905R81111
ANALYTICAL METHOD FOR
ANALYZING FISH FOR ORGANIC COMPOUNDS
Organic Lab Section '
Central Regional Laboratory
U.S. Environmental Protection Agency, Region V
536 South Clark Street
Chicago, Illinois 60605
February 13, 1981
-------
SAMPLE PREPARATION
Whole fish samples are collected, wrapped in aluminum foil, immediately frozen
with dry ice and transported frozen to the laboratory for analysis. In the
laboratory, the fish are chopped into small pieces using a meat cleaver, and
all pieces from one or more fish are collected into a sample. The pieces making
up the sample are then ground, using a meat grinder with a fine cutter. The
ground fish tissue is collected in a staiiless steel pan of appropriate size as
it emerges from the meat grinder. The mixture is made homogeneous by manually
mixing the mass with a spatula. Thirty to fifty gram aliquots are taken from
this mass, wrapped with aluminum foil, labelled and extracted or stored in the
freezer as necessary.
SAMPLE EXTRACTION
An aliquot (30 to 50 grams) of the ground fish is transferred to a 250 ml
flask of a Vir Tis Model 45 Homogenizer, 150 ml of 50% (V/V) acetone/ hexane
solution are added, the mixture is homogenized at full speed for two minutes
after which the solvent is decanted off. This process is repeated two more
times with fresh solvent. The extracts are combined dried with sodium sulfate
and concentrated to 5 to 10 ml using a KudernaDanish evaporator. The remaining
solvent is removed by a gentle stream of dry air. The residue is redisolved
in 10 ml of methylene chloride/ cyclopentane (50/50). The extract is then
divided into two equal portions; (E-I and E-II). One fifth (E-I) is used to
determine the fat content of the sample, the other portion is used to perform
the organic analysis.
FAT DETERMINATION
One fifth of the concentrated extract (E-I) is transferred to a labelled pre-
tared aluminum pan. The solvent is allowed to evaporate (in a fume hood) at
room temperature. When the weight becomes constant, it is assumed that all
the solvent has been evaporated and the residue represents the fat contents of
the sample.
SAMPLE CLEAN-UP FOR ORGANIC COMPOUND ANALYSIS
The second half of the extract (E-II) is loaded into an automatic gel permeation
unit (GPC autoprep 1001) equipped with a 25 X 280 mm column packed with 40
grams of Bio-rad SX-3 beads. The instrument is adjusted to discard the first
80 ml through the column and to collect the next 100 ml. A 50/50 (V/V) methylene
chloride/cyclopentane solution is used to chromatograph the sample clean-up.
The collected fraction is evaporated in a Kuderna-Danish evaporator to 5-10
ml. The remaining solvent is removed with a slow steam of dry air and made up
to 10 ml with n-hexane. (Removal of >99% of the methylene chloride is required
for the gas-liquid chromatography electron capture analysis).
-------
ANALYSIS OF THE BASE/NEUTRAL FRACTION (BN-I)
I. Analysis for Pesticides and PCB's
Approximately 5 ul of the concentrate BN-I are injected into a gas chromatograph
equipped with a single inlet to dual column splitter, dual electron capture
detectors and dual 6' X 6 mm OD glass columns packed with: (a) 1.5% SP 2250/1.95%
and (b) 4% SE 30/6% SP 2401, both on 100-120 mesh Supelcon AWDCMS. Operating
conditions of the instrument are: injector temperature 250°C, column temperature
180°C, detector temperature 325°C, 5% methane/95% argon carrier gas with a
flow rate of 40 ml/min through each column.
Interferences appearing as peaks on high background as well as the general
appearance of the chromatogram are used to determine if further "clean-up" is
necessary. If interferences complicate parameter measurement, additional
clean-up is done with a silica gel column. To perform separation of PCB's
from pesticides with Silica Gel Adsorption Chromatography:
1. Deactivation of Silica Gel
Place about 8 g of silica gel per sample (pre-washed with methylene chloride)
in a glass dish or aluminum foil lined pan no deeper than one-half an
inch. Activate at 180°C for 16 hours. Transfer the silica gel to a glass
stoppered bottle. When cool, add distilled water 1.0% by weight. Store
the well sealed bottle in a dessicator prior to use. Silica gel can be
effectively stored in this manner for several days.
2. Preparation of Chromatographic Column:
a) Fill a Chromatographic column with ji-hexane. Pack the lower one-half
inch of the column with anhydrous sodium sulfate.
b) Weigh out 6 g of silica gel and cover with 60 ml ji-hexane.
c) Carefully add the slurry to the column with gentle tapping, making
sure that there are no air bubbles in the column.
d) Turn stopcock to maximum flow rate.
e) When the silica gel has settled, add sodium sulfate to form a
one-half inch layer atop the silica gel.
f) Turn off the stopcock just as the hexane enters the sodium
sulfate layer. The column is now ready for use.
3. Chromatography of Sample
Quantitatively transfer the sample (1 ml of the 10 ml G.P.C. extract in _n-hexane)
concentrate onto the column with the reservoir disconnected. As the lasT of
the sample passes into the sodium sulfate layer, rinse down the internal wall
of the column twice with Ca (0.25 ml portions of ji-hexane used to wash the
sample container) then assemble the reservoir onto the column.
-------
4. Determination of Elution Volumes
A.
The elution volumes for pesticides and PCB's depend on a number of factors
which are difficult to control; these include variations in:
1. Mesh size of the silica gel ;
2. Adsorption properties of the particular batch of silica gel.
3. Polar contaminants in sample extract.
4. Dimensions of chromatographic column.
Hence, the optimum elution volume must be experimentally determined each time
a variable is encountered. It is advisable to chromatograph a set of standards
prior to each batch of samples to monitor any change in elution patterns.
B.'
To determine the elution volumes, add standard mixtures of Aroclors^ and
pesticides to the column and collect 10 ml sequential aliquots. Analyze each
individual eluate by gas chromatography and determine the cut off volume for
both the ji-hexane (Fraction E) and the 20% ether in benzene (Fraction F).
At the CRL we find that all the PCB's are generally eluted in the first 60 ml
of _n-hexane along with hexachlorobenzene, mirex, aldrin, and heptachlor.
QUANTITATIVE DETERMINATION
Measure the amount of eluate and inject 5-10 ul into the gas chromatograph. If
necessary, adjust the volume of eluate to give a response which can be measured
at the same attenuation as that for the standard.
The quantity of each pesticide was determined by comparison with known U.S.
EPA, Quality Assurance Branch in Cincinnati, primary standards. PCB's were
measured by comparison with the known primary standards utilizing the ten
largest peaks. Area measurement was by electronic integration utilizing a
Hewlett-Packard 3350 Data System which also converted these areas to mg/s'kg
of fish weight. Mirex was utilized as an internal standard after its absence
in the fish was assured.
-------
REPORT;
GENERAL PROCEDURE
Chromatography Clean-Up and Analysis of Fish Samples for PCB's
1. Set up of instrument.
2. Determination of "DUMP", "COLLECT" and "WASH" volume. .
3. Determination of recovery of PCB's in spiked samples.
4. Determination of extent of dross contamination between loops and injector.
1. The instrument was turned on for an hour to remove air bubbles. Flow
rate was adjusted to 5 ml/min (Vernier setting » 34), and all loops
were purged with solvent for 4 minutes each. The injector was removed
and washed with solvent, replaced, and subjected to a couple of 10 ml
washings on Loop 00 in the "LOAD" position. The ballast tubes should
be checked for air bubbles and if any is present should be expelled
to prevent leakage. The outsides of the ends of the loops, which
extend to the collectors, were also washed with solvent.
2. The fish sample, spiked with 0.5 ml of a 1242 PCS Standard, was adjusted
to 8 ml and Injected into Loop #1, followed by 8 ml of solvent in
Loop 2. The instrument was reset to Loop 00 and programmed to Collect
the fractions indicated in Table 1. Fractions 1 thru 7 were evaporated
to constant weight and weighed. Fractions 7 thru 15 were analyzed
for PCB's on Channel 4 and compared to a Standard 1242 PCB sample of
500 ng/ml. The weight of fish oil in mg's versus volume eluted and
the weight of PCB's in micrograms versus volume eluted were plotted
(See Graph 1). From this, it was determined that a "DUMP" volume of
80 ml, a "COLLECT" volume of 100 ml and a "WASH" volume of 50 ml
should be employed.
3. To determine the recovery of PCB's by this method, another fish sample
was obtained, divided into two parts, one-half of which was spiked.
The analysis was performed as in Part 2 to confirm the values of
"DUMP", "COLLECT", and "WASH". However, only the oil present in
Fraction 7 was determined. The total weight of PCB's measured as
1242 were determined on Channel 4 (See Table 2). This gave a spike
recovery of 91.3%.
4. To determine the extent of cross contaminations, the instrument was
set up to dump 80 ml in Loop 1, collect 100 ml in Loop 2 and wash the
50 ml into Loop 3. Loop 4 was washed with 10 ml of solvent and 10 ml
of solvent was injected into Loop 5 from which 180 ml was collected
and concentrated to 10 ml. The results on recovery and contamination
in Loop 5 can be seen in Table 3. The total recovery was only 79%
with some PCB's in Fraction 1. This early elution of PCB's may be
explained by the fact that the PCB standard was in acetone, and this
polarity change compared with the methylene chloride/cyclopentane
solvent may have resulted in a faster elution time and subsequent
lowering of the amount recovered. The cross contamination present
from the large amount of PCB's injected on to Loop 1 should be avoided,
and special attention taken on method blanks run through the G.P.C.
Unit so that Contamination levels are well monitored. In this case,
about 1% contamination occurs.
-------
TABLE 1
FRACTION #
.-
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Standard
500 ng/nl
VOLUME
COLLECTED (ml)
20
20
20
10
10
10
10
10
10
10
10
10
10
10
10
VOLUME g.c.
INJECTED
-
5 ul
1 ul
1 ul
2 ul
2 ul
5 ul
5 ul
5 ul
5 ul
5 ul
TOTAL WT
Oil in mgs
0
0
16.8
181.9
126.8
46.0
1.8
TOTAL WT
1242 PCB in ug
-
-
-
-
0.186
9.740
8.251
3.113
f
1.717
1.027
.206
.197
.117
REMARKS
Area factor
the peak at
R.T. 4.75 inns
= 23.2/ng
Total weight of oil recovered = 373.3 mg
Total weight of oil put on column = (5/8 X 0.679 g) = 0.424 g
% Recovery =
-------
TABLE 2
SAMPLE: 00-02-S02
SPIKE = 0.5 ml of 212 mg/ml 1242
| FRACTION #
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Standard
VOLUME
COLLECTED
(ml)
20
20
20
10
10
10
10
10
10
10
10
10
10
10
10
WT. 1242 PCB
IN SPIKED
SAMPLE ug/ml
0.139
.095
5.168
3.614
.989
.351
.248
.165
.075
500 ng/ml
WT.1242 PCB
in SAMPLE
ug/ml
-
-
0.037
3.288
1.621
.664
.265
.093
.039
no 1242 peaks
no 1242 peaks
WEIGHT
of Oil
nrg
-
2.4, 1.0
•
Total weight sample + spike = 10.844 ug/ml
But, 10 ml sample... = 108.44 ug
Sample weight = 60.06 ug
...Weight of spike = 48.37 ug
Amount added = 212 X 5/10 X .5 « 53.0 ug
% Recovery = 91.3%
-------
Table 3
Fraction Volume Collected PCB's as 1242 in
Total yg weight
1
2
3
A
5
"Dump"
"Collect"
"Wash"
Intermediate
10 ml solvent
80 ml
90 ml
50 ml
10 ml
180 ml cone to 10 ml
present but not calculated as
interference
802 yg
28 yg
non-detected
9.7 yg
oil
Recovered 839.7 yg, 1060 injected
% - 79%
-------
f ••'
------- |