Reaction Pathways of the Diketonitrile Degradate of Isoxaflutole
J. Agric. Food Chem., Vol. 55, No. 5, 2007 1895
Figure 3. Stoichiometry and three potential fundamental reactions
-
proposed for DKN oxidation by hypochlorite (OCl ).
In a separate experiment, ion trap GC/MS analysis was also
conducted to confirm the presence of dichloroacetonitrile (CHCl2CN,
DCAN). The experimental setup included triplicate samples containing
66.6 mg/L (2.78 µmol total) DKN mixed with 53.3 mg/L (5.57 µmol
total) OCl-, added as Ca(OCl)2, in a total of 15 mL of aqueous solution.
Blanks of the same volume included one sample each of HPLC grade
water, 66.6 mg/L DKN only, and 53.3 mg/L OCl- only. All samples
were prepared in 20 mL screw cap Pyrex test tubes with PTFE-lined
septa. The DKN and OCl- samples were allowed to react for a
minimum of 5 min at room temperature (∼25 °C). Extraction of DCAN
from the samples was accomplished using EPA method 551.1 (22).
After the reaction between DKN and OCl- was complete, all samples
were acidified to pH ∼4.5 with approximately 1 g of a phosphate buffer
containing 1% Na2HPO4 and 99% KH2PO4 (w/w) to prevent hydrolysis
of DCAN. A 3 mL aliquot of methyl-tert-butyl-ether (MTBE) was
added to each sample using a syringe and needle. To minimize
volatilization losses of DCAN and MTBE, the septa were pierced with
the needle to deliver the MTBE. The samples were mixed by hand in
an end-over-end fashion for 4 min. To remove the emulsion that formed
(only observed in the DKN and OCl- samples) after the MTBE
extraction step, approximately 6 g of anhydrous Na2SO4 was added to
each sample, and the samples were mixed end-over-end until all Na2-
SO4 was dissolved. A 100 µL aliquot of the MTBE layer was removed
and placed in chromatography vials with septa-lined screw caps for
GC/MS analysis. GC/MS analysis was performed using the following
temperature program: 32 °C held for 4 min to 175 °C at 50 °C per
minute and hold 175 °C for 2 min for a total time of 9.5 min. The
injector temperature was 175 °C. MS conditions were mass scan range,
74-114, and detection mode, selected ion storage (m/z 82 + 84 +
86). All other GC/MS conditions were as noted previously. Confirma-
tion of DCAN was based on retention times established from a DCAN
standard (98% purity, Acros Organics, Morris Plains, NJ) and the
detection of diagnostic chlorinated ions (m/z 82 + 84 + 86) (23).
Figure 2. HPLC/UV chromatogram showing the presence of a stable
unknown intermediate detected when OCl was the limiting reactant.
-
mTorr pressure as the collision gas to generate product ion spectra for
structural determination. Interpretation of spectra and identification of
the unknown were aided by isotope profiling based on the stable
isotopes of Cl (35Cl/37Cl ) 3:1), S (32S/34S ) 95:4.2), and C (12C/13C)
98.9:1.1) and by the use of IsotopeViewer, Version 1.0, mass spectral
simulation software (Thermo Electron Corp., Waltham, MA). Further
analyses to identify products were done by direct infusion MS and MS/
MS, wherein samples were infused (at 10 µL/min) directly into the
electrospray source. An aqueous mixture containing 100 mg/L DKN
and 20 mg/L OCl- was prepared and allowed to react for several
minutes before direct infusion MS analysis.
Gas Chromatography/Mass Spectrometry. Additional mass spec-
tral analyses were conducted by ion trap gas chromatography/mass
spectrometry (GC/MS) to identify possible intermediates or surmised
product compounds that were not detected under the electrospray
ionization conditions, described previously, or were poorly ionized and
additional confirmation was required. To detect possible products and
intermediates, a 50 mL solution containing a mixture of 50 mg/L DKN
and 10 mg/L OCl- was prepared in water. A 50 mL control solution
was also prepared containing 10 mg/L OCl- only. One purpose of this
experiment was to trap and identify possible volatile compounds not
detected by HPLC/MS/MS; therefore, 50 mL of ethyl acetate was added
immediately after DKN addition to extract any nonpolar volatile
products. Both samples (control and DKN spiked) were shaken
intensively for 2 min in a separatory funnel. An emulsion formed in
the DKN sample, which was removed by centrifugation at 600 rpm.
No emulsion was observed in the control sample. Following liquid-
liquid extraction, a 1 mL aliquot of solution was transferred to a
chromatography vial and sealed with a Teflon-lined septum prior to
analysis.
Initially, mass spectra were obtained for the DKN and control
samples and standards of DKN and BA. DKN and BA standards were
10 mg/L in ethyl acetate. Final spectra containing possible products
were obtained by subtracting the control spectrum from the DKN
sample spectrum. In addition, ions considered unique to the reaction
of DKN and OCl- were identified from a combination of the
background subtracted spectrum and the spectra from the DKN and
BA standards. Using these spectra, reconstructed ion chromatograms
were screened for ions of common mass and retention time, and these
ions were eliminated as possible reaction products. Only unique ions
present in the background subtracted spectrum, and not present in the
standard spectra, were considered reaction products. The GC was a
Varian 3400 with a Saturn 2000 ion trap MS detector (Varian, Harbor
City, CA). An HP-1 (Agilent Technologies, Palo Alto, CA) capillary
fused-silica column (0.33 µm film thickness; 12 m × 0.2 mm i.d.) was
used with He as the carrier gas at a flow rate of 1 mL/min to separate
the analytes. A split/splitless injector was used in splitless mode with
an injector temperature of 200 °C and an injection volume of 1 µL. A
multistep temperature program with a total run time of 27 min was
used to separate the analytes. MS conditions were transfer line
temperature, 250 °C; trap temperature, 250 °C; electron impact energy,
70 eV; electron multiplier voltage, -2000 V; mass scan range, 32-
400; and detection mode, full range.
RESULTS AND DISCUSSION
Stoichiometry and Fundamental Reactions. Results of the
iodometric titration showed that the average molar ratio of OCl-/
DKN was 2.06 ( 0.01. From this result, we concluded that the
reaction stoichiometry was essentially 2 mol of OCl- reacting
with 1 mol of DKN. Assuming that 1 mol of BA was formed
per mol of completely oxidized DKN (17), three fundamental
reactions were presumed (Figure 3). Reaction 1 would produce
Cl- while reactions 2 and 3 reflect the possibility of forming
chlorinated byproducts. Chloride was not observed by direct
infusion MS analysis. Thus, reaction 1 was eliminated from
further consideration, and additional studies focused on the
identification of chlorinated and non-chlorinated products, as
well as an intermediate product observed by Lin et al. (17)
(Figure 2), for discerning the pathway(s) for the reaction
between OCl- and DKN.
Identification of Chlorinated Intermediates. As previously
discussed, Lin et al. (17) observed a stable unknown compound
in HPLC/UV chromatograms under conditions in which OCl-
was the limiting reactant (Figure 2). Initial HPLC/MS analysis
of this intermediate product indicated that its molecular weight
was ∼325 g and that its retention time and UV absorbance
suggested an aromatic compound of intermediate polarity to that
of DKN and BA. Given that Cl- was not released into solution