United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
Ada OK 74820
Research and Development
EPA/600/S6-90/004 Sept. 1990
&EPA Project Summary
Laboratory Investigation of
Residual Liquid Organics from
Spills, Leaks and the Disposal of
Hazardous Wastes in
Groundwater
John L. Wilson, Stephen H. Conrad, William R. Mason, William Peplinski, and
Edward Hagan
Organic liquids that are essentially
immiscible with water migrate
through the subsurface under the
influence of capillary, viscous, and
buoyancy forces. These liquids
originate from the improper disposal
of hazardous wastes, and the spills
and leaks of petroleum hydrocarbons
and solvents. The laboratory studies
described in this report examined
this migration, with a primary focus
on the behavior of the residual
organic liquid saturation, referring to
that portion of the organic liquid that
is trapped by capillary forces in the
soil matrix. Residual organic
saturation often constitutes the major
volume of the organic liquid
pollution, and acts as a continual
source of dissolved or vapor phase
organics.
Four experimental methods were
employed. First, quantitative dis-
placement experiments using short
soil columns were performed to
relate the magnitude of residual
organic liquid saturation to fluid
properties, the soil, and the number
of fluid phases present. Second,
additional quantitative displacement
experiments using a long soil column
were performed to relate the
mobilization of residual organic liquid
saturation in the saturated zone to
wetting fluid flow rates. Third, pore
and blob casts were produced by a
technique in which an organic liquid
was solidified in place within a soil
column at the conclusion of a
displacement experiment, allowing
the distribution of fluid phases within
the pore space to be observed. The
columns were sectioned and
examined under optical and scanning
electron microscopes. Photomicro-
graphs of these sections show the
location of the organic phase within
the porous soil matrix under a variety
of conditions. Fourth, etched glass
micromodels were used to visually
observe dynamic multi-phase
displacement processes in pore
networks. Fluid movement was
recorded on film and videotape.
It was found that the spatial
distribution and saturation of organic
liquid within the porous media
depends on a variety of factors,
including: (1) the fluid properties of
interfacial tension, viscosity, and
density; (2) the soil structure and
heterogeneity; (3) the number of fluid
phases present; and (4) the fluid flow
rates. Photomicrographs on a pore
scale show that the residual organic
liquid appears as blobs, films, rings,
and wedges of microscopic size,
depending on these factors. The size,
shape, and spatial distribution of
these blobs, films, rings and wedges
affects the dissolution of organic
liquid into the water phase,
Printed on Recycled Paper
-------
volatilization into the air phase, and
the adsorption and biodegradation of
organic components. These four
processes are of concern in the
prediction of pollution migration and
the design of aquifer remediation
schemes.
This Project Summary was
developed by EPA's Robert S. Kerr
Environmental Research Laboratory,
Ada, OK, to announce key findings of
the research project that is fully
documented in a separate report of
the same title (see Project Report
ordering information at back).
Introduction
Many hazardous waste sites, and most
leaking underground storage tanks,
involve non-aqueous phase organic
liquids. Usually released at or near the
surface, these organic liquid
contaminants move downward through
the vadose zone toward the water table.
Migrating as a liquid phase separate from
the air and water already present in the
vadose zone, some of the organic liquid
is immobilized within the pore space by
capillary forces. The remainder passes
on, and if the volume of organic liquid is
large enough it eventually reaches the
water table. If it is less dense than water
the organic liquid spreads laterally along
the water table (see right side of Figure
1). If the organic liquid is more dense
than water, it continues to move
downward into the saturated zone (the
left side of Figure 1). In both cases the
organic liquid usually migrates down-
gradient with the ambient groundwater
flow, although dense organic liquids may
migrate in other directions as they
encounter dipping barriers. In the
saturated zone, which is mostly below the
water table and includes the capillary
fringe, more organic liquid is immobilized
by capillary forces. Here the immobilized
organics remain as small, disconnected
pockets of liquid, sometimes called
blobs, no longer connected to the main
body of organic liquid. The immobilized
volume is called the residual oil
saturation in petroleum reservoir
engineering and is measured as the
volume of organic liquid trapped in the
pores relative to the volume of the pores.
The final report refers to the immobilized
organic liquid as residual organic liquid.
Organic liquid at residual saturation can
occupy from 15% to 50% of the pore
space in petroleum reservoir rocks under
conditions that are equivalent to those in
the groundwater saturated zones. At a
spill or hazardous waste site the entire
volume of organic liquid can be
exhausted by this immobilization,
although if the volume of organic liquid is
large enough, it continues to migrate
down-gradient where it becomes a threat
to the safety of drinking water or
agricultural water supplies. As described
in detail in sections 9 and 10 of the final
report, the actual spatial distribution of
the residual saturation within the pore
space is completely different in the
vadose and saturated zones.
The organic liquid phase is sometimes
referred to as being immiscible with
water and air. Although that expression is
used here, it is important to realize that
small concentrations of the various
components of the organic phase
volatilize into the air phase and dissolve
into the water phase. A halo of dissolved
organic components precedes the
immiscible phase in its migration (Figure
1). Even when the so-called immiscible
organic liquid has been immobilized by
capillary trapping, the passing
groundwater dissolves some of the
residual. In effect, the organic liquid
phase acts as a continuing source of
dissolved organic pollutants. Similarly, in
the vadose zone, the residual organic
liquid that volatilizes into the air phase
migrates by gaseous diffusion and
advection, becoming a source of organic
contaminants to air or water and a
possible explosion hazard. In large spills
and leaks it is apparent that most of the
liquid organic remains as a liquid, some
is volatilized, and a little is dissolved.
However small in volume, the volatilized
or dissolved components are usually the
ones that cause problems. The liquid
organic phase acts as a reservoir of
additional organic to replenish the air and1
water phases with dangerous and/or toxic
material. Clearly, the source of the
dissolved or gaseous organic
constituents—the liquid organic phase —
must be removed or isolated in order to
restore a polluted aquifer.
There is no wholly effective mechanism
to remove the residual organic liquid.
Waiting for the residual to dissolve can
take several decades. In the vadose
zone, induced volatilization may help
reduce the residual volume for lighter
organics, but is not effective for heavier
ones. Engineered removal is usually
attempted hydraulically, by sweeping the
organic liquid out with water, or
biologically, by encouraging the
consumption of the organic constituents
by the soil rnicrobial community. This last
process, biodegradation, is the focus of
current research and several recent
restoration efforts. It is seldom tried
alone, for the microbes generally
consume only the dissolved organics.
Moreover, some organic chemicals are
extremely resistant to biodegradation.
PCBs, for example, may biodegrade very
slowly, or not at all in the subsurface.
Hydraulic sweeps remain a major
component of any attempt to remove
organic liquids although, commonly,
hydraulic sweeps fail to remove all the
liquid organic phase, often leaving a
significant quantity of residual organic
liquids behind. There is, of course,
another removal option often used for
small pollution events: excavate the site
and dispose of or treat the contaminated
soil. For large sites this alternative is
unfeasible. Since there is no panacea for
the removal of organic liquids,
containment is often adopted as part of a
restoration strategy. Hydraulic
containment, often in combination with
structural barriers such as a slurry wall, is
becoming standard practice.
Scope of Previous Work
Development of improved technologies
to clean up organic pollutants depends in
large part on developing an ability to
understand and predict the migration of
liquid, vapor, and dissolved organics.
Liquid organics move through a water
and sometimes air filled porous soil, as a
separate phase, under the influence of
viscosity, gravity, and capillary forces.
Dissolved organics move in the water
phase and are subject to advection,
dispersion, biodegradation, and
adsorption onto soil particles. Organic
vapors in the air phase are subject to
similar mechanisms. A few of these major
transport mechanisms are fairly well
understood today, principally those
associated with the behavior of dissolved
organics.
In contrast, the organic liquid phase
transport mechanism has been virtually
ignored by the research community in
the United States, although it has been
the subject of empirical studies in
Europe. Recently, however, American
researchers have obtained some
laboratory results." Several investigators
ran gravity drainage experiments on a
long column to relate organic liquid
retention in the vadose zone with grain
size and sorting. Others used a short soil
core centrifuging method to measure
residuals in the vadose zone, or
'The final report provides complete references for
all studies and experiments.
-------
^
ground surface
i J« *\
* • r i
« ' N
residual .
organic
liquid
saturation
capillary
water table
Figure 1. Migration pattern for an organic liquid more dense than water (left), and less dense
than water (right).
employed gamma radiation attenuation
and bulk soil electrical resistivity to
measure three-phase fluid saturations at
various times and at various elevations
above a water table following a simulated
petroleum spill. The experimental
procedure allowed a petroleum spill to be
tracked as it moved through the vadose
zone to the water table. Experiments
have been performed to test the ability of
multiphase flow theory to predict the
infiltration and redistribution of wetting
and non-wetting fluids. They met with
limited success. American researchers
have also used theoretical three-phase
saturation-pressure relationships to
estimate the volume of oil in soils given
observed fluid levels in monitoring wells.
Some simple numerical simulations of
multi-phase transport have been
developed. These focus on immiscible
transport of continuous phases. Residual
organic liquids, trapped by capillary
forces, are often ignored, although they
are sometimes treated as a source of
dissolved contamination. This research
effort mirrors the state of the art of
petroleum engineering's black oil
models. A few researchers have looked
into interphase transfer, including the
volatilization and solution of organic
components, using computer simulations.
This again reflects the state of the art in
petroleum engineering, where so-called
compositional models are used to
examine enhanced recovery techniques.
One study discussed in the final report
proposed a model to estimate the
functional relationships between fluid
pressures, saturations, and permeabilities
of two- or three-phase porous media
systems, and these functional
relationships have been implemented in a
multi-phase numerical flow model. The
model has since been extended to
include the effects of hysteresis and non-
wetting phase trapping. The results of
concurrent laboratory work were used to
validate the model.
Petroleum engineering's long history of
research into improving recovery from
petroleum reservoirs may be applied to
rehabilitating fresh-water aquifers polluted
by organic liquids. Through over forty
years of experimentation, petroleum
engineering has amassed considerable
expertise in multi-phase transport, the
mechanics of oil phase capillary trapping,
and oil recovery. To date, relatively little
of this technology has been applied to
recovering organic hazardous wastes and
petroleum hydrocarbons released in the
near-surface environment. The petroleum
literature on residual oil saturation is
reviewed in papers referenced in the final
report. In groundwater hydrology we too
are concerned with the capillary trapping
of residual saturation, and with its
removal. However, unlike petroleum
engineers, we are also concerned with
the mechanisms that initially brought the
oil into the aquifer in the first place.
Motivation for this Study
Residual organic liquid saturation often
constitutes the major volume of the
organic pollution, and acts as a continual
source of dissolved or vapor phase
organics. In particular, there is a need to
understand how the residual organic
liquid is trapped and how it can be
hydraulically mobilized or otherwise
removed. As shown in Sections 9 and 10
of the full report, the residual organic
liquid appears to form blobs, films,
wedges and rings of microscopic size,
depending on the presence of other
fluids, the port geometry, the surface
wetting of the solids, and soil
heterogeneity. The term wetting refers to
the relative affinity of the solid surface for
-------
the available fluids. Water is normally the
wetting fluid in most soils. Organic liquid
is normally non-wetting relative to water,
and wetting relative to soil gas. The size,
shape, and spatial distribution of these
blobs, films, wedges and rings affects the
dissolution of organic liquid into the water
phase, volatilization into the air phase,
and the adsorption and biodegradation of
organic components. The presence of
residual organic liquid also affects the
relative permeability versus saturation
curves used in numerical simulation
codes of fluid movement and pollution
migration. A paucity of experimental
results regarding these issues makes site
characterization conjectural, predictive
modeling unreliable, and remediation
design of organic liquid leak or waste
sites less effective than might be
possible.
Objectives
The goal of this study was to better
understand the basic physical
mechanisms controlling the movement,
and especially the capillary trapping, of
organic liquids in soils and groundwater.
Emphasis was on relating the various
mechanisms to the issues of contaminant
movement, characterization, and
remediation. This broad goal was broken
down into two sets of specific research
objectives, addressing issues relevant to
the saturated and vadose zones,
respectively:
The Saturated Zone
Assuming that water is wetting and the
organic liquid is non-wetting, our
research objectives for saturated zone
conditions were to:
• conduct a literature review of basic
concepts, including non-wetting phase
capillary trapping and mobilization
mechanisms, and petroleum
experience;
• conduct experiments that permit the
visualization of two-phase fluid flow and
capillary trapping, and record the
visualizations on film and videotape;
• perform a detailed study of two phase
flow capillary trapping and non-wetting
phase residual saturation in a typical
unconsolidated soil, testing the
hypothesis that its behavior can be
predicted from previously published
results from the petroleum engineering
literature;
• compare non-wetting phase residual
saturations for various organic liquids,
testing the hypothesis that residual
saturation is largely independent of
organic liquid composition for expected
conditions in hydrology;
• compare non-wetting phase residual
saturations for various soils, testing the
hypothesis that residual saturations
should be similar in soils that have a
similar grain size distribution;
• investigate how the rate of initial
invasion of a non-wetting organic liquid
may influence irreducible water
saturations and, later, organic residual
saturations
• investigate the possible hydraulic
mobilization of non-wetting phase
residual organic liquid, by increasing
groundwater velocities, testing other
researchers' conclusion that this is
largely an unrealistic aquifer remedia-
tion alternative unless interfacial
tensions are reduced significantly; and
• test the hypothesis that porous media
heterogeneity can dominate displace-
ment and trapping mechanisms.
The Vadose Zone
Our research objectives for vadose
zone conditions were to:
• conduct a literature review of basic
concepts, including capillary trapping
mechanisms, mobilization issues, and
petroleum experience;
• conduct experiments that permit the
visualization of multi-phase fluid flow
and capillary trapping, testing the
hypothesis that spreading organic
liquids typically form a film between
the water and air phases;
• perform a detailed study of capillary
trapping and residual saturation in a
typical unconsolidated soil, testing the
hypothesis that organic liquid residual
saturations are significantly lower in the
vadose zone than they are in the
saturated zone; and
• conduct experiments that permit the
visualization of capillary trapping for
non-spreading organic liquids, testing
the hypothesis that non-spreading
organic liquids behave differently than
non-spreading organic liquids.
Experimental Approach
The problem was approached
experimentally in four ways:
1. Quantitative displacement experi-
ments using short columns were
performed to relate the magnitude of
residual organic liquid saturation to
fluid and soil properties, and to the
number of fluid phases present (i.e.,
both saturated and vadose zone
conditions).
2. Quantitative displacement experi-
ments using long columns were
performed under two-phase saturated
zone conditions, yielding water and
organic liquid relative permeabilities.
In these experiments, reductions of
residual organic saturation were
correlated to the pressure gradient
applied in hydraulic sweeps, and the
potential for hydraulic mobilization of
residual blobs was investigated.
3. Pore and blob casts were produced
for saturated zone conditions by a
technique in which the organic liquid
was solidified in place within a soil
column at the conclusion of a
displacement experiment, allowing the
distribution of organic liquid to be
observed. The polymerized organic
phase was rigid and chemically
resistant. Following polymerization, the
water phase was removed and
replaced by an epoxy resin. The solid
core, composed of soil, solidified
styrene (the organic phase), and
epoxy resin (the water phase), was cut
into sections to show the organic liquid
phase in relation to the soil and the
water phase. The sections were
photographed under an optical
microscope. Although polymerization
only gave a snapshot of the
displacement process, it offered the
advantage of seeing organic liquid in
its natural habitat (i.e. within a soil) as
compared to that observed in etched
glass micromodels. Sometimes,
instead of replacing the water with
epoxy resin, the solid matrix of the soil
column was dissolved with
hydrofluoric acid, leaving only the
hardened organic liquid. The solidified
organic phase was then observed
under a scanning electron microscope
(SEM) and photographed. For vadose
zone conditions, styrene and epoxy
liquids were sequentially applied,
drained and hardened in an attempt to
simulate proper fluid distributions
above the water table. The resulting
pore casts were photographed under
an optical microscope.
4. Etched glass micromodels were
used to observe dynamic multi-phase
displacement processes. Micromodels
-------
provide two-dimensional networks of
three-dimensional pores. They offer
the ability to actually see fluids
displace one another in both a bulk
sense and also within individual pores.
Although displacements are known to
be dependent upon a variety of
factors, this report describes
micromodel experiments that focused
on only three: (1) the fluid flow rate, (2)
the presence of heterogeneities, and
(3) the number of fluid phases present.
The experiments were photographed
and videotaped.
To interpret the experiments in
heterogeneous material, we also
developed a new but simple theoretical
model of multiphase flow and capillary
trapping. The model is based on the
interplay between viscous and capillary
forces.
Organization of the Final Report
Following the introduction are two
sections (2 and 3) that summarize the
report conclusions and recommen-
dations. The next five sections (4 through
8) detail fluid and soil characteristics, and
the experimental methodology, used for
each of the experimental approaches
outlined above. These sections contain
detailed information that may be used by
future investigators wishing to verify or
extend the results of this study. The
reader more concerned with results than
methods can probably skip them. The
last two sections (9 and 10) describe
experimental results for saturated zone
and vadose zone conditions, respectively.
These sections contain a large number of
photomicrographs that visualize multi-
phase flow and residual saturation.
•U.S. Government Printing Office: 1993 — 750-071/60259
-------
-------
-------
John L Wilson, Stephen H. Conrad, William R. Mason, William Peplinski, and
Edward Hagan are with New Mexico Institute of Mining and Technology,
Socorro, NM 87801.
Jerry Jones is the EPA Project Officer (see below)
The complete report, entitled "Laboratory Investigation of Residual Liquid Organics
from Spills, Leaks and the Disposal of Hazardous Wastes in Groundwater,"
(Order No. PB90-235 7971 AS; Cost: $31.00, subject to change) will be
available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Robert S. Kerr Environmental Research Laboratory
U.S. Environmental Protection Agency
Ada, OK 74820
United States Center for Environmental Research
Environmental Protection Information
Agency Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S6-90/004
------- |