility (~10 mg/l) limits O2
loading potential. Even approaches which can deliver large O2
loads are severely limited under these conditions. For example,
H2O2 amendment can supply substantial O2
loads, however, material costs may be prohibitive based on the
stoichiometric relationship between petroleum hydrocarbons (PHCs)
and DO (e.g., a H2O2 concentration of about
20 mg/l can stoichiometrically degrade about 5 mg/l PHCs). In
addition, reduction in permeability can occur if H2O2
decomposes and exsolves O2 bubbles from groundwater,
or reacts with formation materials causing minerals to precipitate
(Baker et al., 1994; Riser-Roberts, 1992). Alternatively, in-situ
air sparging (IAS) can provide large quantities of O2
at relatively low cost; however, IAS is less effective for low
permeability formations due to channelling of air flow through
the formation matrix (Johnson et al., 1993; Nyer et al., 1993;
Angell, 1992).
Research to overcome this limitation has been performed using
nitrate (NO3-) salt amendment to enhance
anaerobic biodegradation. NO3- respiration
(microbial denitrification) is a metabolic analog to O2
respiration in which NO3- (not O2)
serves as the oxidant. Microbial denitrification occurs under
anoxic conditions in which both NO3- and
organic carbon are present with indigenous bacteria capable of
respiring NO3- (Starr et al., 1993; Chapelle,
1993; Ward, 1985). Although denitrifying bacteria can also respire
nitrite (NO2-), nitric oxide (NO), and nitrous
oxide (N2O), the NO3- anion is
preferred because constituent nitrogen is the most oxidized (i.e.,
NO3-, NO2-, NO, and
N2O have N oxidization states of +5, +3, +2, and +1,
respectively). Therefore, NO3- is the most
energetic nitrogen oxide for stimulating microbial denitrification.
Studies demonstrate nitrate salt amendment can enhance denitrification.
For example, Major et al. (1988) showed that nitrate salt amendment
enhanced biodegradation of benzene, toluene, and total xylenes
in laboratory microcosms. Hilton et al. (1992) showed that enhanced
denitrification stimulated VOC biodegradation in a groundwater
system of moderate to low hydraulic conductivity.
NO3- has several advantages over O2
for enhancing biodegradation in-situ. Nitrate salts have aqueous
solubility limits orders-of-magnitude greater than O2,
and can therefore supply significantly larger oxidant loads.
Nitrate salts also are relatively low cost and can be delivered
by a low-technology injection system. In addition, NO3-
respiration is expected to sustain smaller microbial population
densities than aerobic respiration because microbial denitrification
is less metabolically efficient. Hence, biofouling problems may
be less severe. Additionally, nitrate salt also supplies the
inorganic nutrient nitrogen which may stimulate microbial activity.
NO3- respiration also may elevate groundwater
pH because denitrification consumes hydrogen ions. Because VOC-degrading
bacteria are sometimes limited by acidic conditions, elevated
pH may enhance metabolism. Limitations to nitrate salt amendment
include:
- NO3- concentrations above 10 mg/l
exceed regulatory thresholds established by the United States
of America (USA) Environmental Protection Agency (EPA). Hence,
nitrate salt amendment poses regulatory concerns which may restrict
full-scale implementation. However, careful engineering can minimize
risks to sensitive receptors because denitrification consumes
NO3-;
- Biodegradation rates associated with NO3-
respiration are slower than those associated with O2
respiration. Nevertheless, slower rates may be compensated by
greater oxidant loading (Sheehan et al., 1988; Hutchins, et al.,
1991; Lemon et al., 1989);
- Because nitrogen gas (N2) is the ultimate product
of denitrification, NO3- respiration can
exsolve gas bubbles into the formation thereby reducing permeability.
However, this condition may be mitigated by maintaining NO3-
concentrations in excess of nitrogen oxide intermediates, at least
in contaminant source areas, so that the reaction does not proceed
to N2; and
- Denitrification is inhibited in oxic groundwater regimes.
Nevertheless, contaminant plumes are commonly anoxic (Levin et
al., 1993).
This paper presents results from a biotreatability study performed
to evaluate enhanced microbial denitrification as a remedial approach
for an organic solvent-contaminated groundwater system.
BACKGROUND INFORMATION
SETTING
The subject site is occupied by a paint/industrial coatings manufacturer,
and is located in the northeastern USA. Organic solvents are
used at the site for manufacturing processes. A hydrogeologic
investigation was initiated in response to a report of solvent
odors emanating from drainage ditches near the facility. Results
of the investigation indicated that the source of the odors was
an underdrain originating from on-site underground storage tanks
(USTs) which contained organic solvents consisting primarily of
VOCs. The VOC source was identified as releases from these USTs.
The site is situated on a terrace at the base of an approximate
140-foot hill. Subsurface geology consists of about 30-feet of
lacustrine clays/silty clays interbedded with fine sands and silts.
Overburden soils are underlain by metasedimentary rocks including
slate, phyllite, and schist interbedded with quartzite.
The groundwater system is characterized by fine-grained soils
with an estimated hydraulic conductivity of about 1E-5 to 1E-6
cm/s. The hydraulic gradient and effective porosity are estimated
at about 0.10 and 0.33, respectively. Estimates of groundwater
seepage velocities are on the order of 0.001 to 0.01 ft/d. Water
table elevations fluctuate between 5 to 10 feet seasonally. An
on-site piezometer nest indicated groundwater recharge conditions
at the site.
A LNAPL layer was encountered in monitoring wells installed in
the vicinity of the USTs, which varied from an iridescent sheen
to about 6 feet in thickness. Dissolved-phase VOCs were detected
at the site, including aromatic PHCs (£90
mg/l), ethers (£46 mg/l), alcohols
(£1,300 mg/l), ketones (£750
mg/l), and esters (£5,000 mg/l).
REMEDIAL APPROACH
Leaking USTs were closed and a remedial action plan was developed,
consisting of LNAPL recovery and groundwater extraction/treatment.
An LNAPL recovery system was first installed utilizing certain
on-site monitoring wells retrofit with pneumatic pumps. A laterally
extended well (LEW) system consisting of horizontal pumping wells
installed within trenches was then constructed to enhance contaminant
mass recovery. LEW effluent was treated via granular activated
carbon. Monitoring indicated the LEW system enhanced LNAPL recovery
and removed dissolved-phase VOCs, however, system performance
was limited by the low formation permeability which limited the
size of the capture zone. Based on estimated seepage velocities
and a 45-foot flow path, LNAPL recovery could take on the order
of 10 to 100 years, and over a much longer time period for recovery
of dissolved-phase VOCs. In-situ bioremediation was considered
as a remedial technology which may substantially shorten the time
period for site remediation.
Enhanced microbial denitrification was selected based on low permeability
of formation materials and high VOC concentrations. A biotreatability
study was performed, the objective of which was to evaluate whether
enhanced denitrification was an appropriate technology given subsurface
conditions at the site. The following tasks were performed to
satisfy the study objective: (1) groundwater screening for biodegradation
indicator parameters, (2) microbial enumerations of potential
contaminant-degrading bacteria capable of respiring NO3-,
and (3) a bench-scale respirometry/mineralization study.
METHODS
GROUNDWATER SCREENING
Groundwater screening was performed to assess whether subsurface
conditions were suitable for microbial denitrification. Samples
were collected from a monitoring well (MW-1) within the upgradient
portion of the contaminant plume and from background monitoring
well MW-2 for field screening and laboratory analysis for certain
inorganic constituents and VOCs. Only two monitoring wells were
sampled because other on-site wells were dry at the time of the
sampling round. Groundwater screening results are summarized
in Table 1.
MICROBIAL ENUMERATION
Eighteen soil samples from test borings BT-1 through BT-4 (various
depths) and two groundwater samples from monitoring wells MW-1
and MW-2 were collected aseptically from contaminant plume and
background locations to assess the magnitude and distribution
of indigenous denitrifying bacteria. Sample containers were sealed
with paraffin to minimize contamination by atmospheric O2,
and were submitted to University of New Hampshire's Jackson Estuarine
Laboratory of Durham, New Hampshire for enumeration by Most-Probable-Number
(MPN) technique. Results are summarized in Table 2.
RESPIROMETRY/MINERALIZATION STUDY
A bench-scale respirometry/mineralization study was performed
using slurries of contaminated soil/groundwater collected aseptically
from the site to assess if nitrate salt amendment could enhance
VOC metabolism. The respirometry and mineralization study measured
NO3- and substrate (VOC) consumption, respectively.
Sacrificial flasks containing replicate sample slurries were
amended with varying nitrate salt/inorganic nutrient concentrations,
and pH was increased in several flasks. Flasks were incubated
under controlled conditions and were periodically analyzed for
NO3- and COD. A decrease over time in NO3-
concentration (all flasks), and a concomitant decrease over time
in COD (flasks amended with sufficient NO3-
to satisfy a significant portion of the NO3-
demand exerted by the COD) would indicate positive results (i.e.,
enhanced denitrification).
Flask Preparation/Incubation 70 milliliters (ml) of groundwater
and 14 mg of soil were added to each of fifteen sterile flasks.
The following three groups of flasks were prepared to evaluate
various biostimulation strategies:
- Group I (NO3-/nutrient amendment,
pH 5): nine flasks amended with 415 mg/l NaNO3 (303
mg/l NO3-), 480 mg/l K2HPO4
(215 mg/l K, 85 mg/l P);
- Group II (NO3-/nutrient amendment,
pH 7): five flasks amended with 415 mg/l NaNO3 (303
mg/l NO3-), 480 mg/l K2HPO4
(215 mg/l K, 85 mg/l P), and pH adjusted to approximately 7 with
NaOH; and
- Group III (high level NO3-/nutrient
amendment, pH 7): one flask amended with 2,745 mg/l NaNO3
(2,000 mg/l NO3-), 3,190 mg/l K2HPO4
(1,430 mg/l K, 570 mg/l P), and pH adjusted to approximately 7
with NaOH.
Groups I and II NO3-/nutrient concentrations
were selected based on those reported for municipal wastewater
denitrification operations; Group III parameters were selected
to evaluate whether nitrate salt amendment could measurably reduce
COD. The pH of Groups II and III flasks was increased to evaluate
whether pH adjustment could enhance denitrification. Flasks were
incubated in a 60-quart cooler fitted with inlet/outlet ports
through which N2 was flushed at a rate which considered
atmospheric fluctuations in barometric pressure. Incubator temperature
was maintained at about 19 to 21oC.
Biotreatability Study Analytical Methods Analyses for
COD and NO3- were performed at GZA GeoEnvironmental,
Inc.'s Environmental Chemistry Laboratory of Newton Upper Falls,
Massachusetts, USA. A NO3- concentration
versus time plot and a COD versus time plot for the three flask
Groups are provided as Figures 1 and 2, respectively.
RESULTS AND DISCUSSION
GROUNDWATER SCREENING
Field Parameters The pH of contaminated groundwater was
depressed relative to background (Table 1). This condition may
be related to hydrolysis of CO2 (generated during intrinsic
biodegradation) to carbonic acid, and/or hydrolysis of esters
to organic acids and alcohols. Moreover, groundwater pH was relatively
low. Because denitrifying bacteria are pH- sensitive and pH values
less than about 6 can be inhibitory, the acidic groundwater condition
may suppress microbial activity.
The ORP was low, indicating chemically-reduced groundwater conditions.
Moreover, ORP values were in a range suitable for microbial denitrification
-- about -220 to +750 millivolts (Bitton et al., 1984).
The DO concentration of contaminated groundwater was also depressed
relative to background, suggesting that intrinsic aerobic biodegradation
is occurring. The moderate DO concentration at monitoring well
MW-1 indicates oxic conditions exist in the upgradient portion
of the contaminant plume. Because denitrification cannot proceed
in the presence of O2, the DO level likely inhibits
NO3- respiration in this portion of the
plume. However, because the upgradient DO concentration was depressed,
then the downgradient concentration was expected to decrease with
distance from the source along the flow path. Hence, downgradient
locations may be more suitable for microbial denitrification than
upgradient locations.
Inorganic Constituents Inorganic data suggest that these
constituents are generally not limiting with two notable exceptions
(Table 1). First, neither NO3- or NO2-
were detected above analytical detection limits, suggesting that
bacteria capable of respiring NO3- are oxidant
and possibly nitrogen limited. Second, phosphorus was not detected
and also may be limiting. Though analytical testing can establish
oxidant limitation, inorganic nutrient limitation must be assessed
empirically.
VOCs Contaminant signature is consistent with historical
groundwater quality data for the site (Table 1). VOCs were not
detected above analytical detection limits in background groundwater,
but the contaminant plume contained relatively high levels of
certain aromatic PHCs, ethers, alcohols, ketones, and esters.
Published information suggests these VOCs to be biodegradable
under oxic conditions, with aromatic PHCs the most recalcitrant.
Because NO3- respiration is analogous to
O2 respiration, these VOCs may be biodegradable under
denitrifying conditions, however, the high concentrations may
exceed cytotoxicity thresholds. Moreover, limited information
is available on the biodegradability potential of many of these
VOCs via denitrification. Because biodegradation of contaminant
mixtures is site specific, VOC biodegradability potential for
the subject site must also be assessed empirically.
MICROBIAL ENUMERATION
The indigenous denitrifying bacteria population density was low,
suggesting inhibitory subsurface conditions exist at the site
(Table 2). For example, Hutchins et al. (1991) reported population
densities of indigenous denitrifying bacteria at another site
which were about two to five orders-of-magnitude greater than
those at the subject site. Moreover, there was no significant
difference in mean population density between contaminated and
background locations, suggesting that intrinsic microbial denitrification
was not an important VOC attenuation process. Denitrifying bacteria
also were not detected in groundwater samples obtained from monitoring
wells MW-1 or MW-2. These data are consistent with the soil data
indicating that indigenous bacteria are inhibited and are not
effecting microbial denitrification at the site.
RESPIROMETRY/MINERALIZATION STUDY
NO3- concentrations appeared to increase
and not decrease over time (Figure 1). The apparent increase
was attributed to matrix interference with the ISE as previously
discussed (Footnote 9). A calibration check of the ISE using
spiked samples showed that the Days 90 and 107 NO3-
data were about three times actual concentrations. Using a correction
factor of 1/3, actual Day 120 NO3- concentrations
were about 350 mg/l, 203 mg/l, and 2,070 mg/l for Groups I, II,
and III flasks, respectively. Because initial NO3-
concentration for these flasks were about 303 mg/l, 303 mg/l,
and 2,000 mg/l, respectively, the Day 120 data suggest that NO3-
respiration was inhibited, with the possible exception of within
Group II flasks. However, the low population density of indigenous
denitrifying bacteria in combination with the extended duration
of the study suggested that microbial denitrification was also
inhibited in Group II flasks. For example, Major et al. (1988)
and Hutchins et al. (1991) demonstrated significant aromatic PHC
degradation via microbial denitrification in about 60 to 80 days
of nitrate amendment. Because aromatic PHCs are the most recalcitrant
of the VOCs detected at the subject site and the flasks were incubated
for up to 120 days, the data suggest that NO3-
respiration was inhibited for even Group II flasks.
COD also did not significantly decrease over time (Figure 2).
Data indicate that anoxic conditions were maintained in the Groups
I and II flasks because complete stoichiometric biodegradation
via NO3- respiration would remove only about
1 to 2 percent of COD (i.e., about 190 mg/l). A large COD reduction
in these flasks would indicate aerobic respiration or volatilization
but not microbial denitrification. Due to the high nitrate concentration
amended to Group III flasks (high level NO3-/nutrient
amendment, pH 7) which would stoichiometrically remove about 10
percent of COD (~1,300 mg/l), the data also suggest that mineralization
did not occur in these flasks. The agreement between NO3-
and COD data indicates that microbial denitrification was inhibited
during the bench-scale respirometry/mineralization study.
CONCLUSIONS
The significant conclusions of the biotreatability study are that:
(1) the indigenous denitrifying bacteria population is inhibited,
perhaps by low pH, the oxidant/inorganic nutrient-limited condition,
and/or high, potentially cytotoxic VOC concentrations; and (2)
biostimulation by nitrate salt/inorganic nutrient amendment with
or without pH adjustment did not enhance VOC biodegradation via
microbial denitrification during the 120-day respiration/mineralization
study. A bioaugmentation program was considered as a means to
overcome inhibition of indigenous bacteria, however, the program
was not implemented due to the unproven nature of the technology
for in-situ treatment. Based on the results of this study, site
remediation was limited to LNAPL recovery and groundwater extraction/treatment.
REFERENCES CITED
Angell, K.G., 1992, In-situ remedial efforts: air sparging, The
National Environmental Journal, v. 2, Issue 1, p. 20-23
Baker, K.H. and Herson, D.S., 1994; Bioremediation: McGraw-Hill,
Inc., 374 p.
Bitton, G. and Gerba, C.P., 1984, Groundwater Pollution Microbiology:
John Wiley & Sons, Inc., 377 p.
Chapelle, F., 1993, Ground-Water Microbiology & Geochemistry:
John Wiley & Sons, Inc., 424 p.
Hilton, J., Marley, B., Ryther, T., and Forbes, J., 1992, Pilot
test of nitrate-enhanced hydrocarbon bioremediation in a moderate-
to low-permeable aquifer, Petroleum Hydrocarbons and Organic Chemicals
in Ground Water: National Ground Water Association and The American
Petroleum Institute, No. 14, p. 527-540
Hutchins, S.R., Downs, W.C., Wilson, J.T., Smith, G.B., Kovacs,
D.A., Fine, D.D., Douglass, R.H., and Hendrix, D.J., 1991, Effect
of nitrate addition on biorestoration of fuel-contaminated aquifer:
Field Demonstration, Ground Water, v. 29, No. 4, p. 571-580
Johnson, R.L., Johnson, P.C., McWhorter, D.B., Hinchee, R.E.,
and Goodman, I., 1993, An overview of in-situ air sparging, Ground
Water Monitoring Review, v. 13, No. 4, p. 127-135
Lemon, L.A., Barbaro, J.R., and Barker, J.F., 1989, Biotransformation
of BTEX under anaerobic denitrifying conditions: evaluation of
field observations, in proceedings, Focus Conference on Eastern
Regional Ground Water Issues, National Water Well Association,
p. 213-227
Levin, M.A. and Gealt, M.A., 1993, Biotreatment of Industrial
and Hazardous Waste: McGraw- Hill, Inc., 331 p.
Major, D.W., Mayfield, C.I., and Barker, J.F., 1988, Biotransformation
of benzene by denitrification in aquifer sand, Ground Water, v.
26, No. 1, p. 8-14
Nyer, E.K. and Suthersan, S.S., 1993, Air sparging: savior of
ground water remediations or just blowing bubbles in the bath
tub?, Ground Water Monitoring Review, v. 13, No. 4, p. 87-91
Riser-Roberts, E., 1992, Bioremediation of Petroleum Contaminated
Sites: CRC Press, Inc., 197 p.
Sheehan, P.J., Schneiter, R.W., Mohr, T.K.G., and Gersberg, R.M.,
1988, Bioreclamation of gasoline contaminated groundwater without
oxygen addition, in proceedings, Second National Outdoor Action
Conference on Aquifer Restoration, Ground Water Monitoring and
Geophysical Methods: National Water Well Association, v. I, p.
183-199
Starr, R.C. and Gillham, R.W., 1993, Denitrification and organic
carbon availability in two aquifers, Ground Water, v. 31, No.
6, p. 934-1056
Ward, T.E., 1985, Characterizing the aerobic and anaerobic microbial
activities in surface and subsurface soils, Environmental Toxicology
and Chemistry, v. 4, p. 727-737
BIOGRAPHICAL SKETCH
I. Richard Schaffner, Jr. conducted undergraduate
studies in geology and graduate studies in contaminant hydrogeology
at Brigham Young University. Mr. Schaffner is a member of the
National Ground Water Association, the New England Water Environment
Association, and the Northeast Branch of the American Society
for Microbiology. His interests include the fate and transport
of organic solvents in groundwater systems, and the microbial
metabolism of aromatic and aliphatic petroleum hydrocarbons in
the vadose and phreatic zones. Mr. Schaffner is a hydrogeologist
with GZA GeoEnvironmental, Inc., 380 Harvey Road, Manchester,
New Hampshire 03103. Phone No.: (603) 623-3600; E-Mail: rschaffner@gzea.com.
