and GCB to slightly retain those pe

and GCB to slightly retain those pesticides, and in fact, it is more
unexpected that acephate and omethoate were not on the list. This
may be due to saturation of the PSA by formic acid in the initial
extract.
When taking both GC and LC results into account and using only
the more reliable method, only a few pesticides gave demonstrably
worse overall recoveries or RSDs by not using the d-SPE step. These
instances can be discerned for each commodity in Fig. 3, which
include dichlorvos and lindane in apple, deltamethrin in lemon,
and azoxystrobin in wheat grain. All of these instances occurred
due to stronger interferences and or matrix effects when cleanup
was not used. Fortunately, the inaccuracies rarely occurred in both
GC and LC for those 13 pesticide analytes analyzed by each orthogonally selective method. The effects shown in Fig. 2 of losses due
to d-SPE cleanup can also be observed in Fig. 3 for acibenzolars-methyl in apple, spinosad in lettuce, and pymetrozine in all 4
matrices.
To enable direct comparisons, Tables 3 and 4 list the results from
the experiment with and without d-SPE cleanup, respectively, for
each pesticide/concentration/matrix combination. Even after >100
sample injections, we did not observe significant differences in the
data analyses or worse instrument performance after this study.
However, d-SPE cleanup is generally known to reduce matrix interferences, matrix effects, and contamination of the MS system. We
gained slightly higher recoveries of the aforementioned analytes
without cleanup, and this should prove useful in future investigations of this sample preparation approach in FI-MS/MS of more
dilute extracts.
3.4. Validation
Mean recoveries and RSDs for all pesticides, matrices, and
spiking levels are shown in Tables 3 and 4. Similar results
were obtained with or without d-SPE cleanup, independent of
using the IS or not in nearly all cases. In general, however,
the IS provided slightly higher recoveries and more precision,
thus they were used in the reported determinations. The limits of quantification were

Hasil (Bahasa Indonesia) 1:

and GCB to slightly retain those pesticides, and in fact, it is moreunexpected that acephate and omethoate were not on the list. Thismay be due to saturation of the PSA by formic acid in the initialextract.When taking both GC and LC results into account and using onlythe more reliable method, only a few pesticides gave demonstrablyworse overall recoveries or RSDs by not using the d-SPE step. Theseinstances can be discerned for each commodity in Fig. 3, whichinclude dichlorvos and lindane in apple, deltamethrin in lemon,and azoxystrobin in wheat grain. All of these instances occurreddue to stronger interferences and or matrix effects when cleanupwas not used. Fortunately, the inaccuracies rarely occurred in bothGC and LC for those 13 pesticide analytes analyzed by each orthogonally selective method. The effects shown in Fig. 2 of losses dueto d-SPE cleanup can also be observed in Fig. 3 for acibenzolars-methyl in apple, spinosad in lettuce, and pymetrozine in all 4matrices.To enable direct comparisons, Tables 3 and 4 list the results fromthe experiment with and without d-SPE cleanup, respectively, foreach pesticide/concentration/matrix combination. Even after >100sample injections, we did not observe significant differences in thedata analyses or worse instrument performance after this study.However, d-SPE cleanup is generally known to reduce matrix interferences, matrix effects, and contamination of the MS system. Wegained slightly higher recoveries of the aforementioned analyteswithout cleanup, and this should prove useful in future investigations of this sample preparation approach in FI-MS/MS of moredilute extracts.3.4. ValidationMean recoveries and RSDs for all pesticides, matrices, andspiking levels are shown in Tables 3 and 4. Similar resultswere obtained with or without d-SPE cleanup, independent ofusing the IS or not in nearly all cases. In general, however,the IS provided slightly higher recoveries and more precision,thus they were used in the reported determinations. The limits of quantification were <5 ng/g for nearly all pesticides in thestudied matrices, which was the lowest calibrated level in theexperiments. Determinations were made for nearly all analytesat the 10 ng/g spiking level, demonstrating the applicability ofthe method to conduct real-world monitoring in food applications.Fig. 3 displays the overall results for each matrix graphically and details the number of pesticide analytes that met the