United States
Environmental Protection
Agency
Environmental Monitoring and Support
Laboratory
Cincinnati OH 45268
Research and Development
EPA/600/S4-85/073 Dec. 1985
&ERA Project Summary
V-*'''
ry
Validation of Soxhlet Extraction
Procedure for SW-846
Larry C. Michael, M. Arthur Mosely, John W. Mines, and Edo D. Pellizzari
The Soxhlet Extraction Procedure
(Method 3540) was evaluated for the
analysis of hazardous waste compon-
ents in diatomaceous earth filter mater-
ials. Several fortification techniques
were examined to produce a uniform
matrix. A tumbling technique with
methylene chloride spiking was judged
optimal. The accuracy and precision of
the analysis of this fortified matrix were
evaluated. Lower recoveries were deter-
mined for toluene and the xylenes (<
50%); however, for the less volatile
materials, recoveries of 90% or greater
were recorded with an average relative
standard deviation (% BSD) of 7.6%.
Replicate analyses of four industrial
waste mixtures were somewhat less
precise with an RSD of 27%. These
reduced precisions were attributed to a
lack of homogeneity in the sample.
This Project Summary was developed
by EPA's Environmental Monitoring and
Support 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
Adsorption of waste materials on dia-
tomaceous earth is a common practice in
the chemical industry. In the process, a
waste solution is filtered through a bed of
diatomaceous earth to remove the impur-
ities prior to discharge or reuse. The
resultant filter bed contains relatively
large amounts of organic compounds,
typically up to 30 percent of the original
filter weight with individual compounds
ranging up to 10 percent.
The Soxhlet Extraction Procedure has
been used extensively to monitor haz-
ardous waste constituents in diatoma-
ceous earth filter materials, but little
information is available on the accuracy
and precision of these analyses. The
primary objective of this task was to
provide data to demonstrate the validity of
the Soxhlet Extraction Procedure for the
analysis of such filter media.
The validation was conducted in two
phases. Phase 1 involved evaluation of
techniques for fortification of reagent
diatomaceous earth filter material. Twen-
ty organic analytes representative of the
pesticide manufacturing industry were
spiked onto filter material and the produc-
tion of a uniform matrix was assessed.
Utilizing the optimal fortification tech-
nique, the accuracy and precision of the
Soxhlet Extraction Procedure was as-
sessed. In Phase 2, method precision was
further verified by analysis of actual waste
samples from pesticide manufacturers.
Development of Fortification
Techniques
Fortification experiments were con-
ducted to develop a procedure for spiking
diatomaceous earth with analytes that
would produce a homogeneous matrix.
The parameters to be assessed were: (1)
spiking solvent, (2) mixing procedure, and
(3) mixing time. Reproducibility of recov-
ery was the primary measure of homo-
geneity, followed by quantitativeness of
recovery.
Four aliquots of diatomaceous earth
were spiked at the 1 % level (total analyte
mass). In two aliquots, methylene chlor-
ide was the spiking solvent. The samples
were tumbled at 45 rotations per minute
for 0.5 and 24 hours. The third aliquot
was also tumbled for 0.5 hours, but
differed in that methanol was the spiking
solvent. The fourth experiment employed
-------
high-speed blending as the homogeniza-
tion method. Three aliquots of each
homogenized sample were carried
through the Soxhlet extraction. Analyses
were performed by capillary column gas
chrorr.jtography with flame ionization
detection (GC/FID). Recoveries of ana-
lytes were determined by comparison of
concentration found in the extracts with
those in the original spiking solution.
No appreciable difference was shown
between fortification solvents, duration
of homogenization or method of mixing.
Distribution of analytes in diatomaceous
earth was essentially independent of
these parameters. Hence, the most facile
approach, methylene chloride solvent
tumbled for 30 minutes, was used in
subsequent work.
Soxhlet Extraction with
Fortified Filter Materials
Diatomaceous earth samples from two
different suppliers were spiked with each
analyte at three different levels. For
fifteen of the analytes, the levels were
0.1,0.5, and 1 % by weight (w/w); for the
remaining five analytes, concentration
levels were 1, 5, and 10%. Five replicates
for each level and supplier were extracted
and analyzed along with a blank. Method
3540 was followed precisely throughout
extraction and solvent reduction. Internal
standards were added to the concentrated
extract and analyzed by GC/FID.
Soxhlet Extraction with
Pesticides Manufacturing
Waste
The original experimental design in-
cluded analyses of actual pesticide man-
ufacturing filter cakes. When this type of
sample was not available, solid and liquid
wastes from the industry were substi-
tuted. These samples were then incor-
porated into precleaned diatomaceous
earth at a 10% (w/w) level using the
homogenization procedure previously de-
veloped. Aliquots of the mixtures were
dispensed to cellulose extraction thimbles
and extracted by Method 3540. Six repli-
cates of each sample were analyzed by
GC/FID.
p-xylene. These analytes are suspected of
being lost during solvent evaporation. If
one excludes these compounds, the mean
recovery of all other analytes is 90.7
percent. Neither recovery nor % RSD
shows any dependence on fortification
level or diatomaceous earth source.
The analyses of the pesticides waste—
diatomaceous earth mixtures were signif-
icantly less precise than the spiked sam-
ples. The % RSD of the analyses of six
replicate samples ranged from 10 to 46
with a mean of 27%. This is most likely a
result of non-homogeneity of the sample.
In three of the samples, the pesticide
waste was a solid and was, therefore, not
thoroughly distributed throughout the
matrix. However, when a liquid waste
was mixed with diatomaceous earth,
replicate analyses demonstrated high
precision with a mean % RSD of 10
percent.
Results and Discussion
In general, the recoveries of all 20
target analytes for the spiked samples
were quite high and precise. The mean
recovery for all analytes was 83.6 percent
with a mean percent relative standard
deviation of 7.5%. There were distinctly
lower recoveries for toluene, and o- and
-------
L C. Michael. M. A. Mosely, J. W. Mines, and E. D. Pellizzari are with Research
Triangle Institute, Research Triangle Park. NC 27709.
Robert W. Slater, Jr. is the EPA Project Officer (see below).
The complete report, entitled "Validation of Soxhlet Extraction Procedure for
SW-846,"(Order No. PB 86-118 585/AS; Cost: $9.95, 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:
Environmental Monitoring and Support Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S4-85/073
------- |