ENVIRONMENTAL SCREENING OF PESTICIDES: A NEW SCIENTIFIC
INFORMATION SYSTEM TO ENABLE ROUTINE ACCURATE MASS NONTARGETED SCREENING
G. Bondoux1*, J.M. Joumier,1, A. Gledhill2, J. Burgess3
1
Waters European Headquarters, Guyancourt, France
2 Waters Corporation, Manchester, UK
3 Waters Corporation, Milford, MA, USA
* Corresponding author email: gerard_c_bondoux@waters.com
Summary
A new scientific information system has been developed to facilitate the routine use of
LC/Tof for the environmental screening of contaminants by non MS experts. The
sophisticated data filtering criteria are designed to minimise the number of false negative
and false positive results. Application to real environmental samples is presented.
Introduction
LC/MS/MS methods using tandem quadrupole instruments are widely employed for
routine quantitative analysis of pesticides and other contaminants in the environment.
These instruments bring the sensitivity and the selectivity required for the determination
of compounds at very low concentration in difficult sample matrices. However, two major
limitations arise from tandem quadrupole methods: the number of compounds that can
be determined in a single injection is limited (linked to the acquisition speed) and
compounds which are not in the MRM method will be definitely missed (even if present
at a high concentration). Consequently, tandem quadrupoles are ideal for targeted
screening, but inadequate when screening for a very large list of compounds in a single
analysis or when investigating the presence of unknown and unexpected molecules.
With progresses in instrumentation, leading to major gains in sensitivity and linearity,
it’s now possible to use high resolution MS analysers, like TOF instruments, for food
safety and environmental analyses. Regarding food safety, the SANCO Directive, which
sets the identification requirements for pesticide residues, now includes the use of high
resolution instruments. The same directive defines also screening detection limits as
corresponding to a maximum false negative rate of 5 % and recommends a maximum
false positive rate of 5 %. Although addressing the food safety methods, the SANCO
guidelines are universally recognised as references in other application fields, including
environment.
TOF-MS provides accurate mass full spectrum information, at an acquisition speed
compatible with Ultra Performance LC. With MSE, accurate mass data is obtained for
both precursor ions and fragment ions in a single injection. The detection of compounds
is based on their exact mass, together with confirmatory information using the isotopic
pattern, fragmentation and retention time. Recently a new scientific information system,
which acquires and processes these complex UPLC-MSE datasets for accurate and
efficient pesticide identification and quantification, was introduced. This new system
streamlines the workflow and utilises all the available data to ensure that all noncompliant samples are detected whilst minimising false positive results.
Instruments and Methods
The LC/MS method was developed on an ACQUITY I-Class UPLC / XEVO G2-S Qtof
from Waters operated in positive electrospray mode. Instrument control, MS data
acquisition and processing were through the recently introduced software, UNIFI. All
data were acquired in MSE mode. In this mode, the collision cell switches very quickly
from low collision energy for obtaining the molecular weight information of the
compounds, to high collision energy, for the fragmentation information. This provides full
unbiased, accurate mass information on both precursors and products in a single
injection and is easily achievable with the fast acquisition speeds of TOF-MS (30
spectra per seconds), even with UPLC peak widths. Data were searched against a
subset of 479 LC-amenable pesticides extracted from the Scientific Library within UNIFI.
Molecular formulae, fragment ions and retention times were included for identification
and confirmation. The list of the compounds found in the sample based on the exact
mass information was rationalized automatically by using the isotopic pattern, fragments
and retention time. Identified compounds were then quantified against calibration curves
from standards. Further investigation for compounds that were not present in the library
was undertaken using structural elucidation tools including elemental composition,
isotopic pattern matching and halogen filtering. Compounds identified through this
process were subsequently added to the library.
Results and Discussion
The Pesticide Screening Application Solution described above has been used to
screen water samples supplied by a collaborator. Calibration standards and quality
control samples were used to access the performance of the solution. The false negative
and false positive rates defined by the directive SANCO/12495/2011 of 5% were met
using the described workflow. Quantification of low level pesticides was achieved and an
unexpected contaminant that was not initially present in the database was also
identified.
The results will be presented and discussed.
References
Method validation and quality control
Procedures for pesticide residues analysis in food and feed
Document N° SANCO/12495/2011
Supersedes Document No. SANCO/10684/2009, Implemented by 01/01/2012