Page 42 - Volume 1, Issue 2, July 2008

Giannopolitis & Kati 94
phosate and AMPA (125, 500 and 1000 ng/
ml) and supplemental pure water to ob-
tain varying water percentages (0.5, 1, 2,
3 or 4%) in a final volume of 10 ml. A refer-
ence solution in pure water was prepared
for each concentration level. Mixing and
preservation of these solutions was in 20-
ml screw-capped clear glass bottles, un-
less described differently.
Glyphosate and AMPA in the methano-
lic and the respective reference (aqueous)
solutions were quantitatively determined
soon after preparation and the next day.
Analysis was performed using cation ex-
change HPLC and fluorescence detec-
tion following post-column derivatization
with hypochloride and o-phthalaldehyde
(OPA), which is an improved version of the
US EPA method 547 (5, 8).
The instrumentation consisted of a sol-
vent delivery system (LC-10ADVP, Shimad-
zu), an auto-injector (SIL-10ADVP, Shimad-
zu), a cation-exchange column 4x150 mm
in the K+ form connected with a guard col-
umn 3x20 mm K+ form (both from Picker-
ing Laboratories), a post-column derivatiz-
er (PCX5200, Pickering Laboratories) and a
fluorescence detector (RF-10AXL, Shimad-
zu). Each solution was first filtered through
a 0.22 μm disposable syringe filter with a
PTFE membrane, into a 2 ml amber boro-
silicate glass vial and then directly inject-
ed into the HPLC system at 10 μl.
All conducted tests were repeated
three times and results from a typical run
of each test are presented here.
Apparent recovery of glyphosate and
AMPA at the three concentration levels,
from methanolic solutions containing up
to 4% water, are summarized in Figure 1.
Data indicate that there is a significant re-
duction in the apparent recovery of both
compounds from the methanolic solu-
tions, compared to that from the respec-
tive aqueous solutions. Depending on the
analyte, the concentration level and the
water content of the methanolic solutions,
apparent recovery was reduced to a level
varying between about 20%and 70%. The
reduction did not seem to correlate to the
concentration level of glyphosate while it
correlated inversely to the concentration
of AMPA, suggesting that solubility alter-
ations in the methanolic solutions cannot
account for the associated reductions in
apparent recovery.
Many researchers involved in trace
analysis of glyphosate and AMPA in wa-
Figure 1. Apparent recovery of glyphosate and AMPA from
methanol mixed with aqueous stock solutions at three ana-
lyte concentration levels. The methanolic solutions (contain-
ing up to 4% water) were analyzed by directly injecting 10 μl
and compared to similarly treated pure water solutions of the
same concentrations (taken as 100% recovery).