29 / 71
M
astovska
et al
.:
J
ournal of
aoaC I
nternatIonal
v
ol
.
98, n
o
. 2, 2015
503
blank PAH levels were too high to participate in the collaborative
study and to conduct low-level PAH analysis in general.
This was typically due to their location (high environmental
contamination) and/or their laboratory contamination. Some
participants were able to reduce the reagent (procedure) blank
contamination by moving the sample preparation (extraction,
cleanup, and evaporation) away from oil pumps, such as
MS rough pumps. The method requires heating of the used
salts and recommends heating of glassware. Solvents, plastic
material, and equipment may also be sources of PAHs,
including polypropylene centrifugation/extraction tubes. As
one collaborator discovered, simple testing of polypropylene
tubes using ethyl acetate wash/extraction may not reveal PAH
contamination. However, the extraction dynamic during the
actual procedure can release potentially present PAHs into the
extract when the tubes get heated due to the exothermic reaction
caused by addition of MgSO
4
to the water-containing extraction
mixture. For this and other potential contamination reasons, it is
highly important to analyze a reagent blank with every sample
batch.
In addition to the contamination issues, another problem
faced by laboratories less experienced in PAH analysis was
optimization of the evaporation conditions to prevent losses
of volatile analytes, especially naphthalene. Isooctane is used
as a keeper in both evaporation steps, but it did not prevent
significant losses of volatile PAHs in the second evaporation
step in certain laboratories. For this reason, the study direction
team recommended addition of 1–2 mL of ethyl acetate to the
SPE eluent for a better control of the final evaporation process,
which helped in most cases and was added as a recommendation
to the method procedure.
Sixteen laboratories entered the qualification phase, but only
10 of them (listed in the
Acknowledgments
section) completed
the qualification successfully and/or continued in the study. In
many cases, the reason why a participant did not complete the
qualification phase was the availability of resources and not the
ability to qualify for the study.
Collaborators’ Comments
Most of the study participants commented very positively on
the speed and ease of use of the method, especially laboratories
currently analyzing PAHs with much more labor-intensive and
time-consuming methods.
The most frequently reported sources of PAH contamination
were salts (which have to be muffled and stored appropriately).
Some participants had problems with PAH sources in their
laboratories caused by the use of oil pumps in the vicinity of
the space used for the sample preparation. As noted above,
one collaborator discovered PAHs in polypropylene centrifuge
tubes used for practice sample analysis (
Note
: All tubes and
containers used for the test sample storage and preparation were
pretested by the study direction team).
Several collaborators initially had problems with optimization
of the evaporation steps to prevent losses of volatile PAHs.
As noted above, the addition of ethyl acetate in the second
evaporation step resolved this issue in most cases. The majority
of study participants used evaporation with a gentle stream of
nitrogen at room temperature or a maximum of 40°C.
Collaborators noticed differences in the color of extracts of
oyster blank versus fortified samples stored for several months
in a freezer at –20°C. The blank sample produced a dark green
extract, whereas the same blank sample fortified with PAHs
gave a yellow-brown extract. This was not observed for oyster
samples stored for a shorter period of time and/or stored at
–70°C. As discussed below, this observation could be linked to
degradation issues in oysters. Also, the participants noted that
oyster extracts were generally dirtier than shrimp and mussel
extracts, which affected chromatography in some cases.
One collaborator reported the use of a mechanical shaker
instead of hand-shaking. Mechanical shaking is generally
preferred by routine testing laboratories, and this method
modification is acceptable as long as a vigorous and effective
shaking (up and down in the tube) is ensured.
Collaborative Study Results
Tables 3–11 provide the collaborative study results obtained
by the participating laboratories in three blind duplicates (low,
mid, and high fortification level) in shrimp, mussel, and oyster.
Results from Laboratory No. 10 are presented only for mussel
because the Study Directors excluded their oyster and shrimp
data sets due to calibration (standard preparation) issues.
1,7-DMP was used as a homogenization check and
was added to blank mussel and oyster samples at 40 and
80 µg/kg, respectively, during the homogenization step.
The mean concentration value obtained for 1,7-DMP by all
participants (except for sample SFC M4 lost by Laboratory
No. 6 and the result for sample SFC M3 from Laboratory
No. 1, which was removed as an apparent outlier) in all seven
test mussel samples (three blind duplicates and one blank)
was 38.8 µg/kg (RSD = 21.5%,
n
= 68), which corresponds
to mean recovery of 97.0%. In the case of oysters, the mean
concentration value obtained for 1,7-DMP by all participants
(except for Laboratories 1 and 10, for which all 1,7-DMP
results were eliminated as outliers in the Grubbs’ tests applied
Figure 1. Structures of anthracene (Ant), benzo[
a
]anthracene (BaA), and
benzo[
a
]pyrene (BaP).
BaP
Ant
BaA
Figure 1. Structures of anthracene (Ant), benzo[
a
]anthracene
(BaA), and benzo[
a
]pyrene (BaP).