United States
                   Environmental Protection
                   Agency
 Atmospheric Research and Exposure
 Assessment Laboratory
 Research Triangle Park, NC 27711
                   Research and Development
 EPA/600/S3-90/010  May 1990
SEPA          Project  Summary
                    Efficiency of  Dioxin
                    Recovery from  Fly Ash
                    SamplesDuring  Extraction  and
                    Cleanup  Process

                   Joe M. Finkel, Ruby H. James, and Kim W. Baughman
                    The data from investigations of the
                  efficiency of dioxin recovery from fly
                  ash  samples   during  sample
                  extraction  and subsequent column
                  cleanup  of  sample  extracts are
                  discussed in this report. Each step of
                  the  extraction  and  the column
                  cleanup procedures was evaluated
                  by using radiolabeled  14C-dioxins as
                  tracers. Sample extracts and eluate
                  fractions  were analyzed by liquid
                  scintillation counting (LSC)  and the
                  results   confirmed   by  gas
                  chromatography/ mass spectrometry
                  (GC/MS).  Recovery  data  of spiked
                  2,3,7,8-tetrachlorodibenzo-p-dioxin-
                  i4c   (TCDD-1"C)   and  octa-
                  chlorodibenzo-p-dioxin-14C  (OCDD-
                  14C) in carbon-free fly ash and fly ash
                  containing from 0.1  to 10% carbon
                  are discussed. The data indicate that
                  the amount of carbon  in the fly ash
                  may  affect the efficiency of dioxin
                  recovery.  Also, procedural details
                  and  analytical  techniques  are
                  extremely  important  in  the
                  quantification of PCDDs and PCDFs.
                    This Project  Summary  was
                  developed by EPA's  Atmospheric
                  Research  and  Exposure Assessment
                  Laboratory, Research  Triangle  Park,
                  NC, 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
                    Currently, several different analytical
                  procedures are being  used for the
                  extraction, cleanup, and analysis of
                  PCDDs and PCDFs (1, 2). RCRA Method
                  8280  and ASME's draft method entitled
 "Analytical Procedures to Assay Stack
 Effluent Samples  and  Residual
 Combustion Products for Polychlorinated
 Dibenzo-p-Dioxins (PCDD)  and
 Polychlorinated Dibenzofurans (PCDF),"
 have been applied to fly ash samples for
 the determination of PCDDs and PCDFs.
 Using these  procedures,  significantly
 different  levels of PCDDs and PCDFs
 have been found in  similar fly  ash
 samples. These differences may be due
 to changes in  combustion parameters or
 inherent errors in  the analysis methods
 such as  nonreproducible extraction  and
 cleanup procedures  Many investigators
 (3, 4, and 5,)  have noted differences in
 the apparent  recovery efficiencies of
 dioxin by different extraction procedures.
  The sorptivity  of different fly  ash
 samples for  organic compounds in
 general has been shown to vary greatly.
 Studies  with  14C-labeled  PAHs as
 radiotracers   demonstrated that  the
 extraction recoveries of PAHs from fly
 ash are  inversely proportional to  the
 number of rings in  the PAH molecule and
 that  aromatic compounds  are more
 strongly  adsorbed  than  aliphatic
 hydrocarbons (6). These results indicate
 that analyses of some fly ash samples for
 PCDDs and PCDFs  may be seriously
 biased if not  corrected for  extraction
 recoveries or  sorptivity losses as  the
 concentration  levels of PCDDs/PCDFs
 decrease, the number of chlorines added
 to the ring structures increases, or the
 nature of the fly ash changes.
  This report  describes  an effort to
 evaluate various extraction techniques to
 remove dioxins from fly ash and to clean-
 up  sample   extracts  by  column
chromatography.  General  chroma-
tographic media and  eluting solvents

-------
were evaluated. Each step in the sample
preparation was evaluated by using 14C-
radiolabeled 2,3,7,8-tetrachlorodibenzo-p-
dioxin and octachlorodibenzo-p-dioxin as
tracers. Radiolabeled dioxins allowed the
analyst to stop and evaluate each step of
the procedure, each extract, and  each
column  eluate   fraction  by liquid
scintillation counting.  To validate the
radioassay, dioxin was  confirmed by gas
chromatography/mass spectrometry.

Technical Approach
  A single source of  fly  ash  that was
generated  from  a  medium-sulfur,
bituminous coal was used in this study.
The carbon content of the fly  ash was
5%. A carbon-free fly ash was  prepared
by heating the native fly ash for 16 hours
in a muffle furnace set  at 750°G. Fly ash
samples were spiked with either 2,3,7,8-
tetrachlorodibenzo-p-dioxin-Ring-UL-14C
(TCDD-HC)  or  octachlorodibenzo-p-
dioxin-Ring-UL-^C (OCDD^C). Also, fly
ash samples  were spiked with carbon
which  came  from  three commercial
sources.
  The samples were  extracted with  a
Soxhlet extraction apparatus.  Fly  ash
samples (10 g) in the extraction thimbles
were spiked with  either TCDD-14C (88.5
ng; 9,145 pCi)  or  OCDD-14C(230 ng;
10,304 pCi). Extraction times were either
24  or 48 hours, and the  extracting
solvents  were 150  mL of methylene
chloride or toluene. Also, some samples
were extracted  sequentially  with
methylene chloride followed by toluene.
  Three types of chromatography media
were  used  in   the  column  cleanup
procedures  (basic alumina, AGIO; silica
gel, Bio-Sil A; and a mixture of activated
carbon, AX-21 grade/Celite 545).
  The selection  of these columns  and
solvents   was   based   on   the
recommendations  of  the   ASME's
analytical procedures to assay stack
effluent samples  for  PCDDs  and PCDFs.
All eluate fractions were  either saved or
solvent exchanged  with toluene or n-
nonane  and submitted  for  liquid
scintillation counting (LSC) and, in some
cases,  were   analyzed   by  gas
chromatography/mass spectrometry.

Results and Discussion
  In this study,  radiolabeled TCpD-^C
and  OCDD-14C  with liquid scintillation
counting  (LSC) were used to  trace the
PCDDs  through each  of the steps
associated  with   an analytical  method
based on ASME's analysis   of stack
effluents from combustion processes.
The quantification  of TCDD-14C  and
OCDD-14C by  LSC allows for lower
detection limits than does quantification
by GC/MS. Also, the use of radiolabeled
PCDD  is more time and  cost effective
because separation from  impurities
(those  not  radiolabeled) is  relatively
unimportant,  especially  in preliminary
screening or evaluation  of  analytical
procedures.

Soxhlet Extraction
  Three types of fly ash samples  were
prepared from a single source of fly ash
for this investigation: native fly ash,  zero-
carbon fly ash, and carbon-spiked fly ash.
Elemental analysis  of the.native  fly ash
indicated the  presence of approximately
5% carbon (wt/wt). Both the native fly ash
and the zero-carbon fly ash were spiked
with various amounts of carbon. Thus, fly
ash samples containing from 0%  to 10%
carbon  (wt/wt) were extracted  for  24
hours with methylene chloride  followed
by an additional 24 hours with toluene or
were extracted with only  one  of  the two
solvents for  24  hours.  The  recovery
efficiencies  for TCDD-14C from  fly ash
spiked  at 8.85 ppb and for  OCDD-14C
from fly ash  spiked at 23.0 ppb  indicate
that the carbon content of the fly ash and
the nature of the carbon may affect the
extraction  efficiencies  of  PCDDs  if
methylene  chloride is used   as the
extracting  solvent  in the  Soxhlet
extraction procedure.   Also,  the  data
demonstrate that the recovery of OCDD-
14C with toluene was less than that of
TCDD-14C.
Column Cleanup
Chromatography Methods
  When dioxins and  furans  are
determined at parts  per trillion  levels
(ppt), concentration of the sample extract
is required  and extensive  sample
preparation ^prpcedures^are needed.,.J,hjs
is due to the increased ratio of interfering
substances to the compounds of interest
in the  concentrated  extract.  Many
environmental  samples,  after  they
undergo an initial extraction or separation
such as  liquid-liquid or  liquid-solid
extraction  may  require one or more
column chromatography  cleanup
procedures.
  The  selected column  cleanup
procedures, as individual columns and as
columns in  sequence, were evaluated
with TCDD-14C  and  OCDD-14C.  The
elution sequences that  were used  for
TCDD-14C  were  based on  the ASME's
procedures. However,  difficulties  were
noted and  several changes were made
when  we  evaluated  the  cleanup
procedures for   OCDD-14C. A  5 mL
standard of TCDD-14C  (88.5  ng) or
OCDD-14C  (230  ng)  was  solvent
exchanged into either toluene (n-nonane
was used with OCDD-14C) or hexane and
was concentrated to 1 mL. The resulting
concentrate  was applied  to  the
appropriate column.
  The amounts of TCDD-14C recovered
in each  fraction for the various cleanup
steps are summarized  in the  report.
Approximately  80% or  more  of the
TCDD-14C  was  recovered  or was
accountable. However, several  potential
problem areas were noted. Over 70% of
the TCDD-14C immediately eluted from
the silica gel column in the first 10 mL of
hexane.  If the analyst is not aware of this,
a major  portion of the dioxin may be lost.
Thus, all  effluent  from the silica gel
column  should =be.collected,, pooled,: and
concentrated to  1  mL.  The data indicate
that  about 90%  of the TCDD-14C was
recovered from the silica gel column.
  Nonpolar polychlorinated compounds
will be  absorbed on the  basic alumina
column;  thus, this column is often used
as  a second  step in  the  cleanup
procedure. Also,  this column allows for
the separation of  PCBs  from dioxins.
Approximately  90% of TCDD-14C was
associated with the  15%  methylene
chloride in hexane  fraction.  This  step
must be monitored closely because the
activity  and  uniformity of the alumina
may vary with each lot and some of the
dioxin may elute  in the 8%  methylene
chloride in hexane fraction.
  If additional cleanup is required on a
sample  extract, the carbon/ced'fe column
may be used.  Chlorinated dibenzo-p-
dioxins  are  initially adsorbed  on this
column and  the  interferences are
removed  by eluting with solvents of
increasing polarity. The PCDDs can then
be backwashed from  the  column by
jnyerting the column  and eluting with
toluene. ThV analyst must monitor this
procedure carefully. Approximately 20%
of TCDD-14C may be lost if a second 4-
mL portion of toluene were not collected.
  After  evaluating  the column cleanup
procedures using TCDD-14C, we made a
few  modifications  in  the  cleanup
procedures.  The report gives details of
the  modifications and  the  effect on
recovery efficiency  of dioxin. Also, we
increased the volume  of toluene from 4
mL to 10 mL as the last step of the AX-21
Carbon/Celite  545 cleanup  procedure.
Because the first two fractions collected
from this column in our initial evaluation
were free of TCDD-14C, the  1  mL  of
hexane  wash was  combined with the 2
mL of 50% methylene  chloride in hexane
wash.

-------
  The total recovery or accountability of
OCDD-14C varied from 40 to  100%. and
>95% of the spiked  OCDD-14C  was
recovered from the silica gel column.
  The cleanup procedures for OCDD-14C
with  the  basic alumina  column  were
identical with  those used for  TCDD-14C.
Although  a  small  amount,  if any, of
OCDD-14C remained on the column, the
elution with  10 mL of 8%  methylene
chloride in hexane did contain  various
amounts of the OCDD-14C. If this fraction
were  discarded,  >20%  of  OCDD-14C
may be lost.
  The cleanup procedures for OCDD-14C
with   the  AX-21   Carbon/Celite  545
retained various amounts of  the  spiked
standard.  Approximately 80%  of the
spiked OCDD-14C was recovered.
....Although we made a few  changes in
the ASME's  cleanup  procedures, the
evaluation  of these  changes with OCDD-
14C revealed additional areas for possible
problems.  A  major  concern  is the
possible loss of dioxins, especially those
that are highly chlorinated,  during the
basic  alumina column cleanup procedure.
  In an attempt to reduce the amount of
OCDD-14C eluted from the basic  alumina
column with  10  mL  of 8%  methylene
chloride in hexane, we changed to 10 mL
of 5% methylene  chloride  in hexane.
About  3% of OCDD-14C eluted with the
5%  methylene  chloride  in hexane
fraction. This  is a problem that may not
be easily resolved.  Certainly this fraction
must  be  analyzed  for  the presence of
trace   amounts of dioxins,  especially
those  that are highly  chlorinated and
when  recovery of surrogate standards are
low. The results  obtained in  one  study
demonstrate another problem that must
be addressed. The technical skills and
experience of  the analyst may bias the
data. We recommend that the procedures
in this report be implemented  by or under
the close  supervision of analysts with
considerable experience  in  trace
analyses.
Gas Chromatography/Mass
Spectrometry
  Confirmation by GC/MS was used to
validate the  data in this  report. The
presence of TCDD  was confirmed  by
GC/MS. Replicate fly ash samples were
carried through the  Soxhlet extraction
procedures.  After the extracts  were
concentrated to 1  mL, some  of the
samples were submitted for LSC  while
others  were  submitted  for  GC/MS
analysis.   LSC   indicated   that
approximately  93%  of  the  labeled
material was recovered. The recovery of
2,3,7,8-TCDD, as determined by GC/MS,
was 91%.

Conclusions
  If  trace  compounds  in complex
environmental  samples   are  to  be
accurately  determined,  appropriate
analytical procedures must  be  selected
and carefully followed by the analyst. The
extraction and cleanup procedures used
in the  analyses of fly  ash  for the
determination of  PCDD  and  PCDF must
be validated  by  the  laboratory and the
individual   analyst  performing  the
analysis. Laboratory control samples
(spiked  matrix samples) must be carried
through the  entire  analytical  method,
which  includes  sample preparations,
cleanup of sample  extracts, and
instrumental analysis. Procedural  details
and analytical techniques,  such as the
cycling of solvent in the Soxhlet extractor,
transfer of the solutions, the  exchange of
solvents,   the packing  of the
chromatographic  media into columns,
and the volume and  composition of the
extracting  solvents,  are extremely
important.  Variable loss of  dioxins can
occur in each of these steps. All extracts,
eluate fractions, and residues generated
during an analysis need to  be saved  for
additional analysis when the recovery of
spiked standards or  surrogate standards
are extremely low in the initial analysis. If
there  is a  question  about  low  recovery
after the final eluate  is analyzed, the
other eluates may have to be analyzed.
With every  set  of  samples that  are
analyzed,  laboratory control  samples,
surrogate  spiking standards, replicate
analyses, and  method blanks must  be
included. The extraction and cleanup
procedures  to  determine dioxins  in  fly
ash samples must be implemented by or
under close supervision of analysts with
experience in trace analyses.
  The carbon content of fly ash and the
nature  of  the carbon may  affect  the
extraction efficiency of dioxin. As  the
carbon content of fly ash increased, the
efficiency  of  spiked TCDD-14C and
OCDD-14C  recovery  decreased.
Modification and control of the extraction
and cleanup procedures improved  the
efficiency of dioxin recover from the fly
ash samples.      .

References
1. Albro,  P.W.;  Parker, C.E. General
approach to the  fractionation  and class
determination  of complex mixtures of
chlorinated  aromatic compounds. J.
Chromatogr. 197: 155-169; 1980.
2. Smith, L.M.; Stalling,  D.L.; Johnson,
J.L.   Determination  of  part-per-trillion
levels of polychlorinated dibenzofurans
and dioxins  in environmental samples.
Anal. Chem. 56: 1830-1842; 1984.

3. Kooke, R.M.M.; Lustenhouwer, J.W.A.;
Olie,  K.;   Hutzinger,  O.  Extraction
efficiencies  of polychlorinated dibenzo-p-
dioxins   and    polychlorinated
dibenzofurans from fly ash. Anal. Chem.
53:461-463; 1981.
4. Karasek,  F.W.; Clement, R.E.; Viau,
A.C. Distribution  of  PCDDs  and other
toxic compounds generated  on fly  ash
particulates in  municipal incinerators. J.
Chromatogr. 239: 173-180; 1982.
5. Stieglitz, L;  Zwick,  G.;  Roth,  W.
Investigation  of  different  treatment
techniques  for PCDD/PCDF  in fly ash.
Chemosphere 15: 1135-1140;  1986.
6. Griest,  W.H.;  Tomkins,  B.A.; Caton,
J.E. EPRI Report 4792, September, 1986.

-------
   Joe M. Finkel,  Ruby  H.  James, and Kim W.  Baughman  are  with  Southern
         Research Institute, Birmingham, AL 35255-5305.
   Jimmy C. Pau is the EPA Prefect Officer (see below).
   The complete report,  entitled "Efficiency of Dioxin Recovery  from Fly Ash
         Samples During Extraction and Cleanup Process," (Order No. PB90-183
         393 /AS; Cost: $15.00 subfect to change) will be available only from:
            National Technical Information Service
            5285 Port Royal Road
            Springfield, VA 22161
            Telephone: 703-487-4650
   The EPA Project Officer can be contacted at:
            Atmospheric Research and Exposure Assessment Laboratory
            U.S. Environmental Protection -Agency
            Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300

EP/V600/S3-90/010

-------