United States
Environmental Protection
Agency
Risk Reduction
Engineering Laboratory
Cincinnati, OH 45268
Research and Development
EPA/600/S2-88/049 Dec. 1988
Project Summary
Quick Indicator Tests to
Characterize Bentonite Type
Richard M. McCandless, Andrew Bodocsi, and Hinton K. Howard
Ten commercially available
bentonite products representing
unaltered, polymer protected, and
chemically treated sodium
bentonites were used in laboratory
testing. The purpose of the tests was
to identify a quick, reliable, cost-
effective and field Indicator test
procedure to permit
• identification of bentonite type;
• prediction of the hydraulic
conductivity performance of the
bentonite under chemical or
leachate attack, both in pure
form and when admixed with
other soils for use as slurry
trench backfill.
Eighteen test procedures were
applied in either slurry or powdered
form to several bentonite
concentrations, several acetone
concentrations, and three general
hydration cases. Results
demonstrate that gel strength and
apparent viscosity tests may be used
to distinguish between the specific
unaltered, polymer protected or
extended, and chemically treated
bentonites tested.
A total of ninety-five hydraulic
conductivity tests were performed on
various soil-bentonites and
bentonite filter cakes. Results
indicate that the hydraulic
conductivity of bentonite filter cakes
permeated by a mixture of 25%
acetone in water was not
significantly different from baseline
(0% acetone) results. Moreover,
there was no performance distinction
to be made on the basis of bentonite
type.
Selected indicator parameters
were measured on the same slurries
used to prepare filter cakes for
hydraulic conductivity testing.
Results for the 10-minute gel
strength test suggested an apparent
qualitative correlation with hydraulic
conductivity as a function of acetone
concentration.
This Project Summary was
developed by EPA's Risk Reduction
Engineering Laboratory, Cincinnati,
OH, to announce key findings of the
research project that is fully
documented in a separate report of
the same title (see Project Report
ordering information at back).
Introduction
The Land Pollution Control Division,
Containment Branch, of the Hazardous
Waste Engineering Research Laboratory
(now the Risk Reduction Engineering
Laboratory), is conducting research to
evaluate the effectiveness of earthern
barriers to prevent the movement of
contaminated groundwater from
hazardous waste sites. Of particular
interest are the immediate and long-
term performance characteristics of these
barriers under the influence of organic
contaminants.
Various soil-bentonite mixtures are
utilized as barriers in the form of slurry
cutoff walls. The selection and
specification of a particular bentonite for
use in slurry wall construction is often an
application-dependent decision
involving cost and subjective
performance considerations. Sodium
bentonites are classified by vendors as
unaltered, polymer protected or
extended, and chemically treated.
Currently, there is no test protocol to
distinguish between these bentonite type
classifications or to evaluate relative in-
-------
service hydraulic conductivity per-
formance under chemical or leachate
attack.
The purpose of this research was to
identify a quick, reliable, cost-effective
and field-practical indicator test
procedure.
Procedures
The research was performed in three
consecutive phases from June 1, 1984 to
September 30, 1986. A total of ten
different vendor-supplied bentonite
products were studied utilizing eighteen
test procedures, three different bentonite
hydration procedures and eight test
liquids including deionized water,
acetone in various concentrations, and
100% methanol and xylene. The project
involved almost 2000 separate indicator
test runs and a total of 95 hydraulic
conductivity tests.
Phase I involved the application of ten
indicator tests using two hydration
procedures and three test liquids
(deionized water, 25% acetone, 75%
acetone) to determine the rationale for
and scope of subsequent research.
Phase II involved four additional test
procedures, one additional bentonite
hydration condition, three additional
concentrations of acetone (5%, 15%,
50%), and 100% methanol and xylene.
About 64% of all indicator test data
reported were generated during Phase II
in an effort to establish a statistically
viable data base for evaluation.
The focus of Phase III was threefold:
to further examine the most promising
test procedures; to perform a statistical
analysis to determine which tests, if any,
could be used to distinguish between
unaltered, polymer protected or
extended, and chemically treated
bentonites; and to determine the
relationships, if any, between indicator
test parameters and hydraulic
conductivity. Phase III testing involved
nine test procedures, two of the three
bentonite hydration procedures
previously used, and selected test liquids
(deionized water, 25%, 50%, and 75%
acetone, and 100% methanol).
The ten sodium bentonite products
used in this study were classified as
unaltered, polymer protected/extended,
or chemically treated. The term
"unaltered" indicates a naturally
occurring sodium bentonite or a blend of
different grades of naturally occurring
sodium bentonite that has not been
subjected to chemical or physical
alteration or modification. The term
"polymer protected or extended"
indicates a naturally occurring sodium
bentonite that has been polymer
"amended" either to enhance its
resistance to chemical degradation or to
improve its yield for economic reasons.
The category termed "chemically
treated" represents products that are
designed to resist in situ degradation
under chemical or leachate attack.
The majority of indicator tests used in
the course of this study involved the
testing of a bentonite/water slurry or a
bentonite/water/chemical slurry and may
be termed "slurry characterization" tests.
These tests are generally employed in
bentonite manufacture and by
consumers of bentonite products to
monitor key slurry properties such as
viscosity and gel strength. The scope of
testing also included several tests that
are generally associated with either
geotechnical engineering or
environmental science. In addition, a few
new test procedures were developed
in-house or adopted from other
consultants and researchers evaluating
special bentonite testing methods. These
tests and the Phase in which they were
used are shown in Table 1.
The particular procedure used to
prepare samples for testing depended
primarily upon the type of test and
whether or not the bentonite was to be
tested in slurry form. Tests such as free
swell and liquid limit were performed on
powdered bentonite samples as supplied
from the vendor. Other tests, such as the
modified swell test, involved chemical
modification of bentonite prior to testing
in powdered form.
The procedure involving powdered
bentonite was used to prepare both
standard "baseline" (0% chemical)
slurries and chemically "hydrated"
water/chemical solutions. Data for
baseline slurries generally reflect the
normal level of response of the
parameter of interest in typical slurry
applications. Data for the chemically
"hydrated" case were not intended to
reflect a normal slurry application, but
rather, served to represent a different
hydration case purely for type-
distinction purposes.
The procedure involving a water
hydrated bentonite was employed both
to provide a third slurry test case for
type-distinction/characterization
purposes and to reflect the more typical
situation of chemical impact after full
water hydration of the bentonite. This
procedure involved the preparation of a
standard paste for each bentonite which
was allowed to hydrate for several days.
This paste was then mixed in the design
proportion with the various water/acetone
solutions to achieve the desired fi
slurry for testing.
All slurries were cured in airtic
containers in a high humidi
environment for a period of 24 hoi
prior to testing. This curing time v\
maintained in order to standardi
bentonite hydration time and achieve
nearly full bentonite hydration state.
Conclusions and
Recommendations
Within the scope of this study it
possible to distinguish betwe
unaltered, polymer protected/extendi
and chemically treated bentonite typ
on the basis of results from select
indicator tests. Based upon a one-w
analysis of variance of all results
selected tests, the 10-second (
strength test, the 10-minute gel streni
test, and the apparent viscosity test rr
be used to distinguish between the thi
bentonite types by use of a chemic!
"hydrated" 25% acetone slurry havim
bentonite concentration of 5%. Wh
fully water-hydrated prior to exposure
acetone concentrations in the range
0% to 50%, the behavior of the bentor
types is not sufficiently different to per
type-distinction. Although seve
indicator test procedures permit limi
type-distinction, those employing I
rotational viscometer to measure vark
bentonite slurry parameters are the m
useful for type-distinction purposes.
Initial bentonite concentration of
slurry has little, if any, measurable eff
on the equilibrium hydraulic conductiv
of filter cakes formed via slurry filtrati
regardless of the acetone concentrat
of the permeant liquid (up to 50%) e
bentonite type. The greatest reduction
hydraulic conductivity of filter cal<
relative to baseline (0% acetone) rest
is at an acetone concentration of 25
with little or no additional reduction v
increasing acetone concentration up
50%. For the range of aceto
concentrations tested herein, bentor
concentration has a greater effect on
hydraulic conductivity of a bentoni
amended soil than bentonite type. 1
maximum observed difference
hydraulic conductivity between differ
soil-bentonites due to bentonite type
about one order of magnitude.
A quantitative correlation betwe
indicator test parameters studied and
hydraulic conductivity of pure bentoni
and soil-bentonites permeated
varying concentrations of acetone can
be demonstrated; however, qualital
correlation between hydrau
-------
Table 1. Indicator Tests
Indicator Test
Research Phase
1
2.
3.
4.
5.
6.
7.
8.
9.
10
11.
12
13
14
15
16
17
18
Marsh Funnel Viscosity
10-second Gel Strength
10-mmute Gel Strength
Apparent Viscosity
Yield Point
Plastic Viscosity
API Filtrate Loss
Unit Filtrate Loss
Modified Swell
Free Swell (2-hr)
Free Swell (24-hr)
Mud Weight
Specific Conductance
pH
% retained, #200 Sieve
Turbidity
Cracking Pattern
Liquid Limit
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
conductivity and 10-minute gel strength
was observed for filter cake samples
representing each of the three bentonite
types This apparent correlation has no
practical utility since acetone was
observed to have an insignificant effect
on the hydraulic conductivity of bentonite
filter cakes. A similar qualitative
correlation was not observed for soil-
bentonites under any set of test
conditions
Although a type-distinction
capability was demonstrated for the
specific unaltered, polymer
protected-'extended, and chemically
treated bentonites tested herein,
utilization of such should not be initiated
until substantial verification testing is
completed on a much wider sampling of
the population of commercial bentonite
products The question of which
bentonite type yields the best long-term
bentonite-amended clay barrier for
different contaminant therefore remains
unanswered Any future research effort to
resolve this question should focus on
long-term chemical exposure of the
barrier and should employ several
priority pollutants as permeant liquids
instead of an "artificial" chemical
permeant which produces data of highly
limited utility
A qualitative correlation between the
10-mmute gel strength test and the
hydraulic conductivity of filter cakes
permeated by acetone can be
demonstrated, however, such a
demonstration is not possible in the case
of soil-bentonites Apparently there is
no substitute for long-term multiple pore
volume hydraulic conductivity testing;
therefore, the concept of a "quick
indicator," especially m the case of
bentonite-amended clays, may be
tenuous
-------
Richard M. McCandless and Andrew Bodocsi are with the University of
Cincinnati, Cincinnati, OH 45221; and the EPA author Hinton K. Howard
is with the Risk Reduction Engineering Laboratory, Cincinnati, OH 45268.
Walter E. Grube, Jr. is the EPA Protect Officer (see below).
The complete report, entitled "Quick Indicator Tests to Characterize Bentonite
Type," (Order No. PB 88-244 033/AS; Cost: $32.95, sub/ect 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:
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
POSTAGE & FEES PAIt},,
• EPA
/ PERMIT Na. G^3& -.
Official Business
Penalty for Private Use $300
EPA/600/S2-88/049
000031V
------- |