United States
Environmental Protection
Agency
Environmental Monitoring
Systems Laboratory
Las Vegas NV89114
Research and Development
EPA-600/S4-84-057 July 1984
<8>EPA Project Summary
Single Laboratory Evaluation
of the Hydrogen Oxidation
Soil Bioassay
Robert D. Rogers
The Hydrogen Oxidation Soil Bio-
assay was single laboratory tested as a
potential method for monitoring
hazardous wastes and hazardous waste
sites. The bioassay is based on the rate
of hydrogen consumption by soil
microorganisms. Oxidation of
hydrogen to water is inhibited when
various pollutants are present in the soil
and the rate of this reaction can be used
as an indication of potential hazard to
the soil ecosystem.
The single laboratory evaluation
included ruggedness testing, a deter-
mination of method sensitivity and
precision, and tests to determine the
limits of reliable measurement. Since
there was no "true value" or "true
response" to a reference material, the
method's capability for bias (systematic
error) was not determined. Aqueous
solutions of mercuric chloride were
used as sample material during the
evaluation. Some preliminary tests
were also conducted using both organic
compounds and ' actual hazardous
waste samples.
The bioassay was found to be
"rugged" in the sense that modest pro-
cedural variations did not produce an
altered test result. The method's
capability for precision, expressed as a
CV of 7.8 percent, was determined
by conducting 10 separate bioassays
using the same concentration of
mercuric chloride. Within a mercury
concentration range of 10 ppm to 150
ppm, the technique was capable of
distinguishing between concentration
differences of 25 ppm. The limits of
reliable measurement were established
at 10 ppm and 750 ppm mercury when
mercuric chloride solutions are used as
sample material. The complete
Hydrogen Oxidation Soil Bioassay
protocol, the results of chemical
analyses (i.e., gas chromatography/
mass spectrometry, atomic absorption,
and inductively coupled argon plasma
spectroscopy) conducted on actual
samples that were used during the
single laboratory test, and the
preliminary bioassay responses to
different types of sample material are
included in the project report.
Before this bioassay can be considered
for collaborative testing, it will be
necessary to conduct portions of the
single laboratory test again using
sample material that more realistically
simulates either a hazardous waste site
leachate, or an analytical fraction of
actual hazardous waste material.
However, results from this evaluation
suggest that this terrestrial monitoring
technique should ultimately be a
candidate method for collaborative
testing and should be of subsequent
benefit to a hazardous waste
monitoring network.
This Project Summary was developed
by EPA's Environmental Monitoring
Systems Laboratory, Las Vegas. NV, 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
bach).
Introduction
Single laboratory testing is used to
establish the data quality that can be
achieved within a single laboratory. It
also provides a basis for deciding whether
-------
or not a given method merits
collaborative testing. The previously
prepared guidelines for testing biological
methods (EPA-600/S4-83-056) have
been followed during this evaluation.
This approach calls for an identification of
procedural variables that must be
carefully controlled (ruggedness testing),
determination of method sensitivity and
precision, and identification of the limits
of reliable measurement. In addition, if
the response of the evaluated method to a
standard reference material is known
(true response), method bias should also
be determined.
The Hydrogen Oxidation Soil Bioassay
is based on the bio-oxidation of hydrogen
(H2) to water (H20) hy the hydrogenase
enzyme. This enzyme is of microbiolog-
ical origin (their ability to oxidize H2 is
documented). The microbial forms are
assumed to be ubiquitous in soil. If tritium
(designated as either 3H2, T2, or HT; HT is
used in this Summary) is introduced, it is
oxidized in proportion to its abundance
(Figure 1). The product (HTO) from this
reaction can be used to determine the
rate of H2 oxidation in a given soil.
With experimentation, it has been
found that HT oxidation rates in soil
amended with toxic compounds
decreased linearly with the log of the
compound concentration. This has been
shown to occur with toxic liquid, solid,
and gaseous compounds.
During the evaluation it was assumed,
that if the technique was being routinely
used, sample material (e.g., actual
hazardous waste material, leachate,
analytical fractions of hazardous waste
material, water samples that potentially
contain waste site chemicals, etc.) would
be sent to the assay laboratory. Soil from
a hazardous waste site area could
potentially be used as test material, but
the method has not been evaluated for
this type of application. During the single
laboratory test, the method protocol was
strictly followed. A copy of the complete
Hydrogen Oxidation Soil Bioassay
protocol is included in the project report.
Procedure
The soil used for the bioassay is a
Calico series fine sandy loam (Aquic
Xerofluvent) from southeastern Nevada.
Chemicals used as sample material
were all reagent grade and included
mercuric nitrate [Hg(N03)2], cadmium
chloride (HgCI2), silver nitrate (AgN03),
cadmium nitrate [Cd(N03)2], mercuric
chloride (CdCI2), pyrocatechol, m-chloro-
phenol, and p-chlorophenol. Samples
G
,.
.8
.•3
i
0.7
0.6
0.5
0.4
0.3
0.2
0.1
i i r
i i i i i i i
07234567
Time (h)
Figure 1. HT oxidation in soil with time.
10 11
12
which had been collected from an actual
hazardous waste site were also used
during a preliminary evaluation (Table 1).
Testing was initiated by adding 10 ml of
water carrying the toxic compounds,
distilled water if a control, to 100 g of the
air dried soil residing in a 1-L, round
bottom flask. The treated soil was then
incubated at 25°C in the dark for 16
hours. The assay must be conducted in
triplicate (triplicate flasks for each
treatment and control) so that a mean and
coefficient of variation (CV) can be
determined.
Following the incubation period, each
flask was flushed with air for 10 s
(20L/min) and then sealed with a rubber
stopper. Immediately after that, 5 mL of
nitrogen (N2) containing 0.5/^Ci of HT was
injected through the stopper. After
charging, the flasks were returned to the
environmental chamber (25°) for an
additional 2 hours. The HT oxidizing
reaction was stopped at the precise time
(2 hours) by flushing the flasks with air.
Since the consumption of the H2/HT is
rapid and begins immediately, care was
used to charge the flasks sequentially
Table 1.
Major Components of Hazardous Waste Site Material Used During the Method
Evaluation*
Semivolatile Organics (ug/L)
phenol
2-nitrophenol
benzoic acid
pentachlorophenol
alpha-BHC
beta-BHC
gamma-BHC
delta-BHC
fluoranthene
dibenzofuran
trimethylnaphthalene
benzofcjfluoranthene
benzo(a)pyrene
di-n-butylphthalate
2-6-dinitrotoluene
1.2-benzene dicarboxylic acid
Volatile Organics (ug/L)
trichloroethylene
benzene
toluene
methylene chloride
Inorganics (mg/L)
aluminum
copper
iron
sodium
nickel
selenium
zinc
^Hazardous waste sample analysis conducted by Acurex Corporation, Mountain View,
California 94039
-------
with the same order being followed when
the reactions were stopped. To determine
the amount of HT oxidized, the reaction
product, HTO was first recovered from the
soil by distillation (Figure 2). Then the
quantity of HTO in the distillate was
determined by liquid scintillation analy-
sis. An outline for the entire method is
shown in Figure 3.
Treatment effects caused by each test
compound were determined by calcula-
ting a ratio of zero-time reaction rates for
treatments and controls. Reaction rates
were calculated from the exponential
growth model
= P, [1-exp(-P2t)]
where:
Y = tritium content
Distillation
Column
Moisture
Receiving
Trap
1-liter
Boiling
Flask
Figure 2. Distilling apparatus for the col-
lection of tritiated water from
soil exposed to tritium during the
Hydrogen Oxidation SoilBioassay.
P, = the asymptotic tritium content
P2 = the reaction rate parameter
t — time in hours
E = the error function (assumed to
be Gaussian).
By entering the asymptotic HT content
(the amount injected) and the amount of
HTO recovered at the time sampled, the
rate of hydrogen oxidation (P2) can be
calculated. The rate of hydrogen
oxidation in treated soil (P2t or P2
treatment), divided by the rate of
hydrogen oxidation in untreated soil (P2c
or P2 control), is then tabulated for each
treatment concentration. These data are
then graphically plotted.
Results
Preliminary test results confirmed that
the assay was responsive to aqueous
solutions of Hg, Ag, and Cd, to metal
combinations, to phenolic compounds,
and to actual hazardous waste site
material (i.e., metals added to waste site
material to provide a more realistic
sample matrix). Aqueous solutions of
mercuric chloride were used as sample
material during the single laboratory
evaluation because Hg appeared to be
So/7 fWOg)
Amend with test
material in 10 mL H2O
Incubate
16h
more toxic than Ag and Cd, and because
HgCI2 is more soluble than Hg(N03)2. An
aqueous solution of mercuric chloride
might also be used as a much simplified
example of a hazardous waste site
leachate. Qualitative and quantitative
analyses were conducted on all mercuric
chloride sample material to confirm the
chemical composition and to ensure lack
of sample contamination.
The first phase of the single laboratory
evaluation was to determine if minor
departures from the method protocol
would result in an altered bioassay result.
A method's ability to produce an
unaltered test result when subjected to
minor procedural variations is an
indication of method ruggedness. If the
results are altered by small procedural
variations, it is important to emphasize in
the protocol that a specific step must be
strictly followed or, in some cases, to
provide more detail on any quality control
steps associated with the critical pro-
cedure.
The prescribed method procedure and
the corresponding procedural variations
used during the ruggedness evaluation
are summarized in Table 2. The seven
protocol directed procedures (A-G) were
chosen because they are the ones which,
in our judgment, could inadvertently be
altered as indicated by the variations a-g.
Interpret
results
Analyze for
amount
HT - HTO
Prepare
scintillation
cocktail
Add tritium
Incubate
2/7
Collect water
by distillation
Figure 3, Outline of procedures for the Hydrogen Oxidation Soil Bioassay.
3
-------
Table 2. Variations in the Hydrogen Oxidation Soil Bioassay Used to Determine "Ruggedness"
Item Protocol Directed
Variation
1. Length of time a flask is purged with air A. Purge time 10 s
both before introduction of HT and to
flush out remaining HT after incubation.
2. Length of time soil is pre-incubated with B. Preincubation time 16 h
test compound.
a. Purge time 6 s
b. Preincubation time 20 h
3. Length of time soil is incubated in pres- C. Incubation time 120 min
ence of HT.
4. Amount of water, containing test com- D. Amount water 10 mL
pound, applied to soil.
5. Frequency of mixing soil after application E. Frequency of mixing 2 beats/s
of water.
6. Quantity of HTO derived from distillation F. Quantity of HTO 8 mL
which is mixed with liquid
scintillation cocktail.
7. Amount of HTO distilled from soil.
C. Amount of HTO distillate 15 mL
c. Incubation time 135 min
d. Amount water 11 mL
e. Frequency of mixing 1 beat/s
f. Quantity of HTO 7.9 mL
g. Amount of HTO distillate 17 mL
The protocol directed procedures and the
corresponding procedural variations
were then arranged into a series of eight
trials. Each trial consisted of a single
analysis of a single concentration of
HgCI2 (50 ppm Hg) and a pre-selected
combination of procedural variations.
Basically, the procedural variations had
little effect on the assay response and,
based on the ruggedness test results, it
was not considered necessary to revise
the method protocol. The assay is
"rugged" in the sense that modest
variations in method procedure would not
be expected to alter the assay result.
Method precision was determined by
conducting 10 separate tests with each of
the separate determinations represent-
ing a valid test response. Testing was
conducted on alternate days and used 75
ppm Hg as sample material. The average
response to this treatment was 61.6
percent reduction in HT oxidation with a
CV of 7.8 percent.
In the context of a single laboratory
test, a method's sensitivity is defined as
its capability to respond to small changes
in the concentration of a test compound.
The ability of this bioassay procedure to
distinguish between changes in Hg
concentration was initially tested using
one concentration greater than that used
for the precision determination and one
lower. Ten independent analyses were
conducted for each of the new concentra-
tions. If the method can distinguish
between the concentration used during
the precision determination and the two
newly selected concentrations, the
concentration interval is reduced and
additional concentrations are tested. For
this single laboratory test, the process
was repeated until the concentration
interval had been reduced to 25 ppm.
The concentration used for the
precision test was 75 ppm with 10 and
150 ppm being the initially tested
extremes and with 50 and 100 ppm as
midpoints between the reference
concentration and the extremes. It was
therefore possible to determine if the
bioassay could initially distinguish
between Hg concentrations of 75 ppm (65
ppm in the case of the lower concentra-
tion) and then between concentration
differences of 25 ppm.
Results for the sensitivity
determination are included in Table 3.
The bioassay was capable of
distinguishing between Hg concentration
differences of 75 ppm and between
differences of 25 ppm (significant at the 5
percent level). Therefore, the method's
single laboratory capability for sensitivity
has been presented as 25 ppm Hg when
aqueous solutions of mercuric chloride
are used as sample material.
Tests to establish the limits of reliable
measurement should determine the
sample concentration range for which
the method is capable of providing useful
data. In some instances, the single
laboratory test may simply verify that the
method capabilities for sensitivity,
precision, and accuracy (if applicable) do
not deteriorate at the upper and lower
extremes of the detection range.
Three additional concentrations of Hg
were used in addition to those used for
the method sensitivity test. These new
concentrations were 500 ppm, 750 ppm,
and 1,000 ppm. Table 3 presents a
compilation of test data obtained from
both the additional concentrations and
from the concentrations used for the
sensitivity determination. The results
indicate that the method was sensitive to
incremental increases of Hg up to 750
ppm. Results betwen 750 ppm and 1,000
ppm were not statistically distinguish-
able. As noted previously, test results
from the lower concentrations were
distinguishable. Between 10 ppm and
150 ppm of Hg, the method's capability
for precision can be described as having a
CV range of 5.7 to 11.0 percent. Method
capability for precision suffered from 500
ppm to 1,000 ppm. The limits of reliable
measurement forthe Hydrogen Oxidation
Soil Bioassay are presented as 10 ppm
and 750 ppm Hg when aqueous solutions
of mercuric chloride are used as sample
material.
Conclusions
The Hydrogen Oxidation Soil Bioassay
was single laboratory tested as a
potential method for use in hazardous
waste monitoring networks. Preliminary
test results confirmed that the assay was
responsive to aqueous solutions of Hg,
Ag, and Cd, to metal combinations, to
phenolic compounds, and to actual
hazardous waste site material (i.e.,
metals added to waste site material to
-------
Response (% of Control) /jg Hg/g Soil
Determination 10 50 75 100 150 500
1 79 73 63 59 38 8
2 92 68 59 59 41 10
3 85 68 68 59 38 12
4 72 62 54 50 41 11
5 80 80 64 60 38 11
6 79 68 59 55 35 10
7 79 73 63 55 38 12
8 79 68 63 51 35 10
9 79 68 68 63 38 23
10 79 51 55 51 41 8
x 80.3" 67.9b 61.6° 56.2d 38.3e 11.51
SD 5.1 7.5 4.8 4.4 2.2 4.2
CV 6.4 11.0 7.8 7.8 5.7 36.5
"Mean values followed by the same letter are not
provide a more realistic sample matrix).
The bioassay was found to be "rugged" in
the sense that modest procedural
variations did not produce an altered test
result. The method's capability for
precision, expressed as a CV of 7.8
percent, was determined by conducting
10 separate assays using the same con-
centration of mercuric chloride. Within a
mercury concentration range of 10 ppm
to 1 50 ppm, the technique was capable of
distinguishing between concentration
differences of 25 ppm. Limits of reliable
measurement were established at 10
ppm and 750 ppm of mercury. The single
laboratory tested method protocol has, of
course, also has been prepared. Before
this bioassay can be considered for col-
laborative testing, it will be necessary to
conduct portions of the single laboratory
test again using sample material that
more realistically simulates either a
hazardous waste site leachate, or an
analytical fraction of actual hazardous
waste material.
significantly different at the 5% level.
750 7000
2 1
3 2
2 2
3 1
3 1
3 1
3 1
3 1
3 1
1 1
2.6a 1.2a
0.7 0.4
26.9 33.3
Robert D. Rogers is with EG&G Idaho, Inc., Idaho Falls, ID 83415.
W. W. Sutton is the EPA Project Officer (see below).
The complete report, entitled "Single Laboratory Evaluation of the Hydrogen
Oxidation Soil Bioassay, " (Order No. PB 84-2 11 317; Cost: $10.00, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, V A 221 61
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
P.O. Box 15027
Las Vegas, NV 891 14
i
•&U. S. GOVERNMENT PRINTING OFFICE: 1984/759-102/10626
------- |