Grantee Research Project Results

1999 Progress Report: Understanding Seasonal Variation of Bioavailability of Residual NAPL in the Vadose Zone

EPA Grant Number: R827133
Title: Understanding Seasonal Variation of Bioavailability of Residual NAPL in the Vadose Zone
Investigators: Holden, Patricia , Keller, Arturo A.
Institution: University of California - Santa Barbara
EPA Project Officer: Aja, Hayley
Project Period: October 1, 1998 through September 30, 2001
Project Period Covered by this Report: October 1, 1998 through September 30, 1999
Project Amount: $425,000
RFA: EPA/DOE/NSF/ONR Joint Program on Bioremediation (1998) RFA Text |  Recipients Lists
Research Category: Hazardous Waste/Remediation , Land and Waste Management

Objective:

Natural biological attenuation in the vadose zone may occur for a wide range of pollutants. However, the physical, chemical, and biological factors that control natural biological attenuation are not well understood for most pollutants. Reliance on a weakly understood remediation strategy such as natural attenuation leaves at risk underlying groundwater, hydrologically connected surface water, and thus human and ecosystem health. Pollutant bioavailability, as the composite of mass transfer and biodegradation processes, determines natural biological attenuation. Our hypothesis is that factors controlling pollutant bioavailability over time must be understood if natural attenuation, the composite of natural biological and physicochemical processes, is to become predictable and reliable. Further, we propose that an accurate understanding of long-term bioavailability can only be gained in the context of the normal climatic fluctuations that occur seasonally.

Specifically, our objectives are to determine: (1) how cycles in water potential affect residual non-aqueous phase liquid (NAPL) spreading under abiotic conditions, (2) how the presence of bacterial exopolymeric substances (EPS) affect NAPL
mass transfer and biodegradation as a function of water potential, (3) the patterns of microbial activity and growth that occur
with and without residual NAPL as a carbon source and as a result of wetting and drying cycles, (4) the mathematical
representation of the processes that describe residual NAPL bioavailability in unsaturated porous media, and (5) the functional
relationship between the index of bioavailability and environmental conditions such as soil moisture and temperature.

Progress Summary:

To integrate between the mechanistic information that can be derived from controlled
laboratory experiments and the empirical information obtained from an analysis of field site data, we are working at the pore,
core, and field scales.

The studies at the pore scale are being carried out using physical micromodels. The micromodels planned for this work were
designed with the following objectives in mind:

• Produce a realistic representation of actual pore geometry.
• Generate a model with the correct dimensions for pore scale processes.
• Use a media that simulates typical wetting properties of natural media.
• Allow easy reproduction of the micromodels, for comparison studies.

To achieve these objectives, a thin slice of Berea sandstone was imaged through an optical microscope and then digitized. To
improve the connectivity in the micromodel, the digitized image was modified slightly. The pattern then was repeated 100 x 100
times. The digital image of the pore space then was transferred to a chrome plated glass mask, at no magnification from the
original sandstone thin slice. Using technology similar to the manufacture of microchips, the image was photochemically etched
on a silicon wafer, at a constant etching thickness (micromodels have been constructed with etching depths ranging from 15 to
60 ?m, for different studies). Pore diameters range from 3 to 30 ?m. The porosity of the micromodel has been experimentally
determined to be 37 percent. We are able to make identical micromodel reproductions. The silicon surface is oxidized after
etching, leaving a water-wetting silica surface, approximately 10 nm deep.

An industrial microscope (Nikon Optiphot-M) is used to image the micromodel pore space. Because the silicon wafer is too
thick to transmit light; we use reflected light to view the fluids in the pore space. The setup also allows capture of fluorescence
from fluids or tagged colloids in the pore space. The images are captured using a CCD TV-camera (Optronics DEI-470) at a
high resolution (1/2" CCD Color Image Sensor, 470 Line Horizontal Resolution). The video image is acquired by a computer
using a video frame grabber (Flashpoint by Integral Technologies).

To control the moisture content of the gas phase in the micromodel, dry air is passed through a polyethylene glycol (PEG)
solution (M.W. 8000). Different concentrations of PEG are used to produce controlled soil water potentials. We plan to
control temperature in the micromodel using a heating plate, as well as pre-conditioning of the inflowing air by immersing the
flask containing the PEG solution in a constant temperature water bath.

The sequence of experiments at the pore scale is:

• Abiotic cycling of soil moisture in the micromodels to observe the growth and disappearance of water films.
• Abiotic flooding with NAPL into a micromodel at a controlled soil water potential.
• Abiotic drainage of mobile NAPL in the micromodel using either controlled-moisture air or water flooding, until a
  residual NAPL saturation is attained.
• Abiotic observation of NAPL volatilization or dissolution into the surrounding fluid phase(s) as a function of time and
  fluid phase flow.
• Biotic colonization of the micromodel pore space, feeding nutrients and substrate.
• Biotic cycling of soil moisture in the micromodels, observing the effect on microbial growth.
• NAPL flooding into a colonized micromodel.
• Drainage of NAPL from colonized micromodel until residual saturation is achieved.
• Observation of microbial growth during cycling of soil moisture and temperature, where the only carbon source is the
  residual NAPL.

Our approach to understanding how soil bacteria influence residual NAPL bioavailability focuses on the EPS matrix as part of
the microhabitat at the pore scale. Therefore, our main experiments at either the pore or core scale are preceded by generating
EPS variants of the same organism (relative under producer plus relative over producer) so that biodegradation and mass
transfer of a given NAPL can be studied for a model system with only one biotic variable, EPS. In generating the EPS variants,
we have selected to not use random mutagenesis and screening for EPS variants. Our rationale was that random mutagenesis
could alter other physiological attributes of our strain, thus compounding the differences between EPS variants by more than a
shift in EPS production. In lieu of random mutagenesis, we have focused on altering expression of genes for alginate production
in Pseudomonas aeruginosa. Our approaches generally enable us to directly understand the effects of EPS on biodegradation
in the pore and core scale experiments. Additionally, our approach includes streamlining the numbers of model organisms used
for the project. However, we recognize that it may not be possible to acquire or generate a single bacterial strain that is able to
use all of the NAPLs of interest to this project. Therefore, our approach is to try to minimize the numbers of model organisms
while also minimizing the project effort spent engineering organisms.

At the pore scale, we expect our studies to yield insight into mechanisms that control bioavailability of a range of microbially
degradable residual NAPLs. To bridge between the pore scale and the field scale, core scale studies will be performed, using
unsaturated sand as model substrata and mixtures of 14C-labeled NAPLs (toluene and hexadecane). The cores will be
nondestructively characterized using x-ray imaging before the biotic experiments. After characterization, the cores will be
inoculated with microbes and then degradation will be monitored via 14C-CO2 evolution. Our investigations include
predetermining the nutritional conditions in cores that result in biofilm formation. Final NAPL saturation and microbial growth
will be imaged using the x-ray analysis. The results from the pore and core scales will be analyzed quantitatively, to provide a
mechanistic interpretation of the functional relationship between the index of bioavailability and environmental conditions such as
soil moisture.

Field test data from well-monitored sites of intrinsic or managed bioremediation will be used to evaluate the effect that drying
and wetting cycles have on biodegradation rates. We will obtain data from existing sites, and model it numerically, based on our
studies at the pore and core scale, to predict the rate of biodegradation, as measured by CO2 evolution.

The micromodel setup is ready and complete. We have begun with the abiotic cycling of soil moisture regimes, to test the data
acquisition and analysis systems. Two sequences of water film buildup in a completely dry micromodel were assessed, after
flowing through air passed through different PEG solutions. In one of the two sequences, the PEG solution has a concentration
of 330 g/L, resulting in a water potential of ?1.5 MPa. Only the smallest pore throats are occupied entirely by the water phase.
The water film around the grains in the pore space is probably only a few nanometers thick, and thus not directly visible under
these conditions.

The other sequence indicates the buildup of a water film for air at ?0.25 MPa, after passing through a PEG solution with a
concentration of 100 g/L. There is no noticeable difference in the time it takes to build up the water films. However, magnified
images of the pore throats indicate that the water saturation is much higher, filling in some of the smaller pore bodies and leaving
disconnected air bubbles within the pore space.

We also have conducted experiments to visualize the rate of NAPL dissolution and visualization. Toluene was used as the
NAPL. After toluene had moved through some of the pore spaces, the liquid phase was switched back to water. The sequence
of images shows how the NAPL blob shrinks over time. The rate of dissolution can then be calculated. The rate of dissolution
of toluene depends on the interfacial area exposed to the water phase. Toluene is being volatilized by an invading air phase. The
air progressively enters the pore spaces, while the NAPL phase shrinks.

We selected toluene and hexadecane as spreading and nonspreading NAPLs, respectively, for the experimental pore and core
studies. This selection was partly driven by bacterial strain availability and our familiarity with strain and substrate utilization
patterns. The strains that we have been working with for this project are Pseudomonas aeruginosa PG201 and P. putida
mt-2. The versatility of substrate utilization for these two strains was determined and summarized in Table 1. Growth rates in
liquid culture were determined for this project and are outlined in Table 2. Our extensive prior work with these organisms to
date makes their inclusion in this project relatively efficient as opposed to isolating or working with new strains. However,
neither organism is capable of using both toluene and hexadecane as sole carbon sources. To resolve this issue, we have
attempted to mate the two organisms to facilitate transfer of the TOL plasmid into P. aeruginosa. We currently are in the
process of determining the success by expression in liquid culture. Alternatively, we can use both strains in independent studies.
We find this alternative strategy acceptable for accomplishing our goals for the project.

Table 1: Bacterial strains and hydrocarbon utilization in liquid culture
strain	hexadecane	decane	hexane	toluene
Pseudomonas aeruginosa PG201	+	+	-	-
Pseudomonas putidamt -2	-	-	-	+

Table 2: First order growth rate constants in liquid culture by carbon source
STRAIN	Growth rate constant (1/hr, 30 ? C, 150 rpm)
	LB	glucose	decane	hexadecane
Pseudomonas aeruginosa PG201	0.91 ? 0.05	0.47 ? 0.02	0.22 ? 0.01	0.20 ? 0.01
Pseudomonas putida mt-2	1.17 ? 0.03	0.70 ? 0.011	NA2	NA2

¹27? C; previously reported (Holden Ph.D. dissertation, 1995)

²NA = not applicable (see Table 1)

Our work to generate EPS variants has focused on P. aeruginosa. We have acquired a plasmid, SAK5, that causes P. aeruginosa to over-express genes for alginate production. In liquid culture, it is alginate, a uronic acid that is the main exopolymer produced by P. aeruginosa. By triparental mating, we moved the SAK5 plasmid from E. coli into PG201 and have confirmed that total carbohydrate production is higher in PG201::pSAK5. The mass of carbohydrate per mass of protein in liquid culture is 0.55 for PG201 and 1.19 for PG201::pSAK5.

Our biotic micromodel work has confirmed that we can visualize PG201 in the micromodel under saturated conditions.

To determine the conditions in unsaturated sand that affect biofilm formation, we performed batch studies of bacterial growth
and substrate utilization in sand amended with hexadecane (in a carrier or without) and with varying C/N regimes. We also
investigated the effects of atmospheric oxygen on biofilm formation. Our experimental matrix is provided in Table 3. Our
analytical work included total protein analysis, carbohydrate analysis, culturable population size and environmental scanning
electron microscopy. Fixed samples are currently under evaluation for remaining substrate. Differences in the apparent growth
habit were observed when comparing treatments. Our main interest in these results is that C/N ratio appears to affect EPS
production as well as oxygenation. Finally, our images imply that EPS production by PG201::pSAK5 is high in sand. Our
analysis of total protein and carbohydrates of sand cultures will enable our confirmation that the ratio of EPS production to
biomass is similarly high in sand culture as it is in liquid. Our conclusions from this experiment will be used to guide our core
inoculation and substrate application decisions.

Future Activities:

We plan to complete the sequence of abiotic experiments outlined above within the second year of the
project. We plan to complete the colonization studies of the micromodel and begin the experiments on effects of bacteria and
EPS on NAPL flooding and drainage. Finally, we expect to complete wetting and drying experiments with colonized
micromodels.

In the second year, we expect to design and test a prototype core reactor. We have made some progress in this area, but have
not yet completed it. We will perform at least one preliminary experiment with working x-ray and 14C protocols.

We have made contact with several project managers at various field sites to obtain data on the rate of biodegradation of
petroleum hydrocarbon spills as a function of seasonal climatic and soil conditions. We also have proceeded to hire the
post-doctoral researcher who will be in charge of modeling the rate of biodegradation, and make the linkages between the pore
scale and the field scale.

Journal Articles:

No journal articles submitted with this report: View all 20 publications for this project

Supplemental Keywords:

bioavailability, natural attenuation, vadose zone, biofilm, water potential, soil, bioremediation, environmental microbiology, unsaturated zone, diffusion, mass transfer, NAPL, VOC, remediation, groundwater, environmental chemistry and physics, engineering., RFA, Scientific Discipline, Ecosystem Protection/Environmental Exposure & Risk, Waste, Water, Ecological Indicators, Ecosystem Protection, chemical mixtures, Environmental Chemistry, Ecosystem/Assessment/Indicators, Contaminated Sediments, exploratory research environmental biology, Groundwater remediation, Ecological Effects - Environmental Exposure & Risk, Bioavailability, Ecological Effects - Human Health, Hydrology, Bioremediation, Chemical Mixtures - Environmental Exposure & Risk, ecological effects, ecology, bioremediation of soils, sediments, contaminated sediment, biological attenuation, NAPL, exoplymeric substances, seasonal variation, biodegradation, fate and transport, groundwater, water cycles, ecological exposure, natural bioattenuation, vadose zone, chemical transport, sediment

Progress and Final Reports:

Original Abstract

2000 Progress Report

Final Report