3. Results and discussion
3.1. Sample preparation
The key objective of our work was to develop a simultaneous
sample preparation strategy for the isolation of 68 analytes from
paper matrix. When attempting to recover compounds ranging
from relatively polar PFC to apolar PAH, medium polarity solvents
acetone, acetonitrile and 2-propanol were selected in the first
place. In order to examine extraction efficiency, solvents were
added to paper matrix fortified with 0.2 mg/kg target compounds
and the extraction took place in the ultrasonic bath. As Fig. 1 shows,
acetone and 2-propanol yielded comparable amounts of spiked
analytes, while there was a higher proportion of lower recoveries
when using acetonitrile thus confirming its lower efficiency. Mono-PAP and PFAPA were, however, not detected at all regardless the
extractant used. Esparza, Moyano, De Boer, Galceran, and Van
Leeuwen (2011) also reported difficulties when recovering PFAPA
from sewage sludge or sediments and concluded that
tetrahydrofuran-water (1:3) was the optimum solvent composition.
Organic solvents with a certain proportion of water frequently
serve for extraction of other PFC as mentioned in Section 1, but in
this application, undesired complications were expected to take
place when injecting such solvent into the GC system. Samples
were, therefore, first extracted by solvent-water 1:1 (v/v) mixture
and subsequently MgSO4 and NaCl were added in order to separate
organic phase from aqueous. Prior to GC injection, residual amount
of water was removed by MgSO4. Consequently, recoveries of all
spiked compounds including mono-PAP and PFAPA improved
significantly when using acetonitrile (Fig. 1). This observation may
be explained as a result of both salting-out effect, which is an
important feature of QuEChERS method, and matrix hydration effect,
which plays a crucial role e.g. when recovering incurred
pesticide residues (Poulsen, Christensen, & Herrmann, 2009).
Rather surprisingly, 2-propanol provided even poorer results in the
LLE process.