Experiences with application of full scan high resolution MS for pesticide residue analysis

Transcription

1 Experiences with application of full scan high resolution MS for pesticide residue analysis Marc Tienstra, Hans Mol, Paul Zomer? EPRW; Cyprus; 25-May-216

2 Current routine practice Chromatography with mass spectrometric detection Pesticide Residue analysis only GC amenable both GC and LC amenable only LC amenable Pesticide Manual: 163 entries Current golden standard: Targeted quantitative measurement by LC-MS/MS and GC-MS/MS Multi-residue methods, ~15-25 analytes/method Emerging: Non-targeted measurement by LC and GC + full scan MS for better coverage of pesticide scope and easier measurement

3 Selectivity using HR-MS Extract signal of exact mass ± x Da (ppm), e.g. Dimethoate [M+H] ± 5 ppm (±.12 Da) m/z T C 5 H 12 NO 3 PS p dimethoate [ Q m/z C 9 H 6 F 3 N 3 O flonicamid Relative Abundance m/z C 19 H 25 ClN 2 OS pyridaben Time (min) Leek 1 ppb, Full scan m/z 135-; Res = 7,

4 Work flow LC-ESI-Q-Orbitrap-MS Sample Extract QuEChERS extraction (215) GC-EI-Q-Orbitrap-MS Quan (usual suspects): XICs, MEW ± 5 ppm, 2 ions Manual verification of all targets in all samples Extensive AQC: Linearity, LOQ recoveries/rsds Identification: peaks 2 ions fully overlap RT ±.1 min from sequence, ion ratio ±3% (rel) Raw data Qual (others): XICs, MEW ± 5 ppm, 2 ions Fully automated by software Detection: RT ±.5 min from database, peaks detected for both ions

5 LC-HRMS: Identification LC-HRMS LC-ESI-MS: soft ionisation no fragments Generation of fragments: 1) needed for identification, 2) improve screening selectivity For optimum detection and identification: full scan acquisition without and with fragmentation in 1 run Non-targeted fragmentation: two options: AIF and vdia

6 Combined Full scan + vdia acquisition FS: m/z - Ϟ: -2 Ϟ: 2-3 Ϟ: 3-4 Ϟ: 3-4 Ϟ: 5- LC-HRMS FS: m/z - Upper & lower m/z cut-off m/z isolation window

7 Combined Full scan + vdia acquisition FS: m/z - Ϟ: -2 Ϟ: 2-3 Ϟ: 3-4 Ϟ: 3-4 Ϟ: 5- LC-HRMS FS: m/z - Upper & lower m/z cut-off m/z isolation window

8 Combined Full scan + vdia acquisition FS: m/z - Ϟ: -2 Ϟ: 2-3 Ϟ: 3-4 Ϟ: 3-4 Ϟ: 5- LC-HRMS FS: m/z - Upper & lower m/z cut-off m/z isolation window

9 Scan speed sufficient (FS + vdia)? LC-HRMS full scan: no fragmentation m/z 135-@7K Fragments of 95-25@35K Fragments of @35K Fragments of @35K Fragments of @35K Fragments of 495-5@35K 978 ms RT: Relative Abundance 5 ppb Clothiadinin (C 6 H 8 ClN 5 O 2 S) in lettuce [M+Na]+ m/z Time (min) vdia m/z fragment C4H3NCl5 m/z RT: Relative Abundance ppb Profenofos (C 11 H 15 BrClO 3 PS)in lettuce Time (min) [M+H]+ m/z vdia m/z Fragment C 6 H 6 81 BrClO 3 PS m/z

10 Quantitative data review LC-HRMS Tracefinder (Thermo Scientific TM software) Review by pesticide (compound view): For each quan pesticide: click through the samples and check asignment/integration of quantifier (main adduct) and qualifier (fragment), adjust when needed Quantifier OK Qualifier OK

11 Quantitative data review LC-HRMS delete

12 Qualitative screening: method set up LC-HRMS Same raw data, different data review High number of target pesticides, low probability of detection Manual verification of all XICs too time consuming Automated pesticide detection by the software (Tracefinder; Thermo Scientific ) Default settings for pesticide detection (quan module used without quantification) Mass extraction window: Time window: Requirement: Output: exact m/z ±5 ppm database RT ±.5 min signal found for pre-set adduct AND fragment ion report of samples showing only pesticides found

13 Screening: data review LC-HRMS Review by sample (sample view): For each sample, click through the pesticides found: Check: 2 peaks present? Matching peak profile/rt? Optional: isotope pattern, additional fragments Prosulfocarb? Reject X

14 Screening: data review LC-HRMS Terbutylazine? (upon quantification: << 1 ppb)

15 GC-HRMS: introduction Features/specifications: Q Exactive GC Resolution m/z 2 (scan speed) 15,; 3,; 6,; 12, ~ 18 Hz 3 Hz m/z 3-3 Acquisition options: Non-targeted (full scan) / Targeted (SIM, MS/MS, data-dependent MS/MS) Mass accuracy: internal < 1 ppm RMS; external < 3 ppm RMS Dissociation in source EI / PCI / NCI

16 Identification GC-HRMS GC-EI-MS: hard ionisation techique, multiple fragments => one acquisition event suffices 13April15_6K_W O_LM _1 #3316 RT: AV: 1 SB: , NL: 1.3E6 T: FTMS + p EI Full m s [5.-5.] Relative Abundance Mw beta-endosulfan QEGC m/z beta-endosulfan NIST Mw 44

17 Scanspeed and XICs vs TIC Lindane C6Cl3H4 1 ppb in leek Pendimethalin C11H14N3O4 1 ppb in leek GC-HRMS TIC RT: Relative Abundance NL: 1.31E m/z= p EI Full m [5.-5 3April Time (min) 6K 5 Hz

18 Method (datahandling) GC-HRMS Thermo Scientific TM Software: XCalibur TraceFinder 3.3 EFS NIST library Pesticide database with t r and exact masses of (fragment) ions On-going (currently 574 pesticides) Method development: Default MEW: ± 5 ppm Assess sensitivity/selectivity of the different ions available for each pesticide CompoundNa RT Q;q1;q2;q3;q4;q5;q6 Q Q q1 q1 q2 q2 q3 q3 q4 q4 q5 q5 2-phenylphenol ;141;115 C12H9O C12H1O C11H C9H Azoxystrobin ;388;329;172;36;43 C2H14N3O C21H14N3O C19H11O3N C1H6O2N C2H14O4N C22H17N3O Bifenthrin ;166;141;165;153 C14H C13H C13H C12H Biphenyl ;153;152;76 C12H C12H C12H Bitertanol ;168;112;(141;152) C12H1O C8H14ON C4H6ON C11H C12H boscalid ;142;342;344;112 C6H3ONCl C6H3ONCl(37) C18H12Cl2N2O C18H12ClCl(37)N2O C5H3NCl Bupirimate ;193;28;316;166 C1H17N4O3S C1H17N C11H18N3O C13H24N4O3S C8H12ON Captan ;149;17;8 C6H C8H7NO C7H7O C6H chlorfenvinphos ;269;323;295;325 C8H6O4Cl2P C8H6O4ClCl(37)P C12H14O4Cl2P C1H1O4Cl2P C12H14O4ClCl(37)P Chlorobenzilate ;139;253;141;111 C13H9OCl C7H4ClO C13H9OClCl(37) C7H4Cl(37)O C6H4Cl Chloropropylate ;139;253;141;111 C13H9OCl C7H4ClO C13H9OClCl(37) C7H4Cl(37)O C6H4Cl Chlorothalonil ;264;268;124;229 C8Cl3Cl(37)N C8Cl4N C8Cl2Cl(37)2N C8N C8N2Cl Chlorpropham ;171;213;154 C6ClH6N C7H6NO2Cl C1H12ClNO C8H9NCl Chlorpyrifos-ethyl ;199;258;313;125 C5Cl3H2NO C5H2Cl2Cl(37)NO C5H3Cl2NO3PS C9H11Cl2NO3PS C2H6O2PS Chlorpyrifos-methyl ;288;125 C7H7Cl2NO3PS C7H7ClCl(37)NO3PS C2H6O2PS coumaphos ;362;21;19;182 C1H7O2ClS C14H16ClO5PS C1H7O3Cl C2H6O3P C9H7O2Cl Cyhalothrin I ;141;28;181 C8H9ClF C8H7F C14H1ON C13H9O C14H11NO Cyhalothrin-lambda ;141;28;181 C8H9ClF C8H7F C14H1ON C13H9O C14H11NO Cypermethrin I ;127;29;181 C7H9Cl C7H8Cl C14H11NO C13H9O C12H Cypermethrin II ;127;29;181 C7H9Cl C7H8Cl C14H11NO C13H9O C12H Cypermethrin III ;127;29;181 C7H9Cl C7H8Cl C14H11NO C13H9O C12H Cypermethrin IV ;127;29;181 C7H9Cl C7H8Cl C14H11NO C13H9O C12H DDD p,p' ;237;199;212;165 C13Cl2H C13H9ClCl(37) C13H8Cl C14H9Cl C13H DDE p,p' ;248;318;316;176;21 C14Cl2H C14H8ClCl(37) C14H8Cl3Cl(37) C14H8Cl C14H C14H7Cl DDT o,p ;237;199;212;165 C13Cl2H C13H9ClCl(37) C13H8Cl C14H9Cl C14H8Cl C13H DDT p,p' ;237;199;212;165 C13Cl2H C13H9ClCl(37) C13H8Cl C14H9Cl C14H8Cl C13H Deltamethrin cis ;174;172;251;255;181;29 C7H9BrBr(81) C7H9Br(81) C7H9Br C7H9Br C7H9Br(81) C13H9O

19 Quantitative data processing method GC-HRMS Data processing by TraceFinder + manual verification of peak assignment/integration Q q1 q2 q3 alpha-hch 1 µg/kg, XICs exact mass ±5ppm

20 Quantitative data processing method GC-HRMS

21 Qualitative screening Two approaches: GC-HRMS 1. Library: match of EI-spectra deconvolution of HR spectra cleaned spectrum Library search (NIST, PEST library) Software: Report if match (SI) > user threshold Analyst: manual review hits 2. Database: RT + 2 exact masses RT ±.5 min XICs ± 5 ppm 2 ions (w/o ratio criterion) Software: Report if signal is found for both ions Analyst: manual review hits Only briefly assessed, preliminary results

22 Qualitative screening: approach-1 GC-HRMS Orange 1 μg/kg: 73% automatically found

23 Qualitative screening GC-HRMS 16111_66 # RT: AV: 4 SB: , NL: 3.79E5 T: FTMS + p EI Full lock ms [5.-5.] Relative Abundance chlorpyrifos present in orange at 1.4 μg/kg m/z Not automatically found by spectral match _6 # RT: AV: 4 SB: , 15.5 NL: 1.84E7 T: FTMS + p EI Full lock ms [5.-5.] Relative Abundance m/z E:\GC-Orbitrap\...\16111_66 1/13/16 3:3:14 RT: SM: 3B TIC Time (min) easily found through approach-2

24 Qualitative screening: prelim. conclusion GC-HRMS Two approaches: 1. Library: match of EI-spectra deconvolution of HR spectra cleaned spectrum Library search (NIST, PEST library) 2. Database: RT + 2 exact masses RT ±.5 min XICs ± 5 ppm 2 ions (w/o ratio criterion) Software: Report if match (SI) > user threshold Analyst: manual review hits + allows screening for analytes not (yet) included in dedicated HR/AM database - low μg/kg levels in complex matrices are challenging Software: Report if signal is found for both ions Analyst: manual review hits + better screening sensitivity - needs database with t r / exact masses

25 Conclusions For pesticide residue analysis, both LC and GC with MS detection is required Full scan HRMS is more straightforward than targeted triple quad MS/MS and better suited for simultaneous detection of high numbers of analytes (retrospective data evaluation possible) LC: multiple scan events required to obtain fragments; vdia advantagous for sensitivity/selectivity; GC: EI => one scan event suffices Orbitrap MS + use of resolving powers of m/z 2 both for LC and GC: Enables use of MEWs of 5 ppm, => high selectivity (similar to MS/MS) Acquisition speed adequate for routine chromatography Analytica Sensitivity/quantitative Chimica Acta performance fit-for-purpose ( 1 µg/kg) Hans G.J. Mol, Marc Tienstra, Paul Zomer More efficient & effective: LC-Q-Orbitrap-HRMS published 215 Evaluation of gas chromatography - electron ionization - full scan high resolution Orbitrap mass spectrometry for pesticide residue analysis 1 injection LC-HRMS, 1 injection GC-HRMS for both quan and qual screening better coverage of pesticides (and more...) for pesticides amenable to both LC-MS and GC-MS: GC-HRMS and LC-HRMS analysis complement each other GC-Q-Orbitrap-HRMS submitted 216

26 Acknowledgement RIKILT: Hans Mol, Paul Zomer Thermo: Dominic Roberts, Cristian Cojocariu, Paul Silcock (Runcorn, UK) Thank you for your attention!