United States
Environmental Protection
Agency
Environmental Monitoring and
Support Laboratory
Cincinnati OH 45268
Research and Development
EPA/600/S4-85/081 Jan. 1986
&EBA Project Summary
Development of Method for
Semivolatile Organic Priority
Pollutants in Fish
Tina M. Engel, J. Scott Warner, and Thomas A. Pressley
A method has been developed for de-
termining neutral and acidic priority
pollutant compounds in fish tissue. Pri-
ority pollutant compounds are ex-
tracted from the fish tissue during ho-
mogenization of the tissue in
acetonitrile. The acetonitrile extract is
isolated from the tissue after centrifug-
ing and placed in a tumbling bottle con-
taining salted water buffered to pH 4,
methyl t-butyl ether (MTBE), and
petroleum ether. The mixture is equili-
brated by tumbling and a portion of the
organic layer is isolated after drying
over sodium sulfate. The acetonitrile
partitioning step serves to remove a
portion of the extractable fish lipid ma-
terial from the priority pollutant com-
pounds. The organic extract is concen-
trated and the majority of the
remaining lipid material is removed
using a gel permeation chromatogra-
phy (GPC) procedure. The final extract
is concentrated, and priority pollutant
compounds are detected and quanti-
fied in the extract by gas
chromatography-mass spectrometry
(GC-MS). The method yielded a concen-
tration factor of 10 and detection limits
for the neutral and acidic priority pollu-
tant compounds in fish tissue in the
mid to high parts per billion (w/w)
range.
This Project Summary was devel-
oped by EPA's Environmental Monitor-
ing and Support Laboratory, Cincinnati,
OH, to announce the research to de-
velop an analytical method for
semivolatile organic priority pollutants
in fish by capillary column GC-MS. The
research is fully documented in a sepa-
rate report of the same title (see Project
Report ordering information at back).
Introduction
The extraction and cleanup proce-
dures used in the final method were
based on the method entitled "Analysis
of Fish for General Organics by Solvent
Extraction" presented in the EPA publi-
cation entitled "Interim Methods for the
Sampling and Analysis of Priority Pollu-
tants in Sediments and Fish Tissue"
(EPA-600/4-81-055). Analysis proce-
dures used in the final method were
based on EPA Method 625, which speci-
fies the use of GC-MS. Capillary column
GC was used for all analyses. Cleanup
procedures investigated were limited to
the use of acetonitrile partitioning and
GPC. The method was developed to use
a fish sample size of approximately 10
to 25 grams and designed for easy in-
clusion of other semivolatile organic
compounds.
Acetonitrile Partitioning
Two acetonitrile partitioning proce-
dures were examined. The first proce-
dure used separatory funnels for the
partitioning step as specified in the EPA
Interim Method. The second procedure
used tumbling to perform the partition-
ing. Studies were conducted to
determine the partitioning characteris-
tics of the priority pollutant compounds
between the aqueous and organic
phases with and without the presence
of fish lipid material. Organic solvents
investigated included combinations of
hexane, petroleum ether, methylene
chloride, and MTBE. Aceton trile con-
tent of the resultant organic extract was
-------
also determined as well as the fish lipid
removal efficiency observed for each of
the organic solvents.
Partitioning studies demonstrated
that nonpolar solvents such as hexane
or petroleum ether resulted in unac-
ceptable recoveries of polar com-
pounds such as phenols. Use of more
polar solvents such as MTBE resulted in
higher recoveries of polar priority pollu-
tants. Lipid removal studies indicated
that cleanup efficiencies did not vary
significantly regardless of the polarity
of the extraction solvent. Approxi-
mately three-fourths of the lipid mate-
rial remained in the aqueous phase after
partitioning by tumbling. Most of the
acetonitrile must be removed from the
extract prior to introduction onto the
GPC, and acetonitrile is difficult to re-
move using conventional concentration
techniques. In general, the nonpolar ex-
traction solvents contained only a small
percentage of acetonitrile after the par-
titioning step and the polar extraction
solvents contained 20 to 30 percent ace-
tonitrile after the partitioning step. The
optimum situation was observed when
50 percent MTBE in petroleum ether
was used as the extraction solvent. Use
of 50 percent MTBE in petroleum ether
yielded acceptable recoveries of polar
priority pollutant compounds and con-
tained only five percent acetonitrile
after the partitioning step.
Recoveries of neutral compounds
after separatory funnel partitioning
were approximately equivalent to those
observed after tumbling partitioning, in-
dicating that the tumbling procedure
could be substituted for the more com-
monly used separatory funnel partition-
ing technique. Use of the tumbling pro-
cedure greatly reduced the amount of
time and glassware required for the
extraction, avoided the operator-
dependent variability associated with
separatory funnel extractions, and pro-
vided comparable or superior recovery
results.
Gel Permeation Chromatogra-
phy
GPC studies were conducted to evalu-
ate the efficiency of the method for re-
moving fish lipid material from the vari-
ous semivolatile organic priority
pollutant compounds. The GPC column
was packed with BioBeads SX-3 resin,
and 50 percent methylene chloride in
hexane was used as the elution solvent.
Relative elution profiles for fish lipid
material and the semivolatile organic
priority pollutant compounds were
determined.
GPC elution studies indicated that
greater than 99 percent of the lipid ma-
terial could be removed from the fish
extract without loss of even the long-
chain phthalates. Initially, fraction col-
lection was initiated and terminated at
the appropriate time, as indicated by the
GPC elution study results. This sample
collection procedure was suitable dur-
ing the preliminary method evaluation
studies, which used catfish fillets as the
fish matrix. During method validation
and matrix validation studies, it became
apparent that other fish matrixes con-
tained lipid material that eluted later
from the GPC column. Initiation of sam-
ple collection had to be postponed to
avoid excessive amounts of lipid mate-
rial in the final extract, resulting in low-
ered recoveries of long-chain phtha-
lates.
Preliminary Method Evaluation
The draft method was performed as
written as a preliminary evaluation of
the method. Preliminary method evalu-
ation studies were conducted using cat-
fish fillets obtained from a local fish
market. Replicate samples were spiked
with the priority pollutant compounds
at approximately the 2 g/g concentra-
tion level and processed immediately
after spiking. The resultant fish extracts
were analyzed by GC-MS to determine
compound recoveries. In an attempt to
simulate compound incorporation into
the fish tissue, identical samples were
spiked and analyzed after they had been
stored for a 24-hour period to allow in-
corporation of the compounds into the
tissue.
Most of the priority pollutant com-
pounds demonstrated recoveries
greater than 70 percent. Significantly
lower recoveries were observed for the
more polar phenols; recoveries of
20 percent, 53 percent and 28 percent
were obtained for 2,4-dinitrophenol, 2-
methyl-4,6-dinitrophenol, and 4-nitro-
phenol, respectively. Possibly these
components were incorporated into the
fish tissue and degraded by or ineffi-
ciently removed from the tissue.
A general trend of slightly lower re-
coveries was observed in samples that
had been stored for 24 hours after spik-
ing. The trend may have been due to
enzymatic degradation of the com-
pounds or to incomplete extraction of
compounds that became incorporated
into the fish tissue. The experimental
design did not allow for any conclusions
regarding these data, but these data dc
suggest that recoveries observed im
mediately after spiking may be slightly
overstated. Statistically significant de
creases in recoveries upon storage
were observed for only pen
tachlorophenol, bis-(2-ethylhexyli
phthalate, and hexachlorocyclopentadi
ene.
Method Range and Method De-
tection Limit Studies
Method range and method detection
limit studies were conducted to deter-
mine how well the analysis procedure
performed in fish over a range of con-
centrations. Method range and method
detection limit studies were conducted
using a fish matrix consisting of equal
parts of ground catfish, scallops, and
salmon obtained from a local fish mar-
ket. Fish tissue was spiked at concentra-
tions ranging from approximately
20 ng/g to 2 g/g with the priority pollu-
tant compounds, processed, and ana-
lyzed by GC-MS to determine com-
pound recoveries.
A summary of the average recovery
data obtained at each spiking level is
given in Table 1. With the exception of
the more polar phenols, recoveries ob-
tained from the three highest spiking
levels were greater than 50 percent and
repeatable over the spiking range. Virtu-
ally none of the priority pollutant com-
pounds were detected at the low spiking
level and only a few were detected at
the medium-low spiking level.
Matrix Validation Studies
Matrix validation studies were con-
ducted as a comparison to the method
range studies in order to determine if
varying fish matrices might effect the
analytical results obtained. The matrix
validation studies were conducted
using minnows obtained from a local
bait store. The fish tissue was spiked at
the 2 g/g concentrations level with the
priority pollutant compounds, pro-
cessed, and analyzed by GC-MS to de-
termine compound recoveries.
In most cases, recoveries of priority
compounds from the minnow samples
did not vary significantly from those ob-
served from the mixed fish matrix. Re-
covery of di-n-butyl phthalate from the
minnow samples was significantly de-
creased, presumably due to the fact that
collection of the GPC fraction was ini-
tiated later than was done during the
method validation studies.
-------
Table 1. Summary of Results from Method Range Studies
Average Percent Recovery at Given Spike Level (A)
Compound
1 ,2,4-Trichlorobenzene
1,2-Dichlorobenzene
1,3-Dich lorobenzene
1,4-Dichlorobenzene
2,4,6-Trichlorophenol
2,4-Dichlorophenol
2,4-Dimethylphenol
2,4-Dinitrophenol
2,4-Dinitrotoluene
2,6-Dinitrotoluene
2-Chloronaphthalene
2-Chlorophenol
2-Methyl-4,6-Dinitrophenol
2-Nitrophenol
4,4' -ODD
4,4' -DDE
4,4' -DDT
4-Bromophenyl Phenyl Ether
4-Chlorophenyl Phenyl Ether
4-Chloro-3-Methylphenol
4-Nitrophenol
Acenaphthene
Acenaphthylene
Aldrin
Alpha-BHC
Alpha-Chlordane
Anthracene
Benzo (A) Anthracene
Benzo (A) Pyrene
Benzo (B) Fluoranthene
Benzo (G,H,I,) Perylene
Benzo (K) Fluoranthene
Beta-BHC
Bis (2-Chloroethyl) Ether
Bis-2-Chloroethoxymethane
Bis-(2-Chloroisopropyl) Ether
Bis-(2-Ethylhexyl) Phthalate
Butylbenzyl Phthalate
Chrysene
Delta-BHC
Dibenzo IA,H) Anthracene
Dieldrin
Diethyl Phthalate
Dimethyl Phthalate
Di-N-Butyl Phthalate
Di-N-Octyl Phthalate
Endosulfan Sulfate
Endosulfan 1
Endosulfan II
Endrin
Endrin Aldehyde
Fluoranthene
Fluorene
Gamma-BHC
Gamma-Chlordane
Heptachlor
Heptachlor Epoxide
Hexachlorobenzene
Hexachlorobutadiene
Hexachlorocyclopentadiene
Hexachloroethane
Indeno (1,2,3-Cd) Pyrene
Isophorone
Naphthalene
Nitrobenzene
N-Nitrosodiphenylamine
High
63
66
66
67
62
62
171
73
54
58
61
59
36
56
70
65
56
60
60
65
15
61.
64
62
60
57
62
69
61
63
23
64
61
54
57
53
11
79
66
61
91
61
60
50
50
8
63
65
62
73
20
65
61
64
64
58
63
58
61
13
64
777
52
65
60
59
Med-High
79
79
80
86
84
85
104
33
83
82
85
76
54
75
105
96
125
86
86
100
(C)
82
87
87
86
84
89
100
105
104
38
104
93
71
77
66
16
114
114
80
126
79
83
72
49
16
90
102
88
101
31
96
82
81
103
87
90
79
80
13
78
142
70
82
74
74
Medium
51
57
52
55
71
71
157
(C)
63
64
63
57
48
67
88
106
141
76
63
105
(0
78
69
81
67
81
75
90
113
101
44
91
54
59
68
66
20
62
122
65
160
68
97
53
15
26
80
148
70
213
(C)
70
55
72
110
79
87
73
62
(C)
60
121
56
58
74
32
Med-Low
75
74
79
96
47
(C)
(C)
(C)
(C)
(0
81
(C)
(C)
(C)
119
112
(C)
101
80
(C)
85
88
109
77
72
93
85
102
85
(C)
77
(C)
69
101
79
105
458
100
142
263
216
(C)
(C)
(C)
(C)
(C)
{C)
(C)
(C)
44
-------
Table 1. (Continued)
Compound
High
Average Percent Recovery at Given Spike Level (A)
Med-High
Medium
Med-Low
(A) Corrected for the average amount found in the blanks.
(B) Recovery standard added to each fish sample prior to extraction to obtain an indication of method performance.
(C) The compound was not detected.
Lot,
N-Nitrosodipropylamine
Pentachlorophenol
Phenanthrene
Phenol
Pyrene
Trichlorobiphenyl
DlO-Acenaphthene (B)
DW-Fluorene (B)
DW-Pyrene IB)
D3-2,4-Dichlorophenol (B)
54
68
60
38
66
60
57
57
62
57
69
94
86
70
105
85
76
77
85
92
86
775
55
72
94
71
78
83
107
83
(C)
(C)
107
(C)
116
(Cl
80
87
98
(C)
(C)
(C)
5;
{C}
27;
(C)
9t
10t
131
(C)
Tina M. Engel and J. Scott Warner are with Battelle Columbus Laboratories,
Columbus, OH 43201 -2693; the EPA author Thomas A. Pressley (also the EPA
Project Officer, see below) is with the Environmental Monitoring and Support
Laboratory, Cincinnati, OH 45268.
The complete report, entitled "Development of Method for Semivolatile Organic
Priority Pollutants in Fish," (Order No. PB 86-136 058/AS; Cost: $11.95,
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. Envir6fifbental Protection Agency
Cincinnati, OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S4-85/081
0000329 P$
------- |