EB Pesticide Residue Analysis LCGC En

8/18/2019 EB Pesticide Residue Analysis LCGC En

1/38

O

C TO

BER

201

5

Presented in partnership withSponsored content from

PESTICIDERESIDUE ANALYSIS:

TIPS AND TRICKS FOR THE WHOLE WORKFLOW

http://www.thermoscientific.com/en/home.htmlhttp://www.thermoscientific.com/en/home.html


2/382 | October 2015 | LCGC Sponsor’s content

Pesticide residue analysts are challenged to detect, identify, and quantify hundredsof different pesticides from different compound classes at low ng/g concentrationsin a large number of diverse sample types with a fast turnaround, often within 24 hoursof receipt. Analyses must be within budget and in compliance with official methodvalidation, analytical quality guidelines, and accreditation body requirements.

Truly comprehensive monitoring requires analysis by both gas chromatography(GC) and liquid chromatography (LC) techniques, using both targeted and non-targeted approaches. The latter is required to detect illegal usage, because targetedapproaches will not detect pesticides not programmed into the acquisition method.Optimum method performance may also require a separate sample extraction

regimen for different sample types. The analyst’s challenge is to select the optimumcombination of consumable products, instrumentation, and methods to be able todevelop innovative and cost-effective solutions that are fit for purpose.

This e-book, based on a series of four web seminars, provides pesticide residueanalysts with valuable information from Thermo Fisher Scientific about thedevelopment and optimization of methods and workflows for the analysis of pesticideresidues in food. The authors share their knowledge of the critical method controlpoints, assisting analysts in modifying existing methods, optimizing current workflows,

and understanding instrumental and software technologies with the goal of developingmethods to deal with the most complex analytical problems.

INTRODUCTION


3/38

Table of contents

TOC PESTICIDE RESIDUE ANALYSIS: TIPS ANDTRICKS FOR THE WHOLE WORKFLOW

21

GC–MS

Maximizing Efficiency in AnalysisThrough New GC–MS ApproachesRichard Fussell, Dominic Roberts, and Jennifer Massi

31

Data Processing

Processing and Analysis Softwarefor LC–MS/MS and GC–MS/MSKaterina Bousova and Richard Fussell

12

LC–MS

Optimized Workflow for the Analysisof Pesticide Residues Using LiquidChromatography–Mass SpectrometryRichard Fussell, Claudia Martins, and Jennifer Massi

4

Sample Preparation

Sample Preparation Tips and TricksUsing QuEChERS and AcceleratedSolvent ExtractionRichard Fussell, Mike Oliver, Jennifer Massi, and Aaron Kettle


4/38

SPONSORED

Web Seminar:Sample Prep Tips UsingQuEChERS and ASE

Click toview

4 | October 2015 | LCGC Sponsor’s content

Richard Fussell, Mike Oliver, Jennifer Massi, and Aaron Kettle

SAMPLE PREPARATIONTIPS AND TRICKS

USING QUECHERS AND ACCELERATED SOLVENTEXTRACTION

This article begins a four-part tutorial series onworkflow design and optimization for pesticideresidues analysis. The authors provide practicalinformation on food sample preparationwith commonly used techniques includingQuEChERS (Quick, Easy, Cheap, Effective,Rugged, and Safe) and Accelerated SolventExtraction. The authors demonstrate a numberof practical steps to improve pesticide recovery

and extraction efficiency, provide criteria forselecting the appropriate method based onsample type, and discuss ways to develop ormodify methods to meet laboratory-specificchallenges.

Effective sample preparation is a criticalyet challenging step in pesticide residuesanalysis workflows involving the use of gas

chromatography–mass spectrometry (GC–MS)and liquid chromatography–mass spectrometry(LC–MS). Analysts must contend with a diversearray of complex food sample matrices in theeffort to determine an equally diverse andcomplex range of target compounds. Notsurprisingly, sample handling can be a source ofsubstantial experimental error. In the interest ofmaintaining consumer confidence in food safety,

http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%203-24.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%203-24.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%203-24.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%203-24.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%203-24.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%203-24.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%203-24.mp4



DATA PROCESSINGGC–MSLC–MSSAMPLE PREPARATION

pesticide residue laboratories shoulddevelop and routinely use the mostefficient and effective sample preparation

methods possible.In general, food sample matrices are

highly complex and appear in manydifferent physical forms. Many sampletypes require a significant amount oftime-consuming manual processing,extraction, and clean-up before theextracts are ready to be introduced

into a mass spectrometer. To improvequalitative and quantitative performance,the user must try to remove compoundsfrom the sample to minimize matrixeffects that might interfere with theaccuracy of the measurement itself, suchas suppression of the ionization processin LC–MS.

As the demand grows for more sensitive

measurements of a larger number ofpesticide compounds per analysis,historic large-scale extraction methodsthat can create a bottleneck in laboratoryefficiency are being replaced by smaller-scale extraction methods that are morecost effective and more sustainable. Bycontrast to the older methods, which oftenincluded one or more clean-up steps,

some of the newer, faster methods tendto be less selective, and consequently thefinal extracts contain an increased numberand concentration of matrix compounds.Although these modern methods offerproductivity benefits, the high amountof co-extractives, and the high numberof method variations create a dilemma:Which method is the best for the task at

hand? For laboratories facing this dilemma,we focus on two main approaches withthe potential to overcome some of the

challenges: the widely used QuEChERSmethod, and the Accelerated SolventExtraction technique with and withoutautomated evaporation. Each has beenproven to be a practical and effectivesample preparation method for certainsample–target compound combinations.

Sample Processing:e Step Before PrepSignificant errors can be introducedinto the analytical process even beforethe primary sample preparation step.Improperly processed samples canadversely affect the accuracy of the result.Errors introduced at this stage are notcorrected by subsequent steps in the

measurement process.Surprisingly, many analysts go to

great lengths to achieve modest 2–3%RSD improvements in measurementrepeatability but pay relatively littleattention to reducing the much greatererrors that can be introduced duringsample processing. One commonexample occurs with the comminution of

a sample material at room temperature.Ambient conditions give rise to variouschemical and biological reactionsincluding hydrolysis and oxidation,enzymatic activity, and chemicalcomplexation, that can degrade orcomplex the target pesticide in thesample. Lower-temperature comminutionusing liquid nitrogen or dry ice can



DATA PROCESSINGGC–MSLC–MS

minimize these losses (1). In the case ofdry ice, the sample is usually frozen (forexample, placed in a freezer overnight)

and then homogenized in a blender withdry ice to form a dry, homogeneous,and flowable powder. Before analysisthe powder can be divided into testportions and placed in a freezer (typicallyovernight) to allow any excess carbondioxide to dissipate.

Figures 1 and 2 illustrate how cryogenic

processing of sub-samples can improverecovery and homogeneity comparedto room-temperature processing.Unfortunately, analysts often ignore

the errors that can be introduced atthe sample homogenization stage.Such errors are often excluded from

validation experiments and fromroutine recovery, as well as from thecalculation of measurement uncertainty.Laboratories that don’t have on-sitecryogenic capabilities can minimize errorof room-temperature homogenizationby processing the material as quickly aspossible and stirring the homogenate

continuously during withdrawal of sub-samples or test portions. Laboratories canalso improve homogeneity by “double-processing”—in other words,re-processing a large sub-sampleobtained after the initial processing—but this step poses the risk of increaseddegradation of some pesticides.

ExtractionThe goal of the extraction and clean-upstage is to extract analytes with maximumefficiency and with minimal recovery ofmatrix components. Matching the besttechnique to a sample type requiresconsideration of the sample’s physicalproperties, the nature of the targetcompound, and the correct solvent for

the task at hand. Although nearly all ofthe established multiresidue methodsperform equally well for the majority ofpesticides, reduced-scale techniques suchas QuEChERS are becoming increasinglypopular.

QuEChERS overview. QuEChERS(an acronym for “Quick, Easy, Cheap,Effective, Rugged, and Safe”) is

0

50

100

c a p t a n

c a p t a f o l

d i c h l o f l u a n i d

f o l p e t

h e p t e n o p h o s

i s o f e n p h o s

t o l y l f l u a n i d

% S

u r v i v a l R e c o v e r y

Ambient Cryogenic

Figure 1: Cryogenic milling effect on the stability of pesticides.

Fussell et al. J. Agric. Food Chem. 2007, 55, 1062-1070.

Figure 2: Cryogenic milling effect on homogeneity.

Courtesy of Fera, UK


SAMPLE PREPARATION




a streamlined method employingacetonitrile extraction/partitioningand dispersive solid-phase extraction

(dSPE) for the determination of multipleresidues (2). QuEChERS has gained rapidpopularity in pesticide residue analysisbecause it enables laboratories to meetgrowing demands for high samplethroughput, low detection limits, andcost-effective operation. It removesthe need for blending, filtration, and

evaporation steps, which also reducesthe chance of losing volatile analytes.Its use of acetonitrile in the solventextraction stage also helps mitigate theloss of polar pesticides. Because thetechnique is performed at reduced scaleusing less solvent, less equipment, andless laboratory space, the cost of analysisis reduced and environmental impact is

lessened.The original QuEChERS method is

a very simple process (Figure 3). Thehomogenized sample is added to acentrifuge tube, followed by acetonitrile.If the sample is dry, then water should beadded to rehydrate the sample (at least80% water) before addition of acetonitrile.Magnesium sulfate and sodium chloride

are then added (proportion-wise) andmixed in the capped tube by shaking.After centrifugation, the water (lowerlayer) is separated from the acetonitrilephase, an aliquot of which can then betransferred to a QuEChERS clean-uptube containing the appropriate dSPEsorbents, shaken and centrifuged, andan aliquot of the “cleaned-up” sample

is finally transferred to a vial beforeinstrumental analysis. The dSPE clean-upstep is optional. Another time-savings

is the fact that the various reagents canbe purchased conveniently pre-weighed

in sachets or pouches, for example,the Thermo Scientific™ HyperSep™Dispersive SPE materials in metalizedpouches. (See more at: http://www.thermoscientific.com/en/products/quechers-dispersive-spe.html). Variations. There are a number of

variations of QuEChERS, and selectingthe most appropriate option is a matter

2) Dispersive SPEacetonitrile

supernatantcontaining

extractedresidues

Samplematrix

Salts/water

centrifuge

Note: Add sample to the tube, then solvent, then sorbent then mix, to avoidagglomeration

centrifuge

1) Extraction

Figure 3: Steps of extraction and sample clean-upin QuEChERS preparation of a lettuce sample.

Matrix Type ExamplesSorbent Requirements

for Clean-Up

General Matrices•

Apples• Cucumber

• Melon

MgSO4, PSA

Removal of excess water

organic acids, fatty acids, sugars

Fatty Matrices

• Milk

• Cereals

• Fish

MgSO4, PSA, C18

Additional removal of lipids &

sterols

Pigmented Matrices

• Lettuce

• Carrot

• Wine

MgSO4, PSA, C18, GCB

Additional removal of pigments &

sterols

High Pigmented Matrices• Spinach

• Red Peppers

MgSO4, PSA, C18, GCB,

Chlorofiltr™

Additional removal of chlorophyll

Figure 4: Selecting proper sorbent material for clean-up isdetermined by the properties of the sample matrix.


SAMPLE PREPARATION




of evaluating the type of matrix andthe pesticides being measured. Thethree primary variations are currently

supported by commercial accessoriesand consumables (Thermo Scientific,Waltham, MA). The original 2003 method(2) uses NaCl to enhance partitioningof the pesticides into the acetonitrilephase. In an effort to address problemscaused by base-sensitive compounds inthe sample, the AOAC later introduced

a method using NaOAc and acetic acidinstead of NaCl. The European UnionReference Laboratory for Single ResidueMethods has developed a variation usingcitrate buffer salts.

Choosing between the bufferedversions and the original methodis straightforward. If base-sensitivecompounds such as captan,

chlorothanonil, dichofluanid, dicofol,folpet, or tolylfluanid are present, abuffered option will give better recovery.For non-base-sensitive compounds,the recovery differences between themethods are insignificant. In the AOACmethod the presence of acetic acidin samples can reduce the removal ofmatrix co-extractives by PSA, which can

cause elevated baselines in subsequentanalyses.

QuEChERS troubleshooting. Laboratory equipment companies arecontinuously developing new pesticideapplications that demonstrate theoptimal use of QuEChERS in challengingapplications. Our laboratory haspublished a number of technical notes

(3–5) demonstrating the method’seffectiveness with a diverse range offoods including cucumber, grapes, and

lettuce using GC–MS. Although theQuEChERS method is simple, occasionalissues can occur in certain applications—notably diminished analyte recovery.Certain sample types require additionalprocedures. For example, dry-statesamples such as tea or cereal requirehydration with water before extraction;

previously frozen samples must befully thawed (after addition of solvent)and often require extra agitation withan automated shaker; and unstablecompounds prone to breaking downrequire pH adjustment.

Accelerated Solvent ExtractionThe Accelerated Solvent Extraction

technique further expands a laboratory’scapabilities to obtain high-qualityextractions from difficult samples witha low water content and those witha high fat content. The AcceleratedSolvent Extraction method can extractanalytes from the sample and, incertain applications, can also employin-line absorbents to perform clean-up,

minimizing the need for an offline gel-permeation chromatography (GPC) step.

The Thermo Scientific™ Dionex™Accelerated Solvent Extraction systemis available with both single-sample(Thermo Scientific™ Dionex™ ASE™150 Accelerated Solvent Extractionsystem) and 24-sample (Dionex ASE®-350 system) processing capacities. Both

SAMPLE PREPARATION




platforms decrease run times by elevatingthe temperature, thereby increasing theanalyte’s extraction kinetics. The system

also utilizes elevated pressure (1500 psi)to enable extraction solvents to stay inthe liquid state above their atmosphericboiling point. Since the extractionsolvents remain liquid at elevatedtemperature, the extraction efficiency isgreatly improved and the system usessignificantly less solvent than techniques

such as Soxhlet and sonication.Depending on the method, AcceleratedSolvent Extraction reduces extractiontimes to 15–30 min and uses 10–15 mL ofsolvent per sample.

Figure 5 illustrates the six primary stepsused during the Accelerated SolventExtraction technique. The procedureis characterized by one or more static

extraction cycles in which analytes diffusefrom the sample into the extractionsolvent. The diffusion process movestarget analytes down a concentration

gradient. To allow the user to maximizeextraction efficiency, the procedureallows optimization of methods to include

multiple static cycles to exhaustivelyextract samples and produce analyte-richextracts for analysis.

Solvent EvaporationThe Accelerated Solvent Extractiontechnique can perform extraction andcleanup, but isn’t capable of evaporation.

The Thermo Scientific™ Rocket™Evaporator is a centrifugal devicethat evaporates solvent through low-temperature boiling. The system exploitsBoyle’s Law by reducing the pressurebelow atmospheric so the boiling pointof solvents will be lowered. A vacuum-sealed outer chamber containingdeionized water is heated to 40 C to

produce steam for the heat source.Samples are housed in a vacuum-sealedinner chamber that is exposed to the heatsource to evaporate the solvent. Solventvapors are collected and condensedin a cold trap within the instrument sothere is no vapor exposure to the analyst.The Rocket™ Evaporator will evaporatelarge volumes of solvent (up to 400 mL)

to complete dryness or concentratedirection into a GC autosampler vial withfixed endpoint detection.

A wide variety of sample holders isavailable to meet specific requirementsof a method, such as drying orconcentrating an Accelerated SolventExtraction sample. For drying, a toolcalled the Thermo Scientific™ Dionex™

1 min

15-20 min

Total Time

1-2 min

0.5 min

3-5 min

5-9 min

staticcycle

Cell loaded into oven

Fill, heat, equilibrate

Static extraction

Fresh solvent rinse

Nitrogen purge

Filtered extract

Figure 5: Overview of the Accelerated SolventExtraction technique.

SAMPLE PREPARATION




ASE® puck is an adapter that fits into eachof the six rotor positions to hold three ofthe 60-mL ASE® collection vials. The puck

is useful for handling Accelerated SolventExtraction extracts intended to be driedfor reconstitution. For concentrations, theflip-flop style vessel is a modified 60-mL vial with caps on either end. Whenplaced in the extractor, it functions asa conventional collection vial, but afterthe run it can be flipped over, opened,

and fitted with an adaptor that acceptsa GC autosampler vial. These vials canbe placed onto the Dionex™ ASE® puck,enabling the device to concentrate up to18 accelerated solvent extraction samplesinto the autosampler vial.

ConclusionThis overview is intended to help users

take into account several factors thatcould potentially improve analyticalresults when dealing with complexmeasurements of pesticide residues infood samples. An understanding of therequirements and challenges unique toanalytical food testing laboratories canhelp laboratories develop workflows thatdeliver the best possible results in the

most cost-effective, time-efficient mannerfor a wide range of analytical demands.

References

(1) R.J. Fussell, M.T. Hetmanski, R. Macarthur, D. Findlay, F. Smith, Á. Ambrus, and P.J. Brodesser, J. Agric. Food Chem. 55(4), 1062–1070(2007).

(2) M. Anastassiades, S.J. Lehotay, D. Stajnbaher, and F.J. Schenck JAOAC Int. 86(2), 412–431 (2003).

(3) Thermo Fisher Scientif ic Application Note 20549.(4) Thermo Fisher Scientif ic Applicat ion Note ANGSC PESTGRAPES

0700.(5) Thermo Fisher Technical Note 10222.

About the Authors

Richard Fussell focuses on food safety, GC–MS and traceelemental applications in his current role. Before joining ThermoFisher Scientific, he gained more than 35 years of experienceworking for the UK government, including 21 years at the Foodand Environment Research Agency (FERA, formerly called CSL).At FERA, Richard was an advanced research fellow and seniorscientist managing research and collaboration projects focusedon emerging MS techniques for pesticide residues and veterinarydrugs and food.

Since 2010, Mike Oliver has been responsible for the develop-ment and introduction of new innovative technologies to the mar-ketplace, such as the Thermo Scientific Accucore Vanquish UHPLCcolumn and SOLA SPE plates and cartridges. Previously, Mike workedfor nine years for two leading MS vendors, where he was responsiblefor biotechnology sales in the UK and provided application solutionsfor proteomic and metabolic workflows, based on high-resolutionLC–MS platforms. Mike holds a PhD in mass spectrometry and bio-chemistry from the MS Research Unit at the University of Wales inSwansea, UK.

Jennifer Massi is a Market Development Manager for the foodand beverage market in the Chromatography and Mass Spectrometrygroup at Thermo Fisher Scientific Inc. A former genomics scientistwho contributed to the sequencing of the human genome throughthe Human Genome Project, she now works on connecting laborato-ries with food safety testing workflows. Jennifer completed her B.Sc.,Honors, in Biochemistry and Molecular Biology at the University ofCalifornia at Santa Cruz and received an M.B.A., Honors, from SaintMary’s College of California.

Aaron Kettle is a Product Manager for the Thermo Scientificsample prep systems, including Accelerated Solvent Extraction, solid-phase extraction, and the Rocket evaporator. He works for key indus-try experts to develop improvements in these technologies so thatlaboratory and regulatory leaders receive the best possible samplepreparation solution. Over the last seven years, Aaron has held bothsales and marketing positions at Thermo Fisher Scientific (previouslyDionex), and before that, he was a Commissioned Officer in theUS Navy and served as a Technical Director and Biochemist for theDepartment of Defense Forensic Laboratory. Aaron holds an MSc inToxicology from the University of Michigan’s School of Public Healthand an MBA from the Lake Forest Graduate School of Management.

SAMPLE PREPARATION


11/38

http://www.thermoscientific.com/quan-transformed


12/38

SPONSORED

Web Seminar:Workflow Guide for LC–MSin Pesticide Analysis

Click toview


Richard Fussell, Claudia Martins, and Jennifer Massi

OPTIMIZED WORKFLOW FORTHE ANALYSIS OF PESTICIDERESIDUES USING LIQUIDCHROMATOGRAPHY–MASSSPECTROMETRY

This second installment of a four-articleseries of tutorials on pesticide residueanalyses focuses on workflows using liquidchromatography–mass spectrometry (LC–MS). During the past decade, significanttechnological improvements in both samplepreparation and instrumentation havecontributed to dramatic growth in the useof LC–MS in pesticide residues analyses.

Indeed, the technique is now essential for thedetermination of modern pesticides, most ofwhich are more amenable to direct analysis byLC–MS compared to GC–MS. To aid pesticideresidue laboratories in the selection of the bestanalytical method for their specific challenges,this article focuses on practical suggestions foroptimized chromatographic separations andmass detection for a range of pesticides in a

variety of sample matrices.

Rapid technological developments in liquidchromatography mass spectrometry (LC–MS)have brought significant benefit to the analysisof pesticides in food products. With a combinedtotal of around 30 years of involvement inthis field, the authors have employed a widerange of LC–MS approaches, beginning

http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%20Analysis%204-29.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%20Analysis%204-29.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%20Analysis%204-29.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%20Analysis%204-29.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%20Analysis%204-29.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%20Analysis%204-29.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_Pesticide%20Residue%20Analysis%204-29.mp4



DATA PROCESSINGGC–MSSAMPLE PREPARATION LC–MS

with atmospheric pressure chemicalionization with single-quadrupole massspectrometers capable of determining

only two to three compounds in asingle analytical run. By the year 2002,electrospray triple-quadrupole massspectrometers were readily availablewith the ability to quantify around50 pesticides at 10–50 μg/Kg. Soonthereafter, continuing advances in MSalong with the introduction of reduced-

scale sample preparation techniquessuch as QuEChERS opened the door tothe analysis of a much larger number ofpesticides in a greater range of sampletypes.

Today, state-of-the-art triple-quadrupole MS systems such as theThermo Scientific™ TSQ Endura™ triple-quadrupole mass spectrometer and the

Thermo Scientific™ TSQ Quantiva™triple-quadrupole mass spectrometer(Thermo Fisher Scientific, Waltham,MA), enable analysts to detect, identify,and quantify several hundred targetedpesticides at low concentrations (10μg/Kg and below) in complex matricesin a single run. Alternatively, massspectrometers offering high resolution

and mass accuracy, such as the ThermoScientific™ Orbitrap™ -based systems,enable laboratories to combinescreening of unexpected residues withthe identification and quantification oftargeted analytes into a single analysis,dramatically increasing the capabilitiesof LC–MS. The high-resolution approachmeets today’s demand for increased

scope (more analytes), screening for non-targeted compounds and the capabilityfor retrospective analysis should new

information emerge in the future.When evaluating an analytical technique,laboratories take into account a numberof key factors, including methodsensitivity, robustness, and samplethroughput. The challenge is to meet theestablished maximum residue levels, ortolerance values, in the most efficient and

cost-effective manner possible. Selectingthe most appropriate technique fromthe array of sophisticated technologiescan be challenging. This article providespractical insights to help analysts makebetter-informed decisions on methoddevelopment and instrument selection.

Sample Preparation and Separation

The first installment of this series focusedin depth on the QuEChERS approach,which is a generic sample preparationprotocol that has been successfullyvalidated for a large number of pesticidesin a wide range of sample matrices.Laboratory efficiency gains come fromhigher sample throughput with lessdemand for manual labor, laboratory

space, consumables, and, as a result,less waste. QuEChERS has become apopular method for pesticide residueanalysis. One disadvantage of QuEChERScan be the high concentration of matrixcomponents remaining in the extractafter the crude clean-up providedby dispersive solid-phase extraction(SPE). Alternatively, in-line automated



DATA PROCESSINGGC–MSSAMPLE PREPARATION

cleanup approaches such as turbulentflow chromatography (TFC) have beenreported in the literature (1) (Figure 1).

TFC works by combining a column (0.5or 1.0 mm, packed with large particles,typically 50–60 μm) with high flow rates(higher than 1 mL/min), creating a veryhigh linear velocity inside the turbulentflow column. Molecules with low molecularweight will diffuse faster than moleculeswith high molecular weight, and therefore

a separation between matrix componentsand analytes can be achieved. TFC seemsto be more efficient at removing proteinsbased on their size than restricted-accessmedia (RAM) or SPE.

The optimization of the various on-lineextraction steps is crucial. Parameterssuch as mobile phase composition, flow

rates, and extraction time windows willaffect recovery and extraction efficiencyin general. In pesticide analysis TFC mayrequire some minimal prepreparation forcertain sample types. A liquid samplesuch as orange juice may require thatthe chemist adjust the pH, dilute withmethanol, vortex mix, and centrifuge

before TFC separation and MS analysis.Solid samples require homogenizationbefore the dilution, vortex, andcentrifugation steps. Using TFC-LC–MS/MS, researchers have observed recovery

Figure 1: Turbulent flow chromatography has recently emerged as a choice for environmental and food samples.

AnalytesApplication Field

SampleTFC column

Loading Flow RateInjection Volume

DetectionSensitivity

PFOSEnvironmental Analysis

River Water 50 x 1.0 mm, 50µm C181 mL/min

1mL APPI-MS

5.35 ng/L (LOD)

Anti-InfectivesEnvironmental Analysis

Waste Water50 x 1.0 mm, 50µm C18 XL

3 mL/min1mL

ESI-MS/MS15-53 ng/L (LOD)

Enrofloxacin

Ciprofloxacin

Food Analysis

Edible Tissues

50 x 1.0 mm, 50µm C18

Cyclone

5 mL/min

20µL

ESI-MS/MS

25 µg/Kg (LOQ)

QuinolonesFood Analysis

Honey50 x 0.5 mm, 60µm Cyclone

1.5 mL/min160µL

ESI-MS/MS5 µg/Kg (MLOQ)

Veterinary drugsFood Analysis

Milk

50 x 0.5 mm, 60µm Cyclone –Cyclone P (connected in

tandem)

1.5 mL/min50µL

ESI-MS/MS0.1-5.2 µg/L

O.Núñez et al. J Chrom A2012, 1228, 298-323.

LC–MS




rates of 80 to 110% for more than 48pesticides of different classes, mostly at10 μg/Kg (2). The main benefit of more

effective removal of matrix componentsis increased analyte response due toreduced ion suppression and decreasedcontamination of the MS system, resultingin a decrease in the required frequencyof system maintenance. Changing andcleaning the TSQ Endura and TSQQuantiva ion transfer systems when

required is simple and doesn’t requirethe analyst to break the vacuum, thusincreasing system up-time.

Mass Spectrometer SelectionTriple-quadrupole MS can be used toscreen, identify, and quantify a widerange of pesticide compounds of currentinterest. The use of a triple-quadrupole

mass detector offers several advantages,including high sensitivity, compatibilitywith ultrahigh-performance LC, and directcompatibility with extracts prepared bystreamlined methods such as QuEChERS.

In selecting a triple quadrupole system,laboratories should take the natureand volume of their pesticide residueworkload into account. For laboratories in

need of a workhorse system engineeredfor continuous operation, the TSQ Endurais a good choice. By contrast, the TSQQuantiva offers laboratories a higher levelof sensitivity and overall performance forchallenging, complex samples. Figure 2 compares the features and specificationsof these systems.

Triple-quadrupole technology is

the established gold standard and isused routinely on a daily basis in manypesticide residue laboratories around

the world. One drawback is the needfor time-consuming optimization of theconditions to optimize the responsefor each pesticide included in theacquisition method. An alternativeto manual optimization is the use ofa comprehensive database of SRMtransitions included in Thermo Scientific™

TraceFinder™ software.A nominal mass triple-quadrupolesystem configuration is essentially aseries of mass filters. In selected reactionmonitoring mode (SRM), the conditionsof the first quadrupole (Q1) are set toallow precursor ions at a selected m/z ratio to pass through into a collision cell(Q2), where the ions are collided with

gas molecules and hence fragmentedto form product ions. The product ionsof selected m/z pass through the thirdquadrupole (Q3) to the detector. If thefirst mass selective quadrupole in theseries is not selective enough, ionsof masses similar to that of the targetanalyte can be fragmented too, resultingin isobaric interference and uncertainty

regarding the correct identification of theanalyte. In SRM experiments of heavilycontaminated matrices, the analyst canincrease selectivity in a way only possiblewhen using hyperbolic quadrupole rods.An important feature of both the TSQEndura and TSQ Quantiva is hyperbolicquadrupole rods. The use of highly-selective reaction monitoring mode

LC–MS




(H-SRM) increases the selectivity of theanalysis even in highly complex samples,as shown in Figure 3.

Even with H-SRM the triple quadrupoleonly detects those compounds includedin the acquisition method. It is almostinevitable that some pesticides will bepresent in some samples but not included

in the acquisition method and hencenot detected. For this reason there is anincreasing interest in screening approachesusing “full scan” high resolution accuratemass (HRAM) technologies.

Again, mass resolving power is a keyselection criterion. In general, the higherthe mass resolving power used, the betterthe mass accuracy and the possibility

to differentiate between the analyte ofinterest and matrix compounds withmasses close to that of the analyte ofinterest. Laboratories requiring systemsfor both targeted and non-targetedanalysis can choose from a wide range ofMS systems including the Orbitrap line,the Q Exactive Plus, and the Q Exactive

HF systems. Most recently, the Q ExactiveFocus was introduced for pesticideresidue analysis. This system has thesame sub-1-ppm mass accuracy of otherQ Exactive systems across a range of 50to 2000 m/z , with resolution of 70,000FWHM at 200 m/z .

The European Reference Laboratory forPesticides in Fruit and Vegetables recently

Figure 2: A technical comparison of the TSQ Endura and TSQ Quantiva systems.

LC–MS




demonstrated the use of the Q Exactivesystem for the analysis of pesticideresidues in fruits and vegetables. Anassessment of the effect of mass resolvingpower on the number of pesticides thatcould be detected at different analyteconcentrations with a mass tolerance

of 5 ppm showed that higher resolvingpower becomes increasingly importantas the concentration of the pesticidedecreases toward the 10-ng/g level (3). Asthe resolving power was increased from17,500 and 70,000 the percentage oftotal number of identified pesticides alsoincreased. The increased resolution helpsin terms of linearity, both at low and high

concentration levels, and also improvesthe method reproducibility (Figure 4a).In tea, a more complex matrix comparedto tomato, this difference is more evident(Figure 4b). A resolving power of 70,000FWHM was required to detect 95% ofthe target pesticides at the lowest (10-

ppb) concentration. This underlines someof the tradeoffs associated with methodselection. A targeted analysis might offerextra sensitivity, but at the expense of theability to conduct non-targeted screening.According to the European guidelinesfor Method Validation and QualityControl Procedures for Pesticide ResiduesAnalysis in Food & Feed (4), high-

Figure 3: Results from a 2012 study published in the Journal of Chromatography A showing background-free

peaks even in highly complex samples. (1)

O.Núñez et al. J Chrom A ., 2012, 1249, 164-180.

m/z

•

Remove background

interferences

• Smooth baselines

• Improve % CV

• Increase signal-to-noise

LC–MS




Figure 4: The importance of high mass resolving powers used in conventional and complex sample types (tomato and tea).

Data presented at EPRW 2014

Data presented at EPRW 2014

a)

b)

LC–MS




resolution accurate mass identificationcriteria should include at least two ions,one of which should ideally be the quasi-

molecular ion. Mass accuracy should bebelow 5 ppm, and at least one fragmention should be detected. Resolution of atleast 20,000 FWHM is specified, althoughno specific corresponding mass has beenspecified. The data obtained using theOrbitrap technology is compliant withthese criteria, and typically exceeds

the requirements, providing increasedconfidence in the identification andquantification of residues.

ConclusionThis article has discussed the criticalimportance of adapting an analyticalmethodology for maximum performancein the demanding applications of

pesticide residue analysis. We haveemphasized that a wide range ofinstrumental technologies andconfigurations can complement oneanother in laboratories faced with agrowing demand to perform efficient,cost-effective analytical measurementsof an ever-growing range of pesticidecompounds.

Reference

(1) O. Núñez, H. Gallart-Ayala, C.P.B. Mart ins, and P. Lucci, J. Chro-matogr. A 1228, 298–323 (2012).

(2) L. Hollosi, K. Mittendorf, and H.Z. Senyuva Chromatographia 75,

(23-24), 1377–1393 (2012).(3) Ł. Rajski, M. Gomez-Ramos, Amadeo R. Fernandez-Alba J. Chro-matogr. A, 1360, 119–27 (2014).

(4) SANCO/12571/2013. Method Validation and Quality Control Pro-cedures for Pesticides Residues Analysis in Food and Feed: http://ec.europa.eu/food/plant/pesticides/guidance_documents/docs/qual-control_en.pdf(last accessed August 2015).

About the Authors

Richard Fussell focuses on food safety, GC–MS and trace

elemental applications in his current role. Before joining ThermoFisher Scientific, he gained more than 35 years of experienceworking for the UK government, including 21 years at the Foodand Environment Research Agency (FERA, formerly called CSL).At FERA, Richard was an advanced research fellow and seniorscientist managing research and collaboration projects focusedon emerging MS techniques for pesticide residues and veterinarydrugs and food.

Claudia Martins is the environmental and food safety applica-tions program manager with Thermo Fisher Scientific. She has adiploma in biochemistry and a PhD in analytical chemistry from theUniversity of Manchester in the United Kingdom. During her many years at Thermo Fisher Scientifi c she has held various marketing and

sale support roles in which she has focused on assisting customers inmethod development for environmental and food analysis.

Jennifer Massi is a Market Development Manager for the foodand beverage market in the Chromatography and Mass Spectrometrygroup at Thermo Fisher Scientific Inc. A former genomics scientistwho contributed to the sequencing of the human genome throughthe Human Genome Project, she now works on connecting laborato-ries with food safety testing workflows. Jennifer completed her B.Sc.,Honors, in Biochemistry and Molecular Biology at the University ofCalifornia at Santa Cruz and received an M.B.A., Honors, from SaintMary’s College of California.

LC–MS

http://www.thermoscientific.com/QExactiveGC


20/38

http://www.thermoscientific.com/QExactiveGC


21/38

SPONSORED

Web Seminar: MaximizingGC–MS Efficiency

Click toview


Richard Fussell, Dominic Roberts, and Jennifer Massi

MAXIMIZING EFFICIENCYIN ANALYSIS THROUGH

NEW GC–MS APPROACHESThis article reviews the limitations andadvantages of different approaches to thedevelopment of effective and complete gaschromatography-mass spectrometry (GC-

MS) workflows for pesticide residues in food.It addresses some of the generic-but-criticalaspects of GC methods, including injectorparameters and column configurations, andalso discusses the latest developments in GCtriple-quadrupole MS and Thermo ScientificGC Orbitrap MS technologies.

One of the many challenges in the analysis of

pesticide residues in food is the determinationof a large number of different pesticides in awide range of diverse matrices. In pesticideapplications, chromatographers encountersimple matrices with relatively high water contentsuch as fruits, but they must also be preparedto analyze more complex matrices such asherbs, spices, and tea. Regardless of the matrix,analytical methods must accurately detect and

quantify pesticides at low concentrations whilealso offering high throughput, fast sampleturnaround, and low analysis costs.

Liquid chromatography-MS (LC–MS) and LC–MS/MS technologies have progressed rapidlyin recent years, prompting some people toquestion whether there is still a need for GC–MS in pesticide residues applications. However,some compounds do not give a good response

http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%203_Pesticide%20Webinar%20Series.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%203_Pesticide%20Webinar%20Series.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%203_Pesticide%20Webinar%20Series.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%203_Pesticide%20Webinar%20Series.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%203_Pesticide%20Webinar%20Series.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%203_Pesticide%20Webinar%20Series.mp4



DATA PROCESSINGLC–MSSAMPLE PREPARATION GC–MS

using LC–MS with atmospheric ionization.In fact, GC–MS offers better detectabilitythan LC–MS for a number of compounds.

Therefore, GC–MS capability remainsessential to achieve a comprehensivemultianalyte method for pesticide residueanalysis.

One of the most important recentdevelopments in GC–MS technology wasthe introduction of GC triple-quadrupoleMS, which provides better selectivity and

higher signal-to-noise (S/N) performancethan is possible with single-quadrupoleMS. A quadrupole is a nominal mass filterthat allows ions of a specified mass-to-charge ratio (m/z ) to pass through. In atriple-quadrupole system, ions producedby ionization in the source initially passinto the first quadrupole, Q1. Systemparameters are set to allow the allow the

precursor ions of preselected m/z to passthrough into the second quadrupole, Q2.In Q2, the precursor ions are fragmentedby collision with neutral gas molecules andwith collision energy applied. This is theprocess underpinning selected reactionmonitoring (SRM), which is the basisof GC–MS/MS analysis. Some extractscontain thousands of different compounds

and it is often the case that a matrix peakand the pesticide of interest will share thesame nominal mass and co-elute at thesame retention time, passing through Q1together. Using selective ion monitoring(SIM), or single-quadrupole MS, all of theco-eluting ions are monitored effectively atthe end of Q1, and that results in chemicalinterference with the pesticide signal.

By contrast, in triple-quadrupole MS,we use SRM. The pesticide and the matrixwill have different chemical structures,

so they will typically produce differentproduct ions when fragmented in Q2under controlled conditions. Only theunique product ions from the pesticide ofinterest pass through Q3 and reach thedetector. The result is a cleaner signal,which leads to improved selectivity andS/N compared to single-quadrupole MS.

Injection MethodsAlthough fundamental to any GC–MSmethod, sample injection is the stepwhere most problems with analysisby GC–MS originate. The two mostcommonly used injection techniques aresplitless, and programmed temperaturevaporization (PTV) (Figure 1). Each

has advantages and disadvantages inpesticide residue analysis, and carefuloptimization of the selected method isrequired for best results.

Splitless injection is the simplest andprobably most widely used method,

Figure 1: Schematic of a PTV injector system.

• Minimal thermal mass forfast cooling and heating

• Injection volumes from nano literup to large volume

• Cold injection technique

• Clean step possibility for keepingthe liner inert

• Multiple injection modes

OVENcolumn

Liner

Cooling by fan Heater element

Inlet CarrierSeptum Purge

Split line

Slide courtesy of Thermo Fisher Scientific



DATA PROCESSINGLC–MSSAMPLE PREPARATION

offering excellent sensitivity andrepeatability for low sample volumes.During sample injection, the split

line is closed so the entire sample istransferred to the head of the column.In the solvent focusing mode the initialoven temperature is set 10–20 °C belowthe solvent boiling point. The solventand analytes are refocused into a tighterband at the head of the column. At apredetermined time, set to allow transfer

of analytes to the column, the split vent isopened to “clean” the injector. The useof a pressure purge during injection helpsto compress the solvent vapor in the liner,and can allow slightly increased sampleinjection volumes and help to protectpesticides prone to thermal degradation,especially in the “hot” splitless mode.

Although less commonly used in

pesticide residue analysis due to limitedsensitivity, the technique of a splitinjection can be used to overcomematrix problems. During split injection, aproportion of the injected sample is splitor diverted to waste. In split injections,the sample is swept by the carrier gasthrough the liner and split betweenthe column and split-line in the ratio

of at least 1:5 and often much higher.This “shoot-and-dilute” approach helpskeep the inlet and GC liner clean fora longer period of time and increasescolumn lifetime. Split injections yieldvery reproducible results and good,sustainable peak shapes. The residencetime of the analyte in the inlet is reduced,and that can lead to less thermal

degradation of certain problematicpesticides. The main disadvantage is thatless analyte is injected into the column

so the detector response is decreasedproportionately.

In PTV injections the sample isinjected into the liner at a relativelylow temperature with precisely time-controlled evaporation of solvent (towaste); this allows large volume injection(typically 5–10 µL).The liner design is

selected to help with retention of theinjected sample in the liner and toincrease the surface area for evaporation.After the evaporation step, a smallvolume of sample (typically less than 1µL) is left in the liner. The injector porttemperature is then increased rapidlyup to 300 °C to transfer the analytesto the head of the analytical column.

Because of the precise control of the PTVtemperature profile, the PTV techniquecan help minimize analyte discriminationduring injection. This results in improvedrecovery of thermally labile pesticides andfewer adverse effects from non-volatilespresent in the sample during injection.These features make PTV especially well-suited for trace pesticide analysis.

GC Liner SelectionLiner selection is an important facet ofoptimizing a robust GC–MS method.In pesticide residue analysis, the widelyused QuEChERS extraction methodtypically yields extracts in acetonitrilesolvent. Using acetonitrile is not ideal forGC as it often involves injecting a polar

GC–MS




solvent onto a mid-polar column, whichwill often result in poor peak shapes.Acetonitrile also has a high solvent

expansion coefficient, which thereforelimits the volume that can be injected.When selecting a GC liner for such anapplication, it is important to considerfactors such as: the type of injection to beused; the internal diameter and volumeof the liner, the liner construction, theliner packing (if required), deactivation,

and other features. Acetonitrile has alarge expansion volume, and the linervolume must therefore be sufficient toaccommodate the sample in gaseousform. If the internal diameter is toonarrow, the sample will expand beyondthe liner’s capacity, resulting in samplelosses, peak tailing, and poor peak-areareproducibility. If the diameter is too

wide, however, a large dead volume slowsdown sample transfer time and, again,contributes to peak tailing. Various liner types are suitable for

pesticide residue analysis by GC–MS. Insplit injection, liners that are open-endedenable the split flow to pass across thebottom of the liner. By contrast, splitlessinjection requires a tapered line to funnel

the sample onto the GC column.The use of baffled liners is recommend

for PTV as they create a turbulent flowin the liner, which aids sample mixing;this improves reproducibility. Linersare available empty (no packing), withinternal protrusions (e.g., asymmetric orbaffled), or packed with materials suchas Carbofrit® or deactivated wool (glass

or quartz for example). The advantageof injecting a sample onto the wool ina packed liner is that it increases the

surface area for complete vaporization ofthe sample before it reaches the column.This improved vaporization improvesreproducibility and lower boiling pointdiscrimination. The wool can alsoclean the injection needle and reducethe chance of any particulate materialreaching the column. However, the wool

is also a potential source of active sitesand regular injector maintenance shouldbe performed. For the wide range ofavailable Thermo Scientific consumablessee www.thermoscientific.com/gcconsumables

Avoiding System ContaminationThe analysis of dirty extracts will

eventually give rise to a build-up of matrixwhich will cause counter-productive shiftsin peak shape and retention time. Thissituation highlights the need for regularinjector maintenance to ensure goodsystem performance.

The use of column backflushing is anoption to improve method robustness.Co-injected matrix components can

cause matrix enhancement by protectinganalytes from absorption or degradationon active sites. Some analysts perform“equilibration” injections of matrix todeactivate the inlet system prior toanalysis.

However, sample matrix componentscan also create new active sites in theinlet and column head, compromising

GC–MS




chromatographic performance.Backflushing, a technique that involvesreversing the carrier gas flow at aspecified time after injection to backflushthe retention gap and liner, can help toprevent less volatile matrix components

from reaching the GC column. It isimportant not to reverse the gas flowbefore the last analyte of interest hasentered the column. A T-junction, whichconnects the analytical column and theretention gap, and a pressure valve,can be operated to reverse the gasflow through the retention gap. Gasflow through the analytical column is

maintained at all times. Chromatogramsin Figure 2 illustrate the retention ofanalytes and removal of high-boilingpoint components when the column wasbackflushed 10 min after sample injection.For challenging samples, backflushing

will provide better retention timeprecision and better spectrum quality andquantification accuracy. It will also reduceanalysis time, increase column lifetime,and minimize the need for systemmaintenance.

Increasing Productivity State-of-the-art detection technology,

GC–MS

Figure 2: Effect of backflushing during analysis of a pear extract in ethyl acetate. Without backflushing (top), peaks from severalhigh-volume components are present. Those peaks disappear with backflushing 10 min post injection (middle) and when astandard is injected (bottom).

• No BKFL

• BKFL ON 10 min afterinjection of sample

• BKFL ON 10 min afterinjection of standard

Full scan data acquisition – Trace GC w PTV-BKF – 30 m TR-Pesticides, 5 m pre-column 0.53

mm ID

Area of high boiling matrix




available in systems such as the ThermoScientific™ TSQ™ 8000 Evo GC–MS/MS,can play an important role in productivityfor the routine laboratory with highsample throughput. Shorter GC runtimes, greater selectivity, and the ability

to capture more data in a single runhelp laboratories to run more samplesin less time. The TSQ 8000 Evo featuresenhanced technologies that can reduceanalysis times without compromisingdata quality. In particular, a fast collisioncell provides high speed SRM transitionsand rapid data acquisition; this enablesacquisition and analysis of more

pesticides in a single analytical run. Suchfeatures make it possible to take fulladvantage of fast triple quadrupole GC–MS in pesticide residue analysis.

Recently, the use of the TSQ 8000Evo GC–MS/MS (coupled to a ThermoScientific™ TRACE™ 1310 GC andThermo Scientific™ TriPlus RSH™autosampler configured for liquid

injection) with shorter, narrow borecolumns (20 m × 0.18 mm × 0.18 µmcolumn) allowed faster GC gradientswhile shortening the GC run timefrom 37 min to 11 min—more than a3-fold improvement in speed for the

determination of 144 pesticide compoundsin baby food, as shown in Figure 3.

This system not only is able to achievefaster analysis times and measure morepesticides; it also offers more transitionsfor better selectivity (especially forcomplex matrices) and enables full-scan and SRM acquisition in the sameexperiment. In the analysis of baby foods

the narrow-bore column provided narrowpeaks and reduced column bleed, bothof which improved the S/N performance.The fast acquisition rates providedsufficient data points, even for narrowpeaks, to give excellent quantification fora large number of compounds in a shortanalytical run time.

Dedicated SRM method development

Figure 3: Column length and diameter can be reduced to decrease analysis times without impacting analytical data.

Full scan144 pesticides in baby food @ 0.2 mg/kgTG-5 SILMS, 30m × 0.25 mm × 0.25 mGC run time: ~37 min

Full Scan144 pesticides in baby food @ 0.2 mg/kgTG-5 SILMS, 20m × 0.18 mm × 0.18 mGC run time:




software, Thermo Scientific™ AutoSRM,reduces the time and effort required toadd new compounds into the acquisition

method. The software guides the userthrough the steps from precursor ionselection to product ion selectionand to collision energy optimization,resulting in an optimized method withminimum operator input. The softwarealso automates the creation of samplesequences and data layouts.

Timed SRM is an alternative to classicalsegmented SRM for laboratories seekingconsistent and rapid acquisition of largeamounts of data. Timed SRM can reducecomplexity through automated, optimizedtargeting of a particular compound. Oncethe user has entered the retention timeand the time required to acquire thepeak, timed SRM automatically optimizes

the instrument duty cycle and sensitivitywithout the need for spreadsheets, dwelltime calculations or segment breaks.This ensures that compound detectionis optimized for maximum sensitivity andallows more compounds to be added tothe method without compromising theresponse for analytes. Also, it increasesresilience to small retention time shifts

and prevents wasteful, unnecessary MSscanning for compounds during timeswhen they are not eluting.

The TSQ 8000 Evo GC–MS/MS isadept at analyzing an increased numberof pesticides in a single run, therebyreducing the number of injectionsrequired to cover all of the targetpesticides with the required number of

SRM transitions for identification. Thisincreases the instrument’s availability forother work and reduces maintenance

requirements. In the analysis referencedearlier, the system also showed improvedselectivity for different matrices withno loss in peak area response whenincreasing the number of SRM transitionsfrom two to six. Linearity was alsoconsistent whether 2 or 6 transitions wereused. With this capability, a chemist can

select the most intense interference-freetransition for quantitation without havingto perform a second injection to monitordifferent transitions.

Many of these productivity-enhancingperformance benefits are embedded inthe Thermo Scientific™ TSQ™ 8000 EvoPesticide Analyzer solution. This systemis preloaded with pesticide-specific

GC–MS/MS acquisition and processingmethods, analytical column consumables,tutorials for method development, and acompound database of more than 600compounds with retention times and pre-optimized SRMs. The automated PesticideAnalyzer system is suitable for lessexperienced users but can be customizedto meet a wide range of advanced user

needs.

GC and High-Resolution MSRecent advances in high-resolutionaccurate mass spectrometry coupled toGC create potential for new approachesin the analysis of GC-amenable pesticides.For example, the Thermo Scientific™Q Exactive™ GC Orbitrap™ GC–MS/

GC–MS




MS system, operating in full-scan mode,allows the detection, identification, and

quantification of an unlimited number oftarget pesticides and screening for non-targeted compounds in a single run. Thesystem covers a broader analytical scope byacquiring and automatically processing full-scan data, enabling the user to substantiallyincrease the scope of the method withouta corresponding increase in analysis time.

Another major advantage compared totriple-quadrupole systems is the possibilityof retrospective data evaluation. Theuser can revisit older data to search fornew compounds that were not on theoriginal target compound list or knownor recognized at the time of acquisition.The Q Exactive GC is capable of selectivedetection of targeted analytes in complex

matrices with high resolving power up to120,000 (FWHM m/z 200) and excellentmass accuracy (typically less than 1 ppm).

The sensitivity and specificity of the

Q Exactive GC for routine screeningof pesticides in complex matrices has

been demonstrated by the analysis ofacetonitrile QuEChERS extracts of wheat,horse feed, and leeks containing 55different pesticides at concentrationsbetween 0.5 and 10 ppb. The goal wasto establish the lowest level at whicheach pesticide could be detected in suchcomplex matrices.

Automated sample injection wasachieved using the TriPlus RSHautosampler. Separation was achievedusing a TRACE 1310 GC with a ThermoScientific™ TraceGOLD™ TG-5SilMS 15m × 0.25 mm, 0.25 µm column with a fastGC method to achieve 11 min analysistime. For detection, a Q Exactive GCOrbitrap mass spectrometer in electron

impact full scan mode was used. Data wasacquired at 60 K resolving power (FWHMm/z 200) and at 15, 30, and 120 K. Dataprocessing was performed using Thermo

Figure 4: Lowest detected concentrations of our 55 pesticides in (a) wheat and (b) horse feed.

Lowest detected std

Half MRL

MRL

Lowest detected std

Half MRL

GC–MS




Scientific™ TraceFinder™ 3.3 software.The selected identification points for

a positive screen were a retention time

match within 20 s of the pesticide standardand accurate mass with less than 2 ppmmass error. The presence of fragment ions,isotopic pattern, a NIST match, and ionratios was used for identification and tominimize both false detects (to reduce theneed for operator intervention) and falsenegatives (increased confidence in results).

Results demonstrating increased selectivityand sensitivity in wheat and horse feed aresummarized in Figure 4, which shows thelowest detected concentrations for each ofthe 55 pesticides in both of these complexsamples.

ConclusionChemists using state-of-the-art GC–MS and

GC–MS/MS systems can perform efficientand robust pesticide analysis. Carefulmethod optimization is essential, especiallywith respect to injection parameters.Routine pesticide analysis with the TSQ8000 Evo system reduces analytical costswhile increasing sensitivity, analysis speed,and the ease of database management. Bymaximizing the use of available dwell time

through timed SRM, users can maximizesensitivity in data acquisition to increasethe range of pesticides measured. High

resolution MS offers efficiency gains byallowing simultaneous targeted analysis withthe quantitative, sensitivity, and selectivity

performance of triple-quadrupole MS. Italso provides unrivalled capabilities forscreening a higher number of pesticides withexcellent selectivity provided by unparalleledresolving power and mass accuracy.

About the Authors and Contributors

Richard Fussell focuses on food safety, GC–MS and trace

elemental applications in his current role. Before joining ThermoFisher Scientific, he gained more than 35 years of experienceworking for the UK government, including 21 years at the Foodand Environment Research Agency (FERA, formerly called CSL).At FERA, Richard was an advanced research fellow and seniorscientist managing research and collaboration projects focusedon emerging MS techniques for pesticide residues and veterinary

drugs and food.

Dominic Roberts is a senior applications specialist in chromatog-raphy and mass spectrometry at Thermo Fisher Scientific. Dominic isan experienced analytical chemist and mass spectrometrist, with morethan 15 years of experience using state-of-the-art analytical systems.Dominic has worked in both regulatory and industrial environmentsgaining strong hands-on experience with GC–MS and LC–MS systems,

using them to investigate a variety of chemicals in food, biological,and environmental samples. The majority of his experience comesfrom 10 years working at FERA as an analytical chemist, focusing onmethod development using LC–MS and GC–MS systems to detectfood contaminants. Dominic currently works with GC–MS technologies,including high resolution accurate mass MS to develop and implementanalytical methodologies for the detection and quantification of tracechemicals, including pesticides in food and environmental samples.

Jennifer Massi is a Market Development Manager for the foodand beverage market in the Chromatography and Mass Spectrometrygroup at Thermo Fisher Scientific Inc. A former genomics scientistwho contributed to the sequencing of the human genome throughthe Human Genome Project, she now works on connecting laborato-

ries with food safety testing workflows. Jennifer completed her B.Sc.,Honors, in Biochemistry and Molecular Biology at the University ofCalifornia at Santa Cruz and received an M.B.A., Honors, from SaintMary’s College of California.

GC–MS


30/38

http://www.thermoscientific.com/TSQ8000Evo


31/38

SPONSORED

Web Seminar: DataProcessing and AnalysisSoftware for LC–MS/MSand GC–MS/MSClick to

view


Katerina Bousova and Richard Fussell

PROCESSING ANDANALYSIS SOFTWARE

FOR LC–MS/MS ANDGC–MS/MS

In LC–MS/MS and GC–MS/MS analysis,processing and evaluation of the raw massspectrometry data can be the most time-consuming part of the analytical process.These steps also create opportunities toachieve substantial improvements in sampleturnaround times. This article reviewsenhanced software capabilities now availableto increase the productivity and efficiency of

pesticide residues analysis workflows.

State-of-the-art instruments combining liquidchromatography (LC) or gas chromatography(GC) with tandem mass spectrometry (MS/MS) are powerful and widely used tools forthe analysis of pesticide residues in food. Inhighly regulated, high-throughput laboratoryenvironments, scientists are under pressure

to process a growing number of more diversesamples as rapidly and economically as possible.One of the major time commitments with LC–MS/MS or GC–MS/MS is the task of processingand evaluating the large volume of raw MS dataproduced. Fortunately, automated solutions arenow available to streamline this process for gainsin efficiency, productivity, data security, andregulatory compliance.

http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%204_Pesticide%20Webinar%20Series.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%204_Pesticide%20Webinar%20Series.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%204_Pesticide%20Webinar%20Series.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%204_Pesticide%20Webinar%20Series.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%204_Pesticide%20Webinar%20Series.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%204_Pesticide%20Webinar%20Series.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%204_Pesticide%20Webinar%20Series.mp4http://www.learnpharmascience.com/tablet-apps/issue13/LCGC_EBook_Part%204_Pesticide%20Webinar%20Series.mp4



GC–MSLC–MSSAMPLE PREPARATION DATA PROCESSING

One such software solution is theThermo Scientific™ TraceFinder™software package (Thermo Fisher

Scientific, Waltham, MA), which offersenhanced capabilities that address thoseneeds in both targeted quantitation andscreening of non-targeted compounds.Using this platform as an example ofcurrent capabilities, this article reviewskey software and data processingrequirements in LC–MS/MS and GC–

MS/MS pesticide residue workflows,emphasizing new functionalities thattransform a once time-consuming processinto an opportunity to improve overalllaboratory productivity and performance.

In addition to automated data processing,analysts rely on the system softwareto control instruments, to streamlinethe development and optimization of

methods, to alert them to analyticalfailures, and to create custom reports thathelp them deliver on the expectations oftheir customers and auditors. In LC–MS/MS or GC–MS/MS work, chemists find itespecially advantageous when softwaregives them the flexibility to work withmultiple instrument platforms and tocontrol how they manage and display the

complex data sets involved. Designedfor routine quantitative analysis of largesample batches as well as for screening fornon-target compounds, the TraceFindersoftware features new security, reporting,and compliance features with potentialto significantly increase efficiency inenvironmental and food safety laboratories.

Data Acquisition ModeAs stated, TraceFinder is ideal forprocessing large batches of sample

data in routine quantitative analysis. Thesystem uses batch templates to simplifyset-up and control of the data acquisitionmode. Analysts can populate thesetemplates with sample lists built eitherfrom scratch or using a batch acquisitionwizard tool within the software. Also,the system automatically populates

the template with all required qualityassurance (QA) and quality control (QC)samples and calibration standards.

TraceFinder’s new security featurescreate an unambiguous audit trail. Ifactivated, the security system allowsmanagers to assign very specific rolesand access rights to specific users from anextensive list of preset permissions. The

system helps regulate and document whohas access to start the batch, who hasaccess to the results, or who can changethe method. Controls can be set torequire users to input the reasons for anychanges and to acquire their electronicsignatures. Although only one aspect ofa total laboratory compliance program,these new security and audit features

offer substantial support for regulatedlaboratories.

Method DevelopmentFigures 1a–c present TraceFinder’sdisplay of key method developmentfunctions such as method selection,detection settings, calibration options,and QA/QC. TraceFinder is available in



GC–MSLC–MSSAMPLE PREPARATION

four industry-specific variants, includingan environmental and food safetyversion with many specialized features

for pesticide residues work. The platformincludes several databases of compoundswith data acquired in selective reactionmonitoring (SRM) or with high resolutionaccurate mass (HRAM) mass spectrometry(extracted ion chromatogram, or XIC).The GC–MS/MS database containsmore than 1000 compounds (SRM

experiments) including pesticides,environmental, contaminants and PCBs.The LC–MS/MS database contains morethan 600 compounds (SRM experiments)covering mainly pesticides, mycotoxins,and veterinary medicines while theHRAM database contains around 1700compounds (XIC experiments) includingfragments. To create a new method, the

user selects one of five different methodprocedures from a Create Master Methoddialogue box. For explanatory purposes,the most convenient of these tools isthe option to Select Compounds fromthe Compound Database (Figure 2). Inthe latter case, the user selects targetcompounds from the database, selectsthe correct instrumental method, and

provides a raw data file of previouslymeasured standards. Next, the analystdefines the desired detection settings,type of calibration level, QA/QC criteria,and other parameters. The last stepis to define the report to be createdautomatically after data processing. Inall, the entire process takes only a fewminutes.Figure 1a–1c: TraceFinder’s tools for method selection,

calibration, and QA/QC.

Figure 1a: Detection Settings

Figure 1b: Calibration Options

Figure 1c: Criteria for Quality Assurance and Quality Control

Change in the settingscan be easily applied forall peaks in method orall peaks in compound

ExternalInternal

EstimatedStandard Addition

DATA PROCESSING




For calibration calculation, thesoftware allows users to select external,internal standard calibration, estimated,or standard addition. Each targetcompound can have its own initial

calibration setting. Modifications arepossible during set-up or in acquisitionmode after viewing results from anactual calibration batch. Changing is alsopossible during data review, creatingoptions to compare and choose betweencalibration options.

For QA/QC, the system allows theoperator to set limits and ranges that

enable automated review of data andresults. Copy-and-paste functions andother shortcuts are useful for copying gridvalues from column to column or fromone master method to another.

Data Review and AnalysisThe TraceFinder data review screendisplays the target compound list, thesample set, and compound details suchas quantitative peaks, confirming ions, ionoverlay, and calibration data. The datareview window is customizable—the sizeor position of individual display panes can

Figure 2: Compound database view of data acquired in SRM mode, including precursor mass,product mass, polarity, and parameters for confirming peaks.

1. Select target compounds from CDB

2. Choose the correct instrumental method

3. Associate a raw data file4. Define the detection settings, type of

calibration, criteria for QA/QC

5. Define reports

DATA PROCESSING




Figure 3a–e: Versatile data review features

Figure 3b: Data flagging system Figure 3c: View by compound—a compound across all samples

Figure 3d: Data review—ion overlayFigure 3e: View by sample —all compounds in a sample

Figure 3a: Easy data review

COMPOUNDS

QUANTITATIVE PEAK IONS ION OVERLAY CALIBRATION

SAMPLE SET WITH FLAGS

Sample- orCompound-

Centric

Data Flag Customization gives the laboratory

the ability to create meaningful visualization of

the rules developed for data review

DATA PROCESSING




be altered, and manual changes can bemade to parameters such as integrationor calibration. The software automatically

recalculates results reflecting thesechanges.

The user can view a full data overviewon one screen, or can customize thedisplay to hide individual fields that aren’tof interest. Another option is to display allthe peaks in a given sample. Dependingon the size of the user’s computer

monitor, it can require a number of screenpages to display that many peaks. Thesystem uses multiple dockable panels tofacilitate a holistic view of multiple peakdisplays and other larger data sets, ineffect creating multiple monitors on oneworkstation for easier review.

Because many analysts spend so muchof their time reviewing data, innovations

that add efficiency to the process cantranslate into significant productivitygains. One such feature in TraceFinderis its novel flagging system, whichidentifies and highlights batch resultsthat don’t match specified requirements.User-customized flags in the data resultspane change colors based on criteriaset in the master method. Additionally,

the sample peaks are enclosed in a redbox when ion ratios are not within thedefined ranges. Users can edit flag coloror shape, change or delete certain errorconditions or change their priority, reviseflag rules, and modify the color of theerror indicator icon. The flagging systemlinks to the user-security feature which,when activated, restricts access to the

ability to edit flag features. Figures 3a–e provide representative views of some ofthe features described above.

Preparing Methods forNon-Targeted ScreeningAs stated above, TraceFinder softwarenot only supports routine quantitationof target compounds but also is usefulfor the automated identification ofnon-target compounds. The most

important step in preparing a screeningmethod is the determination ofsettings for compound identificationand confirmation. The other primaryconsideration is selection of one or moreappropriate compound databases forlibrary searching. TraceFinder searchesselected libraries and reports the highestscoring matches back to the user as a

percentage value. The unknown’s likelyidentity is narrowed down and furtherverified through a process of comparingmultiple forms of measurement data—retention times, isotopic patternmatching, or library segment matching,for example. High-resolution MSinstruments controlled by TraceFindersoftware can be used to accomplish both

screening and quantitation in a singleworkflow, as shown in Figures 4a–b.

There are many high-resolution MSlibraries available, including spectrallibraries containing data from morethan 600 pesticide compounds. Theselibraries are continuously updated andcan be accessed at http://planetorbitrap.com/compoundlibrary. In the library

DATA PROCESSING




Figure 4a–b: Combined workflow for quantitative analysis and screening

Figure 4a: Quantitation and Screening in a Single Workflow (purple flag indicates detection of a compound for which the

standard is not available and thus calibration and quantification is not possible)

Figure 4b: Quantitation and Screening in a Single Workflow (green flag indicates detection of a compound for which the

standard is available to permit calibration and quantification)

DATA PROCESSING


38/38


you can find, for each compound, fiveto six high-resolution MS/MS spectracollected at different collision energies.

TraceFinder is also available with anEnvironmental and Food Safety HRAMMS/MS spectral library that containsover 1500 compounds and 8900 spectra.[http://www.thermoscientific.com/en/product/high-resolution-3.html]

Reporting

TraceFinder includes a wide range ofstandard reports as well as customizablereporting templates. Using a MicrosoftExcel-like report generator, users canedit or add fields and calculations to thestandard reports or create original reportformats that match specific needs. Amongthe exhaustive list of templates are toolsfor batch, breakdown, internal standard

summary, and quantitative reports.

ConclusionTraceFinder is designed for routine, high-throughput GC–MS/MS, LC–MS/MS,and high-resolution MS qualitative andquantitative workflows. It provides thetools necessary to maximize productivitywith quick method development, efficient

data review, and custom reporting. It

includes Intelligent Sequencing with anoption to specify the actions you wantthe application to take when there is an

acquisition failure, and an audit viewer toview the audit log files in order to trackuser access and user modifications to thedata. There is functionality to exchangethe data between TraceFinder and aLaboratory Information ManagementSystem (LIMS). Indeed, TraceFinder hasthe security and auditing features to help

meet regulatory requirements.

About the Authors

Katerina Bousova studied food chemistry and chemical analysisat the faculty of food and biochemical technology at the Institute ofchemical technology in Prague, finishing her Master’s degree therein 2006. From 2006 to 2010, she worked as a chemical analyst at thedepartment of chemistry at the state veterinary Institute in Prague.In 2010, Thermo Fisher Scientific established a new laboratory, theFood Safety Response Center, in Dreieich, Germany, where Katerinastarted working as an application specialist. She became a memberof a three-member team focused on the rapid development of ana-

lytical methods to respond to emerging contaminations in food. In2013, the name of the group changed to the “Special Solutions Cen-tre.” Katerina is still responsible for developing new analytical meth-ods, but is also giving presentations about new trends in food safety,providing training for customers, and preparing various workshops inthe area of food testing.

Richard Fussell focuses on food safety, GC–MS and traceelemental applications in his current role. Before joining ThermoFisher Scientific, he gained more than 35 years of experienceworking for the UK government, including 21 years at the Foodand Environment Research Agency (FERA, formerly called CSL).At FERA, Richard was an advanced research fellow and seniorscientist managing research and collaboration projects focused

on emerging MS techniques for pesticide residues and veterinarydrugs and food.

DATA PROCESSING

Documents

EB Pesticide Residue Analysis LCGC En