EFS Applications

Thermo ScientificLC-MS Applicationsfor Food, Beverage,and Water Testing

Includes sample preparation, chromatography, and mass spectrometry conditions

Contaminants

Drug Residues

Natural Compounds

Pesticides

Part of Thermo Fisher Scientific

Table of Contents

Click on a compound class at right to go to related notes.

Click the Search button to search for any word or phrase.

ContaminantsAcrylamideMelamine & Cyanuric AcidMycotoxinsPhenolic PollutantsToxins

Drug ResiduesAntibioticsAntibiotics & AntimicrobialsPharmaceuticalsPharmaceuticals, Personal CareProducts, and Pesticides

Natural CompoundsFlavanoidsFlavonoids

PesticidesPesticidesPesticides, Mycotoxins, Plant toxins

Index

t tt

tt

tt

tt

tt

tt

ContaminantsAcrylamide

319: Quantitation of Acrylamide in FoodSamples on the Thermo Scientific TSQQuantum Discovery by LC/APCI-MS/MS

t

Quantitation ofAcrylamide in Food Samples ontheTSQ Quantum Discovery by LC/APCI-MS/MSKevin J. McHale, Witold Winnik and Gary Paul; Thermo Fisher Scientific, Somerset, NJ, USA

Key Words

• TSQ QuantumDiscovery

• Hypercarb

• Acrylamide

• Quantitation

• TripleQuadrupole MS

ApplicationNote: 319

Introduction

Acrylamide has been identified as a potential human carcinogen. This is important not only because acrylamideis a common industrial chemical, but acrylamide has beenshown to be present at significant levels in food samples,1

particularly cooked foods high in carbo hydrates. Thishas led many government health agencies around theworld to assess the risk of short- and long-term exposureto acrylamide in humans.

This has led to the development of LC-MS/MS method-ology for the quantitative analysis of acrylamide in food-stuffs.1-5 While a GC/MS protocol for the analysis ofacrylamide exists, this method requires extensive samplecleanup and chemical derivatization.6 The advantage ofLC-MS/MS is that chemical derivatization is not necessaryprior to acrylamide analysis.

To date, most LC-MS/MS methods for the assay ofacrylamide have utilized an electrospray ionization (ESI)source for the production of acrylamide ions.1-4 Yet it iswell-known that ESI-MS is problematic when highlyaqueous solutions, such as those required for the reversed-phase LC separation of acrylamide, are used.7 On the otherhand, water does not pose a problem for the formationof a stable corona discharge used in APCI. One publishedreport has demonstrated that APCI is a viable ion sourcefor the production of acrylamide ions for LC-MS/MSdetection.5 Furthermore, a study comparing ESI and APCIion sources for the LC-MS/MS analysis of acrylamideshowed that under the same chromatographic conditions,APCI-MS/MS yielded an improved detection limit.8

This report presents data acquired on the ThermoScientific TSQ Quantum Discovery for the analysis ofacrylamide. A simple LC-MS/MS method using theAPCI source is used to measure acrylamide, via selectivereaction monitoring (SRM), over a wide concentrationrange. A small selection of food samples was analyzedfor acrylamide content following extraction with water.To preclude the need for a time-consuming solid-phaseextraction procedure, a column-switching method wasemployed to selectively “fractionate” acrylamide frompolar matrix interferences prior to LC-MS/MS detection.

Goals1. Development–A sensitive and rugged LC/APCI-MS/MS

assay for the analysis of acrylamide

2. Application–An on-line column-switching technique to aqueous food extracts as an alternative to solid-phaseextraction (SPE) cleanup

3. Measurement–Acrylamide content in selected foodsamples

Experimental

Chemicals and Reagents: Acrylamide (>99.0%) waspurchased from Fluka (Buchs SG, Switzerland). 2,3,3-d3-acrylamide (98%) was obtained from Cambridge IsotopeLaboratories (Andover, MA, USA). HPLC grade waterwas acquired from Burdick and Jackson (Muskegon,MI, USA). All chemicals were used as received withoutfurther purification.

Sample Preparation: Standards were prepared by dilutionof a stock solution of 1.0 mg/mL acrylamide or 1.0mg/mL d3-acrylamide in water. The stock solutions werestored at 4°C for a period of no longer than two weeks.

Two brands of potato chips and two brands of break-fast cereals were purchased and stored at room tempera-ture until processed. After homogenizing approximately50 grams of a food sample, two grams were weighed intoa 35 mL polypropylene centrifuge tube. Aqueous extrac-tion of acrylamide was initiated by the addition of 20 mLwater containing 1000 ng d3-acrylamide as the internalstandard (final concentration = 50 ng/mL). The samplewas vortexed for 30 s then subsequently centrifuged at18,000 g for 15 minutes. Ten milliliters of the supernatantwas decanted into a clean 35 mL centrifuge tube andcentrifuged at 18,000 g for 10 minutes. Prior to analysis,0.49 mL of the aqueous extract was filtered through a0.45 µm centrifuge filter (Millipore Corp., Bedford, MA,USA) at 9,000 g for 5 minutes.

Sample Analysis: LC experiments were conducted with theThermo Scientific Surveyor HPLC system. A ThermoScientific Hypercarb 2.1× 50 mm column was utilized asthe analytical LC column. Separations of acrylamide wereachieved under isocratic conditions using 100% water asthe mobile phase at a flow rate of 0.4 mL/min. The injec-tion volume for all LC experiments was 10 µL.

Administrator

Text Box

Search | Contents | Index

To eliminate the need for solid phase extraction (SPE)purification prior to the analysis of the food sampleextracts, a column-switching LC method was employed.Briefly, the sample extract was loaded onto a 2.1× 50 mmThermo Scientific Aquasil C18 column, which was posi-tioned before a 6-port switching valve. The eluent fromthe C18 column was diverted to waste except for theperiod when acrylamide eluted from the C18 column,whereby the valve was switched to the Hypercarb columnfor MS/MS detection. This column-switching methodrequired a second Thermo Scientific Surveyor MS pump,which also delivered 100% water at 0.4 mL/min. BothSurveyor MS pumps and the 6-port switching valve werecontrolled using Thermo Scientific Xcaliburversion 1.3software.

The experimental conditions for the TSQ QuantumDiscovery were as follows:

Source: APCIIon polarity: PositiveVaporizer Temperature: 375°CDischarge Current: 5 µAIon Transfer Capillary Temperature: 250°CSource CID Offset: 6 VScan Mode: Selective Reaction MonitoringQ2 Pressure: 1.0 mTorr argonSRM Transitions: m/z 72 →55 for acrylamide;

m/z 75 →58 for d3-acrylamideCollision Energy: 13 eVScan Width: 1.0 uScan Time: 0.3 s (each SRM transition)Q1, Q3 Resolution: Unit (0.7 u FWHM)

Results and Discussion

Prior to the acquisition of acrylamide standards, it wasimportant to determine if there was any detectable nativeacrylamide contribution originating from the deuteratedinternal standard. As shown in Figure 1, there is no acrylamide signal observed for the m/z 72 → 55 SRMtransition at the same retention time as the 50 ng/mL d3-acrylamide standard.

The limit of quantitation (LOQ) for acrylamide onthe TSQ Quantum Discovery was 0.25 ng/mL acrylamideor 2.5 pg on column (Figure 2). This compares favorablyto LOQs previously reported by other research groups,including an 8-fold improvement over the mass LOQ byLC/ESI-MS/MS (20 pg)1 and a 40-fold improvement overthe concentration LOQ on the TSQ 7000 (10 ng/mL),5

which used an LC/APCI-MS/MS method.The calibration curve for acrylamide from 0.25 ng/mL to

2500 ng/mL is displayed in Figure 3. This calibration curvewas generated using the column-switching LC method justprior to the acquisition of the food extracts data. A linearregression fit to these data using 1/x weighting yielded thefollowing equation: y = 5.5997 × 10–4 + 0.0206125x. Thecorrelation coefficient for this curve was r2 =0.9999, indi-cating excellent linearity across the four orders of magni-tude dynamic range. Table 1 summarizes the statisticalresults for the acrylamide calibration curve. At the LOQ,

Figure 1: SRM chromatograms for 50 ng/mL d3-acrylamide

Acrylamide72 → 55

d3-Acrylamide75 → 58Area = 1507019

2100

2000

1900

1800

1700

1600

1500

160000

140000

120000

100000

80000

60000

40000

20000

Inte

ntis

yIn

tent

isy

RT: 1.95

Time (min)0.0 0.5 1.0 1.5 2.0 2.5 3.0

Figure 2: SRM chromatograms for 0.25 ng/mL acrylamide (LOQ)with 50 ng/mL d3-acrylamide

Acrylamide75 → 55Area = 26998S/N = 17


6500

6000

5500

5000

4500

4000

3500

500000

400000

300000

200000

100000

Inte

ntis

yIn

tent

isy

RT: 2.91

RT: 2.95

Time (min)1.5 2.0 2.5 3.0 3.5 4.0

Figure 3: Calibration curve for acrylamide using column-switchingLC method with APCI-MS/MS detection

0 500 1000 1500 2000 25000

10

20

30

40

50

0 2 4 6 8 10

ng/mL

ng/mL

0.00

0.05

0.10

0.15

0.20

0.25

Are

a Ra

tio

Are

a Ra

tio

Administrator

Text Box


the accuracy, as a percent relative error, is 1.1% and theprecision, as a percent coefficient of variance (%CV), is12.1% for five replicate injections. Above the LOQ, therelative error varied from -2.9 to +2.4% and the %CVranged from 0.5 to 6.7%.

Results obtained from the aqueous extract of PotatoChip 2 are presented in Figure 4. By utilizing a C18column positioned before a switching valve to selectivelyelute acrylamide onto the Hypercarb column, backgroundinterferences are reduced. Unlike most of the other acry-lamide reports where SPE cleanup was used followingextraction of the sample with water,1-4 the column-switching LC method employed here provides an on-linemeans of acrylamide fractionation. This has the advantageof minimizing sample losses during SPE and greatlyreduces sample preparation time.

To monitor the consistency and reproducibility of thecolumn-switching LC-MS/MS method, a 1 ng/mL acry-lamide standard was analyzed immediately following eachfood sample. An example of this quality control standardanalyzed after the Potato Chip 2 sample is shown inFigure 5. Although the baseline for the m/z 72 → 55 SRMtransition is somewhat elevated near the retention time foracrylamide, the calculated concentration for this standardis 0.99 ng/mL, equating to a relative error of -1.0%.

Table 2 reports the results for four different foodsamples that were assayed for acrylamide using thecolumn-switching LC method and MS/MS detection.The acrylamide concentrations in each food sample werecalculated by multiplying the measured solution concen-tration from duplicate injections by the extraction volumeand dividing by the food sample mass that was extracted.The determined acrylamide concentrations correlated wellto those reported elsewhere for these classes of food.1-5

Nominal Mean Conc. % Rel. Number of(ng/mL) (ng/mL) Error % CV Replicates

0.250 0.253 1.1 12.1 50.500 0.485 -2.9 6.7 51.00 1.00(4) 0.4 4.6 55.00 4.86 -2.7 0.9 510.0 10.2 2.1 0.7 5100 101 0.7 0.5 5500 512 2.4 0.8 31000 1006 0.6 0.6 32500 2481 -0.8 0.6 3

Table 1: Statistical data for the calibration curve of acrylamide

Potato PotatoCereal 1 Cereal 2 Chip 1 Chip 2

Injection 1 17.17 ng/mL 55.93 ng/mL 57.11 ng/mL 29.18 ng/mL

Injection 2 17.00 ng/mL 56.18 ng/mL 56.52 ng/mL 29.14 ng/mL

Mean 17.09 ng/mL 56.06 ng/mL 56.82 ng/mL 29.16 ng/mL

Extraction Vol. 20.0 mL 20.0 mL 20.0 mL 20.0 mL

Mass Sample 2.003 g 2.007 g 2.021 1.995

Acrylamide Conc. 171 ng/g 559 ng/g 562 ng/g 292 ng/g

Table 2: Results of acrylamide assay from food samples

Figure 4: SRM chromatograms of the Potato Chip 2 sampleaqueous extract

250000

200000

150000

100000

50000

RT: 2.97


Acrylamide72 → 55Area = 3268462

400000

350000

300000

250000

200000

150000

100000

50000

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

RT: 2.93

Inte

ntis

yIn

tent

isy

Time (min)

Figure 5: 1 ng/mL acrylamide standard analyzed directly afterduplicate injections of the aqueous extract of the Potato Chip 2sample

RT: 2.94

RT: 2.91


Acrylamide75 → 58Area = 115877

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

400000

350000

300000

250000

200000

150000

100000

50000

450000

20000

18000

16000

14000

12000

10000

8000

6000

Inte

ntis

yIn

tent

isy

Time (min)

Administrator

Text Box


In addition to these

offices, Thermo Fisher

Scientific maintains

a network of represen -

tative organizations

throughout the world.

Africa+43 1 333 5034 127Australia+61 2 8844 9500Austria+43 1 333 50340Belgium+32 2 482 30 30Canada+1 800 530 8447China+86 10 5850 3588Denmark+45 70 23 62 60 Europe-Other+43 1 333 5034 127France+33 1 60 92 48 00Germany+49 6103 408 1014India+91 22 6742 9434Italy+39 02 950 591Japan +81 45 453 9100Latin America+1 608 276 5659Middle East+43 1 333 5034 127Netherlands+31 76 587 98 88South Africa+27 11 570 1840Spain +34 914 845 965Sweden / Norway /Finland+46 8 556 468 00Switzerland+41 61 48784 00UK +44 1442 233555USA +1 800 532 4752

www.thermo.com

AN62606_E 01/08S


Thermo Fisher Scientific,San Jose, CA USA is ISO Certified.

Legal Notices©2008 Thermo Fisher Scientific Inc. All rights reserved. All other trademarks are the property of Thermo Fisher Scientific Inc. and its subsidiaries. Thisinformation is presented as an example of the capabilities of Thermo Fisher Scientific Inc. products. It is not intended to encourage use of these products inany manners that might infringe the intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are availablein all countries. Please consult your local sales representative for details.

View additional Thermo Scientific LC/MS application notes at: www.thermo.com/appnotes

Conclusions

An LC-MS/MS method has been developed for the meas-urement of acrylamide on the TSQ Quantum Discovery.Using APCI for the analysis of acrylamide from 100%water, an LOQ of 0.25 ng/mL acrylamide or 2.5 pg oncolumn was achieved. Incorpo ration of a column-switching LC method prior to MS/MS detection of acry-lamide eliminated the need to purify food sample extractsby SPE. The method was successfully demonstrated forthe analysis of four brands of food samples using TSQQuantum Discovery in conjunction with a column-switching LC method.

References

(1) Tareke, E.; Rydberg, P.; Karlsson, P.; Eriksson, S.; Tornqvist, M.J. Agric. Food Chem., 2002, 50, 4998-5006.

(2) Rosen, J.; Hellenas, K. Analyst, 2002, 127, 880-882.

(3) Musser, S. M. http://www.cfsan.fda.gov/~dms/acrylami.html

(4) Becalski, A.; Lau, B.P.; Lewis, D., Seaman, S.W. J. Agric. Food Chem.,2003, 51, 802-808.

(5) Brandl, F.; Demiani, S.; Ewender, J.; Franz, R.; Gmeiner, M.; Gruber, L.;Gruner, A.; Schlummer, M.; Smolic, S.; Stormer, A.; Wolz, G. Electron. J.Environ. Agric. Food Chem., 2002, 1(3), 1-8.

(6) Castle, L.; Campos, M.J.; Gilbert, J. J. Sci. Food Agric. 1993, 54,549-555.

(7) Ikonomou, M.G.; Blades, A.T.; Kebarle, P. J. Am. Soc. Mass Spectrom.,1991, 2, 497-505.

(8) McHale, K.J.; Winnik, W.; Paul, G. Proceedings of the 51st ASMSConference on Mass Spectrometry and Allied Topics, Montreal, 2003.

Administrator

Text Box


ContaminantsMelamine & Cyanuric Acid

424: Analysis of Melamine and Cyanuric Acid in Food Matrices by LC-MS/MS

t

Analysis of Melamine and Cyanuric Acidin Food Matrices by LC-MS/MSPeterVarelis, National Center for Food Safety and Technology, Illinois Institute of Technology.Jonathan Beck, Kefei Wang, and Dipankar Ghosh; Thermo Fisher Scientific, San Jose, CA


Key Words

• TSQ QuantumUltra

• Cyanuric Acid

• Food Safety

• LC-MS/MS

Introduction

In March 2007, several North American manufacturersof pet food voluntarily issued nationwide recall notices forsome of their products that were reportedly associated withrenal failure in pets1. The raw material wheat gluten, usedto manufacture the pet food, was imported from Chinaand was identified as the source of contamination2.

Although initial reports suggested that contaminationwas confined to pet food, further investigations revealedthat melamine-tainted fodder may have been used to feedanimals intended for human consumption.3,4,5 In particular,it was discovered that melamine-contaminated ingredientshad been used to prepare feed for chickens, swine, andcatfish.3,4 Consequently, the U.S. Food and Drug Administra - tion (FDA)3 and the U.S. Department of Agriculture (USDA)4

have developed methods for the analysis of melamineresidues in animal tissue. Both methods use tan dem massspectrometric detection and employ dispos able strongcation exchange solid phase extraction (SPE) car tridgesto prepare samples for liquid chromatographic analysis.

Experimental

Chemicals and reagentsUnless stated otherwise, all organic solvents used in thiswork were HPLC grade quality and were purchased fromThermo Fisher Scientific (Fair Lawn, NJ, USA). Melamine,cyanuric acid, and 30% (w/w) aqueous ammonia werepurchased from Sigma (St. Louis, MO, USA). The internalstandards 13C3-melamine and -cyanuric acid were preparedusing 13C3-cyanuric chloride, which was also obtained fromSigma. 18 MΩ water was obtained from a Milli-Q™ (Milli -pore Corporation, Billerica, MA, US) purification system.

Calibration StandardsIndividual solutions (1000 µg/mL) of cyanuric acid andmelamine were prepared by dissolving the crystalline com -pounds in 50% (v/v) aqueous methanol. Aliquots (1 mL)of these solutions were combined and then diluted with1:3 water-acetonitrile, respectively, to obtain a 10 µg/mLstock solution, from which eight working standards in therange of 1-1000 ng/mL were prepared by serial dilutionswith acetonitrile. Calibration standards were preparedby adding 50 µL of the stock solution of the internalstandards to 1 mL of each of the eight working standards.

Sample PreparationSolid samples were homogenized using an Ultra-Turrax®

(IKA®-Werke GmbH & Co. KG, Staufen, Germany)homogenizer. After extraction into aqueous 1:1Water:MeOH, and addition of the internal standards,the samples were prepared by offline ion exchangechroma tography using SPE cartridges.

Liquid ChromatographyAliquots (10-25 µL) of the above extracts were chromato -graphed on a Thermo Scientific BioBasic AX analyticalcolumn (2.1×150 mm, 5 µm), which was kept at 30°C inan oven. The initial mobile phase was composed of ace-tonitrile-isopropanol-50 mM aqueous ammonium acetatein the ratio of 85:10:5, respectively, and was pumpedthrough the column at a flow of 400 µL per minute.

After 5 min, the mobile phase composition and flowwere immediately changed to 9:1 water-acetonitrile, and500 µL per minute, respectively. These conditions weremaintained for 5 min before returning the mobile phaseto the initial composition. After 5 min of equilibration,the flow through the column was returned to 400 µLper minute. The column effluent was diverted to wastefor the first 1.5 minutes and then switched to the detectorfor the remaining run time.

MS Conditions – Melamine

MS: Thermo Scientific TSQ Quantum UltraSource: Heated Electrospray (H-ESI)Ionization: Positive ESISheath Gas: 65 unitsAuxiliary Gas: 35 units at 250°CIon Transfer Tube Temp: 350°CScan Time: 200 ms/transitionQ1/Q3 Resolution: 0.7 FWHM

SRM Transitions:Melamine 13C3

Melamine: (Internal Standard):m/z 127→68 @ 32 eV m/z 130→70 @ 32 eVm/z 127→85 @ 18 eV m/z 130→87 @ 18 eV

QED-MS/MS Conditions: Collision Energy: 30 eVReverse Energy Ramp (RER): 50 eV

Administrator

Text Box


of 6

MS Conditions – Cyanuric Acid

MS: TSQ Quantum UltraSource: Heated Electrospray (H-ESI)Ionization: Negative ESISheath Gas: 75 unitsAuxiliary Gas: 10 units at 250°CIon Transfer Tube Temp: 350°CScan Time: 200 ms/transitionQ1/Q3 Resolution: 0.7 FWHM

SRM Transitions:Cyanuric Acid 13C3

Cyanuric Acid: (Internal Standard):m/z 128→42 @ 17 eV m/z 131→43 @ 17 eVm/z 128→85 @ 11 eV m/z 131→87 @ 11 eV

Results

A chromatogram showing a standard mixture of bothmelamine and cyanuric acid, along with their associatedinternal standards, is shown in Figure 1. Calibrationcurves ranging from 1-1000 ppb are shown in Figure 2and Figure 3 for melamine and cyanuric acid, respectively.The calibrations are linear over the entire range, and aclose-up of the lower portion of the calibration curve(1-100 ppb) is shown in the same figure.

Melamine and cyanuric acid were spiked into a matrixof catfish and processed as described in the method sectionabove. A chromatogram of this sample, spiked at 10 ppb formelamine and 50 ppb for cyanuric acid, is shown in Figure4. Very low noise is observed, emphasizing the effectivenessof the cleanup procedure for a complicated matrix.

0 1 2 3 4Time (min)

0

50

1000

50

1000

50

100

Rela

tive A

bund

ance

0

50

1004.30

4.753.743.543.34

4.29

3.35 4.554.082.22

1.570.30 1.691.200.83 2.37 2.57

2.22

2.922.860.22 0.55 1.09 1.841.61

Cyanuric Acid

Cyanuric Acid 13C3

Melamine

Melamine 13C3

Figure 1. Melamine, cyanuric acid, and their internal standards at a concentration of 1 ppb.From top to bottom, cyanuric acid, cyanuric acid 13C3, melamine, and melamine 13C3.

MelamineY = 0.0118128*X R^2 = 0.9998 W: Equal

0 200 400 600 800 1000ng/mL

0

2

4

6

8

10

12

Heig

ht R

atio

MelamineY = 0.0118128*X R^2 = 0.9998 W: Equal

20 40 60 80 100ng/mL

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

Heig

ht R

atio

Figure 2. Calibration curve for melamine from 1-1000 ng/mL. The left figure shows the entire calibration range, while the right figure shows the expandedrange from 10-100 ng/mL.

Administrator

Text Box


of 6

Additionally, full spectra data was collected using thestandard Quantitation-Enhanced Data-Dependent MS/MS(QED-MS/MS) scan function. QED-MS/MS works bymonitoring SRM data, and when the response of a partic-ular SRM reaches a threshold level, the full scan MS/MSis activated. The resulting full scan spectra for melamineat 100 ppb and its internal standard are shown in Figure 5.The full scan data allows for further confirmation ofresults by eliminating “false positives” and also providesthe opportunity to perform a library search. When a fullscan QED-MS/MS spectra collected from a catfish sample

spiked at 10 ppb was searched against the library, thelibrary search returned melamine as the most likely hit.The results of the library search are shown in Figure 6.The spectrum of the sample and the spectrum that isstored in the library are visible in the lower left quadrantof the figure. The top spectrum is the catfish sample, whilethe lower spectrum is the reference spectrum. There isgood agreement between the two spectra, even thoughthe reference spectrum was generated using standardswithout matrix.

Cyanuric_acid

Y = 0.00790388*X R^2 = 1.0000 W: Equal

0 200 400 600 800 1000ng/mL

0

1

2

3

4

5

6

7

8

Area

Rat

io

Cyanuric_acid

Y = 0.00790388*X R^2 = 1.0000 W: Equal

0 20 40 60 80 100ng/mL

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Area

Rat

io

Figure 3: Calibration curve for cyanuric acid from 1-1000 ng/mL. The left figure shows the entire calibration range, while the right figure shows the expandedrange from 1-100 ng/mL.

0 1 2 3 4Time (min)

0

50

1000

50

1000

50

100

Rela

tive A

bund

ance

0

50

1004.32

3.15 3.47 4.77

4.31

4.574.133.733.422.30

1.730.18 0.58 1.16 2.67 2.93

2.31

2.031.66 2.691.491.010.14

Cyanuric Acid50 ppb spiked in Catfish

Cyanuric Acid Standard

Melamine10 ppb spiked in Catfish

Melamine Standard

Figure 4: Chromatogram of cyanuric acid and melamine spiked into catfish matrix, at a level of 50 ppb for cyanuric acid,and 10 ppb for melamine

Administrator

Text Box


of 6

Catfish Sample

Library Spectrum

Figure 6: Library search results for melamine spiked at 10 ppb into a catfish matrix. Melamine is the top hit in the search list.

30 40 50 60 70 80 90 100 110 120 130 0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rela

tive

Abun

danc

e

44.17

130.06

30 40 50 60 70 80 90 100 110 120 130

m/z m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100 43.16

68.08 127.05

Rela

tive

Abun

danc

e

70.12

87.06

85.07

Melamine 13C3 Melamine

Figure 5: QED-MS/MS spectra for melamine 13C3 (left) and melamine (right). Unique, rich, library-searchable spectra are collected in the same chromatographicrun, allowing both quantitative and confirmatory full scan data in the same file.

Administrator

Text Box


of 6

Conclusion

A simple, sensitive, and specific method for the detectionand quantitation of melamine and cyanuric acid in foodmatrices has been demonstrated. The method is robustand allows for the analysis of a large number of samples,without degradation in column performance. Additionally,full scan spectra for Q3 are collected in the same chro-matographic run using the QED-MS/MS scan function,permitting a library search of the results to eliminate anyfalse positives.

References:1 Weis, E. Nestlé Purina, Hills join pet food recall. USA Today. Available at

http://www.usatoday.com/news/nation/2007-03-31-pet-food-recall_N.htm.Accessed 10 December 2007.

2 Aarthi Sivaraman. Melamine in pet food, wheat gluten fromChina: FDA. Reuters. Available athttp://www.reuters.com/article/domesticNews/idUSWEN594320070330.Accessed 10 December 2007.

3 Weise, E. and Schmit, J. Melamine in pet food may not beaccidental. USA Today. Available atwww.usatoday.com/money/industries/2007-04-24-fda-pet-food-probe_N.htm.

4 Fish on U.S. fish farms fed melamine-contaminated feed;FDA discovers contaminated food products from China mislabeled.American Veterinary Medical Association. Available atwww.avma.org/press/releases/070508_petfoodrecall.asp.Accessed 10 December 2007.

5 Melamine contaminant found in chicken feed. Available atwww.sciencedaily.com/releases/2007/05/070502072434.htm.Accessed 10 December 2007.

Administrator

Text Box


AN62732_E 05/08S


Legal Notices©2008 Thermo Fisher Scientific Inc. All rights reserved. Milli-Q is a registered trademark of Millipore Corporation. Ultra-Turrax is a registered trademark ofIKA-Werke GmbH & Co. KG. All other trademarks are the property of Thermo Fisher Scientific Inc. and its subsidiaries. This information is presented as anexample of the capabilities of Thermo Fisher Scientific Inc. products. It is not intended to encourage use of these products in any manners that might infringethe intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are available in all countries. Please consultyour local sales representative for details.

View additional Thermo Scientific LC/MS application notes at: www.thermo.com/appnotesThermo Fisher Scientific,San Jose, CA USA is ISO Certified.

Tap our expertise throughout the life of your instrument. Thermo Scientific Services

extends its support throughout our worldwide network of highly trained and certified

engineers who are experts in laboratory technologies and applications. Put our team

of experts to work for you in a range of disciplines – from system installation, training

and technical support, to complete asset management and regulatory compliance

consulting. Improve your productivity and lower the cost of instrument ownership

through our product support services. Maximize uptime while eliminating the

uncontrollable cost of unplanned maintenance and repairs. When it’s time to

enhance your system, we also offer certified parts and a range of accessories

and consumables suited to your application.

To learn more about our products and comprehensive service offerings,

visit us at www.thermo.com.

Laboratory Solutions Backed by Worldwide Service and Support








www.thermo.com

Administrator

Text Box


ContaminantsMycotoxins

377: Analysis of Mycotoxins in VariousCattle Forages and Food Matrices with the Thermo ScientificTSQ QuantumDiscovery MAX

t

Analysis of Mycotoxins in Various CattleForages and Food Matrices with theTSQ Quantum Discovery MAXRobert Huls1, Richard Zuiderent1, and Dipankar Ghosh2

1Thermo Fisher Scientific BV, Breda, The Netherlands; 2Thermo Fisher Scientific, San Jose, CA, USA


Key Words

• TSQ QuantumDiscovery MAX

• Food andEnvironmental

• H-SRM

• Quantitation

Introduction

Mycotoxins are toxic metabolites produced by certainspecies of fungi that can infect and colonize on variousagricultural crops in the field and during storage.Environmental factors such as temperature and humidityinfluence the occurrence of these toxins on grains, nutsand other commodities susceptible to mold infestation.In addition, any crop that is stored for more than a fewdays is a target for mold growth and mycotoxin formation.

Most mycotoxins are relatively stable compoundsthat are not destroyed by food processing or cooking.Although the generating organisms might not surviveprocessing, the toxin can still be present. Mycotoxins posea potential threat to human and animal health through theingestion of contaminated food products. Mycotoxins canhave both chronic and acute effects on human and animalhealth. They can be teratogenic, mutagenic, or carcinogenicin susceptible animal species. They are linked to variousdiseases in domestic animals, livestock, and humans inmany parts of the world. Most mycotoxins are toxic invery low concentrations and therefore require sensitiveand reliable methods for their detection.

This application note describes an LC-MS/MS methodfor the determination of mycotoxins in various cattleforages. Using this method it is possible to simultaneouslymeasure the following 12 mycotoxins within 12 minutes:Nivalenol (NIV), Deoxynivalenol (DON), Aflatoxin G1,Aflatoxin G2, Aflatoxin B1, Aflatoxin B2, Fumonisin B1,Fumonisin B2, Diacetoxyscripenol (DAS), T2-Toxine,Ochratoxin A, and Zearalenon (ZEN). See Figure 1.

The Thermo Scientific TSQ Quantum Discovery MAXtriple quadrupole system has been evaluated for round-the-clock analysis of different mycotoxins. Multiplesamples with different matrices (cattle forages, foodmatrices) have been analyzed.

Goal

To demonstrate that the TSQ Quantum Discovery MAX™,with its H-SRM capabilities and H-ESI source, is ideallysuited for the rigorous demands of high-throughputanalyses of mycotoxins in various matrices.

Experimental Conditions

Sample Preparation The samples analyzed were various extracts of cattleforages and food products. The following sampleextraction procedure, adapted from TLR InternationalLaboratories, was used. To begin, 25 g of groundedsample was dissolved in 100 mL of acetonitrile:water(80:20 v/v). The extract was then mixed for two hours.Afterwards, the extracts were filtered and diluted fourtimes with water. The resulting solution wasacetonitrile:water 20:80 v/v.

HPLCHPLC analysis was performed using the Thermo ScientificSurveyor HPLC System. Each 20 µL sample was injecteddirectly onto a Thermo Scientific Hypersil GOLD 100 × 2.1 mm, 5 µm analytical column. A gradient LCmethod used mobile phases A (0.1% formic acid in ace-tonitrile) and B (0.1% formic acid in water) at a flow rateof 0.3 mL/min. The gradient is described in Figure 2.

Mass SpectrometryMS analysis was carried out on a TSQ QuantumDiscovery MAX triple stage quadrupole mass spectrom-eter with a heated electrospray ionization (H-ESI) probe.The MS conditions were as follows:Ion source polarity: Positive ion modeSpray voltage: 4000 V Vaporizer temperature: 300°CSheath gas pressure (N2): 30 units Auxiliary gas pressure (N2): 30 unitsIon transfer tube temperature: 350°CScan Type: SRM

Administrator

Text Box


of 6

The MS conditions and the H-SRM transitions wereobtained by automatic optimization with the auto-tunesoftware. Figures 3 and 4 show two examples of thecollision energy optimization. Figure 5 summarizes allof the H-SRM transitions that were used.

Two product ions were measured for all compounds;one was used as the quantifier ion and the other was usedas the qualifier ion. In this way, the ion ratio confirmationwas done as an identity confirmation. See Figure 5 forfurther details.

Deoxynivalenol Nivalenol Diacetoxyscripenol T2-Toxine

Zearalenon Ochratoxin A Fumonisin B1 Fumonisin B2

Aflatoxin B2 Aflatoxin B1 Aflatoxin G1 Aflatoxin G2

Figure 1: Structures of 12 mycotoxins

Column: Hypersil GOLD™ 100 x 2.1 mm, 5 µm particlesFlow: 0.3 mL/minInjection volume: 20 µLColumn temperature: 30°CTotal time: 12 min

Figure 2: LC/MS conditions

Administrator

Text Box


of 6

Figure 3: Optimization of collision energies of Aflatoxin A1/B1/G2

Figure 4: Optimization of collision energies of Fumonsin B1/B2

Figure 5: H-SRM transitions

Product Ion Product Ion Conc RangeCompound Precursor (Quan) (Qual) RT (ppb)Aflatoxin B1 313 241 285 4.8 0.1–100Aflatoxin B2 315 259 287 4.5 0.1–100Aflatoxin G1 329 243 283 5.0 0.1–100Aflatoxin G2 331 245 275 4.8 0.1–100Fumonisin B1 722 334 352 4.4 0.1–1000Fumonisin B2 706 336 318 4.8 0.1–1000Ochratoxin 404 239 221 5.6 0.1–1000Zearalenon 319 187 185 5.7 0.1–100Deoxynivanlenol 297 249 231 1.2 0.1–100Diacetoxyscripenol 367 307 289 4.4 0.1–100

???

RT: 1.16MA: 107932

RT: 1.19MA: 101932

RT: 4.40MA: 6233

RT: 4.42MA: 4475

1.58

3.814.84

4.94

4.36

3.80

4.01

4.72

5.485.88

4.92

4.45

SRM H-SRM H-SRMSRM

100

95

90

85

80

75

70

65

60

55

50

45

40

35

30

25

20

15

10

5

00 1 2

Time (min)

Rela

tive

Abun

danc

e

Rela

tive

Abun

danc

e

Rela

tive

Abun

danc

e

Rela

tive

Abun

danc

e

3 4 5

100

95

90

85

80

75

70

65

60

55

50

45

40

35

30

25

20

15

10

5

0

100

95

90

85

80

75

70

65

60

55

50

45

40

35

30

25

20

15

10

5

0

100

95

90

85

80

75

70

65

60

55

50

45

40

35

30

25

20

15

10

5

00 1 2

Time (min)

3 4 53

Time (min)4 5 6 3

Time (min)4 5 6

Nivalenol in corn, 0.5 ppm Diacetoxyscripenol in flour, 10 ppb

Figure 6: Comparison of SRM and H-SRM data for two samples

a b

Administrator

Text Box


of 6


The TSQ Quantum Discovery MAX offers the uniquecapability of highly selective reaction monitoring (H-SRM).Setting the resolution of Q1 at 0.1 FWHM helps todecrease the background noise and eliminate isobaricinterferences. This improves the signal-to-noise ratio andresults into a lower limit of quantification. Figure 6compares SRM and H-SRM data for two samples.

The calibration curves were generated by dilutions inacetonitrile:water 20:80 v/v. Figure 7 presents the linearfit calibration curves for five mycotoxins using H-SRM.

The calibration curves have R2 values that are greaterthan 0.998, which indicate excellent linear fits over thedynamic range.

The mycotoxin levels found in the various matriceswere in the expected range. For example, in a QC-sampleused as an internal control, Aflatoxin B1 was expectedon a level of 5 ppb (in extract). The detected amounts(ppb in solution) are presented in Table 1. This level forAflatoxin B1, is subjected to EU legislation as the lowlimit of quantification.

Fuminosine_B1Y = -6629.54+96887.2*X-3.66559*X^2 R^2 = 0.9998 W: 1/X

0 100 200 300 400 500 600 700 800 900 1000 1100ng/ml

0

5000000

10000000

15000000

20000000

25000000

30000000

35000000

40000000

45000000

50000000

55000000

60000000

65000000

70000000

75000000

80000000

85000000

90000000

95000000

100000000

Are

a

Ochratoxine_AY = -11201.9+225947*X-28.6401*X 2̂ R^2 = 0.9996 W: 1/X

0 200 400 600 800 1000 1200ng/ml

0

10000000

20000000

30000000

40000000

50000000

60000000

70000000

80000000

90000000

100000000

110000000

120000000

130000000

140000000

150000000

160000000

170000000

180000000

190000000

200000000

210000000

220000000

230000000Diacetoxyscripenol_(DAS)

Y = -3583.47+11136.8*X R^2 = 0.9977 W: 1/X

0 500 1000 1500 2000 2500 3000 3500 4000ng/ml

0

2000000

4000000

6000000

8000000

10000000

12000000

14000000

16000000

18000000

20000000

22000000

24000000

26000000

28000000

30000000

32000000

34000000

36000000

38000000

40000000

42000000

44000000

46000000

48000000

Are

a

Deoxynivanlenol_(DON)Y = -12942.9+4471.77*X R^2 = 0.9991 W: 1/X

0 100 200 300 400 500 600 700 800 900 1000ng/ml

0

200000

400000

600000

800000

1000000

1200000

1400000

1600000

1800000

2000000

2200000

2400000

2600000

2800000

3000000

3200000

3400000

3600000

3800000

4000000

4200000

4400000

Are

a

Fuminosine_B2Y = 565.239+135163*X-0.586979*X^2 R^2 = 0.9998 W: 1/X

0 100 200 300 400 500 600 700 800 900 1000 1100ng/ml

0

10000000

20000000

30000000

40000000

50000000

60000000

70000000

80000000

90000000

100000000

110000000

120000000

130000000

140000000

Are

a

Fuminosine_B1Y = -6629.54+96887.2*X-3.66559*X^2 R^2 = 0.9998 W: 1/X

0 100 200 300 400 500 600 700 800 900 1000 1100ng/ml

0

5000000

10000000

15000000

20000000

25000000

30000000

35000000

40000000

45000000

50000000

55000000

60000000

65000000

70000000

75000000

80000000

85000000

90000000

95000000

100000000

Are

a

Ochratoxine_AY = -11201.9+225947*X-28.6401*X 2̂ R^2 = 0.9996 W: 1/X

0 200 400 600 800 1000 1200ng/ml

0

10000000

20000000

30000000

40000000

50000000

60000000

70000000

80000000

90000000

100000000

110000000

1 000000

130000000

140000000

150000000

160000000

170000000

180000000

190000000

200000000

210000000

220000000

230000000

Are

a

Diacetoxyscripenol_(DAS)Y = -3583.47+11136.8*X R^2 = 0.9977 W: 1/X

0 500 1000 1500 2000 2500 3000 3500 4000ng/ml

0

2000000

4000000

6000000

8000000

10000000

12000000

14000000

16000000

18000000

20000000

22000000

24000000

26000000

28000000

30000000

32000000

34000000

36000000

38000000

40000000

42000000

44000000

46000000

48000000

Are

a

Deoxynivanlenol_(DON)Y = -12942.9+4471.77*X R^2 = 0.9991 W: 1/X

0 100 200 300 400 500 600 700 800 900 1000ng/ml

0

200000

400000

600000

800000

1000000

1200000

1400000

1600000

1800000

2000000

2200000

2400000

2600000

2800000

3000000

3200000

3400000

3600000

3800000

4000000

4200000

4400000

Are

a

Fuminosine_B2Y = 565.239+135163*X-0.586979*X^2 R^2 = 0.9998 W: 1/X

ng/ml

10000000

20000000

30000000

40000000

50000000

60000000

70000000

80000000

90000000

100000000

110000000

120000000

130000000

140000000

Are

a

Fumonisin B1Range 0.1–1000 ppbR2=0.999Weighting=1/x

Fumonisin B2Range 0.1–1000 ppbR2=0.999Weighting=1/x

DeoxynivalenolRange 0.1–1000 ppbR2=0.999Weighting=1/x

Ochratoxin ARange 0.1–1000 ppbR2=0.999Weighting=1/x

DiacetoxyscripenolRange 0.1–1000 ppbR2=0.999Weighting=1/x

Figure 7: Calibration curves for five mycotoxins

Sample Detected Amount (ppb)Sample-01 1.19Sample-02 1.28Sample-03 1.43Sample-04 1.25Sample-05 1.15Sample-06 1.37

Average 1.28RSD 0.1RSD% 8.3%Average in Extract 5.11

Table 1. Detected Amounts of Aflatoxin B1 in Solution

Administrator

Text Box


of 6

For the analysis of mycotoxins in various matrices,the heated electrospray ionization (H-ESI) probe providessignificant advantages. The dual desolvation zone designincreases the ionization efficiency and helps to get rid ofthe clustering solvents. (See Figure 8.) This leads to highersignals with better %RSDs. The H-ESI probe also handles

higher LC flows (up to 1 mL/min) without losing ioniza-tion efficiency. This helps to speed up the method withoutthe need to split the LC flow. Figures 9 and 10 describethe increased sensitivity of the H-ESI probe with twosamples of mycotoxins.

Figure 8: H-ESI – Heated Electrospray Ionization probe

ESI Aflatoxin B10.5 ppb

Aflatoxin B20.5 ppb

H-ESI

RT: 4.78MA: 159292

RT: 4.47MA: 125169

RT: 4.55MA: 533364

RT: 4.85MA: 521821

H-ESI is 3 times moresensitive than ESI3.0

100

90

80

70

60

50

40

30

20

10

03.5 4.0 4.5

Time (min)5.0 5.5 6.0 3.0 3.5 4.0 4.5

Time (min)5.0 5.5 6.0

100

90

80

70

60

50

40

30

20

10

0

100

90

80

70

60

50

40

30

20

10

0

100

90

80

70

60

50

40

30

20

10

0

Figure 9: Increased sensitivity with the H-ESI probe in Aflatoxin data

Administrator

Text Box


Conclusion

LC-MS/MS is a major technique for all kinds of environ-mental safety and food control laboratories. The TSQQuantum Discovery MAX is the workhorse of the TSQQuantum series for round the clock productivity. Matrixeffects are always an issue with LC-MS/MS methods.However, this application note shows that the TSQQuantum Discovery MAX can help overcome these effectswith its unique features of H-SRM and H-ESI. The resultspresented here were obtained without extensive prepara-tion. A wide range of matrices were analyzed and excel-lent results were obtained.

Acknowledgements

Drs. Ing. Harm Janssens and BSc Gerard Franken of TLR InternationalLaboratories, www.tlr.nl, are acknowledged for supplying the basis for thismethod, which they developed for feed.

ESIFumonisin B10.5 ppb

Fumonisin B20.5 ppb

H-ESI

H-ESI is 3 times moresensitive than ESI

100

90

80

70

60

50

40

30

20

10

0

100

90

80

70

60

50

40

30

20

10

0

100

90

80

70

60

50

40

30

20

10

0

100

90

80

70

60

50

40

30

20

10

0

3.0 3.5 4.0 4.5Time (min)

5.0 5.5 6.0 3.0 3.5 4.0 4.5Time (min)

5.0 5.5 6.0

RT: 4.64MA: 14939

RT: 5.01MA: 23282

RT: 5.04MA: 71964

RT: 4.67MA: 53001

Figure 10: Increased sensitivity with the H-ESI probe in Fumonisin data

Thermo Finnigan LLC,San Jose, CA USA is ISO Certified.


©2007 Thermo FisherScientific Inc. All rightsreserved. All trademarks arethe property of Thermo FisherScientific Inc. and its sub-sidiaries.

Specifications, terms andpricing are subject to change.Not all products are availablein all countries. Please con-sult your local sales repre-sentative for details.




a network of represen-



Australia+61 2 8844 9500Austria+43 1 333 50340Belgium+32 2 482 30 30Canada+1 800 532 4752China+86 10 5850 3588Denmark+45 70 23 62 60 France+33 1 60 92 48 00Germany+49 6103 408 1014India+91 22 6742 9434Italy+39 02 950 591Japan+81 45 453 9100Latin America+1 608 276 5659Netherlands+31 76 587 98 88South Africa+27 11 570 1840Spain+34 91 657 4930Sweden / Norway /Finland+46 8 556 468 00Switzerland+41 61 48784 00UK+44 1442 233555USA+1 800 532 4752

www.thermo.com


AN62247_E 04/09S

Administrator

Text Box


ContaminantsPhenolic Pollutants

411: Analyzing Phenolic Pollutants in WaterUsing U-HPLC

t

Analyzing Phenolic Pollutants in WaterUsing U-HPLCN. Jones, L. Pereira, D. Milton, Thermo Fisher Scientific, Runcorn, UK

Key Words

• Hypersil GOLDColumns

• Accela HighSpeed U-HPLC

• Surveyor PlusHPLC

• Phenols

• US EPA and EUStandards

• Water Pollutants


Overview

This study demonstrates analysis optimization byvariation of column chemistry, and the viability ofreducing stationary phase particle size to significantlyincrease analysis speed, while maintaining separationparameters and increasing sensitivity.

Introduction

Phenolic compounds are of particular environmentalimportance due to their relatively high toxicity at lowlevels and their presence in environmental waters andorganic matter, following degradation of a range ofindustrial products such as pesticides and herbicides, aswell as naturally occurring humic substances and tannins.

Previous studies(1,2) have shown that reversed-phaseliquid chromatography (RP-LC) coupled to atmosphericpressure chemical ionization mass spectrometry (APCI-MS)can effectively separate and detect a range of phenoliccom pounds at low ppb levels, following various extractionmethods. Such methods provide a realistic alternative totraditional analysis approaches using gas chromatography(GC), which involve lengthy sample prepa ration/analysistimes and difficulty in derivatization of certain phenols.

In this study, the effect on the separation and analysisspeed of a number of priority phenols cited within theU.S. Environmental Protection Agency (EPA) and EuropeanUnion (EU) lists of priority pollutants(3) have been assessedby changing the chemistry and reducing the particle size ofthe stationary phase.

Materials and Methods

HPLC ColumnsThe effect of particle/column size variation on analysisspeed and separation efficiency was studied using thefollowing Thermo Scientific Hypersil GOLD columns andexperimental conditions:

150 × 2.1 mm (5 µm particle size)100 × 2.1 mm (3 µm particle size)100 × 2.1 mm (1.9 µm particle size).Mobile Phase: A) 0.1% Acetic Acid in Water

B) 0.1% Acetic Acid in Methanol.Temperature: 60°CDetection: UV Diode array (270-320 nm), Gradients, flow rates and injection volumes are listedin Table 1.Phenols were prepared at a concentration of 5 ppmin Water:Methanol (95:5).

Stationary phase chemistryThe effect of stationary phase chemistry on the separationof five phenols (2-Chlorophenol, 4-Chlorophenol, 2-Nitrophenol, 4-Nitrophenol and 2,4-Dinitrophenol),using 1.9 µm particles, was studied using three columntypes (all 100 × 2.1 mm):

Hypersil GOLD™

Thermo Scientific Hypersil GOLD aQ (polar endcapped C18)Thermo Scientific Hypersil GOLD PFP (perfluorinated phenyl).Analysis conditions were equivalent to those describedwithin U-HPLC Method 1 (Table 1).

InstrumentationA Thermo Scientfic Surveyor Plus HPLC system was usedfor 5 and 3 µm particle analyses, and a Thermo ScientificAccela U-HPLC system was used for 1.9 µm analyses.

Results

Effect of particle/column size on analysis speed and qualityThe analysis times of eleven priority phenolic pollutantswere significantly improved by reducing column dimensionsfrom 150 to 100 mm and particle size from 5 µm to 3 µm.Further improvements were achieved by changing to 1.9 µm particles, using the Accela™ High Speed LCSystem.

Administrator

Text Box


Typical chromatograms demonstrating improvementsin analysis speed are provided in Figures 1 to 3.

Analysis time was further reduced by increasing theflow rate of the U-HPLC analysis to 1000 µL/min,without any adverse effects on resolution (Figure 3). Thisis illustrated in the Table inset in Figure 3, which indicatespeak width and resolution values for all separations.

Stationary phase chemistryThe Hypersil GOLD 1.9 µm phase produced the optimaloverall separation of the chloro- and nitrophenols underthe standard conditions used.

The Hypersil GOLD PFP™ (perfluorinated phenyl)phase showed superior selectivity between chlorophenolcomponents, likely due to the unique selectivity enabledby the presence of fluorine in the stationary phase.However, the separation performance between thechloro- and nitrophenols was slightly reduced.

Chromatograms illustrating the effect on the sepa -ration of using different stationary phases are given inFigure 4, along with resolution values between 4- and2-chlorophenol (RS 6,7) and between 2-Nitro and4-Chlorophenol (RS 4,6).

Method A (Column 150x2.1 mm, 5 µm). Flow = 600 µL/min. Injection Volume = 5 µL.

Method B (Column 100x2.1 mm, 3 µm). Flow = 600 µL/min. Injection Volume = 1 µL.

UHPLC Method 1 (Column 100x2.1 mm, 1.9 µm). Flow = 600 µL/min. Injection Volume = 1 µL.

UHPLC Method 2 (Column 100x2.1 mm, 1.9 µm). Flow = 1000 µL/min. Injection Volume = 1 µL.

0.0 5 0.0 5 0.0 5 0.0 5 1.5 5 1.0 5 1.0 5 0.6 5 19.5 95 13.0 95 13.0 95 7.8 95 21 95 14.0 95 14.0 95 8.4 95 21.01 5 14.01 5 14.01 5 8.41 5 22.5 5 15.0 5 15.0 5 9.0 5

Time (min) Eluent B (%) Time (min) Eluent B (%) Time (min) Eluent B (%) Time (min) Eluent B (%)

Table 1: HPLC and U-HPLC gradients, flow rates, and injection volumes.

2 4 6 8 10 12 14 16

Time (min)

1

2

3

4

5,6

7 8

9

10 11

5 µm Particles

Rs (4,5) = 1.7

Pentachloropenol112,4,6–Trichlorophenol102,4–Dichlorophenol92,4–Dimethylphenol82–Chlorophenol74–Chlorophenol64–Methylphenol52–Nitrophenol42,4–Dinitrophenol34–Nitrophenol2Phenol1

No. Compound

Figure 1: Separation of priority phenolic pollutants. using a 5 µm particle packed column.

Administrator

Text Box


1 2 3 4 5 6 7 8 9 10 11 12

1 2 3 4 5 6 7 8 9 10 11 12

Time (min)

Time (min)

3 µm Particles

Rs (4,5) = 1.5

Rs (4,5) = 1.5

1.9 µm Particles

Figure 2: Chromatographic effect of variation in column dimensions (3 and 1.9 µm, 100 x 2.1 mm).

1.0 2.0 3.0 4.0 5.0 6.0 7.0Time (min)

1.30.06U-HPLC 21.9 µm. 100 x 2.1 mm

1.50.09U-HPLC 11.9 µm. 100 x 2.1 mm

1.50.13B3 µm. 100 x 2.1 mm

1.70.14A5 µm. 150 x 2.1 mm

Resolution(peaks 4,5)

Average PeakWidth (mins)Method

ColumnDimensions

Figure 3: Increased throughput using U-HPLC and 1.9 µm particles. Comparison of peak width (at 10% height) and resolution.

Administrator

Text Box


Figure 4: Comparison of 1.9 µm stationary phase chemistries for the separation of chloro- and nitrophenols.

AN62531_E 01/08S


Legal Notices©2008 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific Inc. and its subsidiaries. This informationis presented as an example of the capabilities of Thermo Fisher Scientific Inc. products. It is not intended to encourage use of these products in any mannersthat might infringe the intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are available in allcountries. Please consult your local sales representative for details.


Conclusions

A number of priority phenols can be successfullyseparated in shorter analysis times by transferring toU-HPLC methods using Hypersil GOLD 1.9 µm particlecolumns, without losing any significant resolution.

The increased peak efficiency observed for 1.9 µmparticle packed columns indicates that low level phenolanalyses in environmental matrices described in previousstudies,(1,2) would be further enhanced with increasedsensitivity.

Different column chemistries create important differ -ences in selectivity for method development purposes, whichmay aid studies involving, for example, the separation ofhalophenols using a Hypersil GOLD PFP phase.

References1 M.C. Alonso, D. Puig, I. Silonger, M. Grasserbauer, D. Barcelo,J Chromatogr. A, 823 (1998) 231-239.

2 J. Martinez Vidal, A. Belmonte Vega, A. Garrido Frenich. Analytical& Bioanalytical Chem. 1. 379 (2004) 125-130.

3 EPA Method 625, Phenols, Environmental Protection Agency, Part VIII,40 CFR Part 136, Washington DC, 1984.

Additional InformationFor additional information, please browse our Chromatography ResourceCenter which can be accessed from: www.thermo.com/columns








www.thermo.com


3.5 4.0 4.5 5.0

Time (min)

5.5 6.0 6.5 7.0 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5

Time (min)

3.5 4.03.0 4.5 5.0 5.5 6.0

Time (min)

Hypersil GOLD 1.9 µm Hypersil GOLD PFP 1.9 µm Hypersil GOLD aQ 1.9 µm

Rs (6,7) = 8.8Rs (4,6) = 4.2

Rs (6,7) = 15.5Rs (4,6) = 0.9

Rs (6,7) = 7.9Rs (4,6) = 1.1

2

3

4

6

7

6

4

7

2

3

2

3

4

6

7

Administrator

Text Box


ContaminantsToxins

379: Analysis of Microcystins from Blue-green Algae Using the Thermo ScientificTSQ Quantum Ultra LC-MS/MS System

t

Analysis of Microcystins from Blue-green AlgaeUsing the TSQ Quantum Ultra LC-MS/MS SystemMihoko Yamaguchi, Thermo Fisher Scientific, Yokohama, Japan


Key Words


• Blue-green algae

• LC-MS/MS

• Microcystin

• SRM

Introduction

Overgrowth of algae is a common problem in manywetlands with advanced stages of eutrophication (theenrichment of chemical nutrients containing nitrogen orphosphorus in an ecosystem). This often results in a thick,colored layer on the water's surface, known as an algalbloom. Some of the algae that grow in these bodies ofwater, known as Cyanobacteria or blue-green algae,produce toxic compounds known as microcystins.

Microcystins have a ring peptide structure consisting ofseven amino acids, and more than 80 homologs are known.One of the most widely studied of the microcystins isknown as Microcystin-LR, and is shown in Figure 1. Manyof the microcystins are particularly toxic to the liver. (SeeReferences.) Among them are Microcystin-LR, YR andRR, which have been detected in wetlands in Japan. Thisapplication note reports on the analysis of these micro-cystins by using LC-MS/MS.

Method

HPLC: HTC PAL Autosampler and Thermo ScientificSurveyor MS pumpColumn: Thermo Scientific HyPURITY

C18 2.1× 50 mm, 5 µMobile Phase A: Water with 0.1% Formic Acid Mobile Phase B: AcetonitrileGradient: 30%B (0.5 min) ➝ 80%B (in 3 min) ➝ 80%B

(2 min hold) ➝ 30%B (7 min hold)Injection Volume: 20 µLFlow: 0.2 mL/minColumn temperature: Room temperature

MS: Thermo Scientific TSQ Quantum UltraIonization: Positive ESISpray voltage: 5000 VSheath gas: 45 arbitrary unitsAuxiliary gas: 15 arbitrary unitsSweep gas: 2 arbitrary unitsCapillary T: 350°CSource CID: OffCollision gas: Ar, 1.2 mTorrScan Time: 0.15 secSRM setting: 519.9 ➝ 135.0 @ 32 V (RR)

995.7 ➝ 135.0 @ 65 V (LR)1045.8 ➝ 135.0 @ 70 V (YR)

SRM Chromatogram (STD 1.0 ppb)

The SRM chromatograms for 1.0 ppb standards areshown in Figure 2. The linear calibration curves of thestandards (0.1 ppb–1.0 ppm) are shown in Figure 3.

HCO 3 O

NH

HN

H3C

OOH C

O

N

HC 3

HC 2

O HN

H3C ONHH

N HC 3

HC 3O

OOC HO

HN

NH

HC 3HC 3

NH

H2N

O

HC 3

Figure 1: Microcystin-LR

R T : 3 .00 - 6 .0 0

3.0 3.5 4.0 4.5 5.0 5.5 6.0Time

0

1

2

3

4

5

6

7

8

9

10

Rela

tive

Abun

danc

e

4.35 Microcystin-RR4.72 Microcystin-YR

4.78 Microcystin-LR

Figure 2: SRM Chromatogram (RT 4.35: Microcystin-RR, RT 4.72:Microcystin-YR, RT 4.78: Microcystin-LR)

Administrator

Text Box



Conclusion

Microcystin-LR, YR and RR can be quantitativelyanalyzed over four orders of dynamic range (0.1 ppb –1.0ppm) by using the TSQ Quantum Ultra triple quadrupoleLC-MS/MS system.

References

T Ohta; R Nishiwaki; M Suganuma; J Yatsunami; A Komori; S Okabe;M Tatematsu; H Fujiki. 1993. Significance of the Cyanobacterial CyclicPeptide Toxins, the Microcystins and Nodularin, in Liver-Cancer. MutationResearch, 292:286-287.

JG Pace; NA Robinson; GA Miura; CF Matson; TW Geisbert; JD White.1991. Toxicity and Kinetics of [H-3] Microcystin-Lr in Isolated Perfused RatLivers. Toxicology and Applied Pharmacology, 107:391-401.

R Nishiwaki; T Ohta; E Sueoka; M Suganuma; K Harada; MF Watanabe;H Fujiki. 1994. Two significant aspects of microcystin-LR: Specific bindingand liver specificity. Cancer Lett, 83:283-289.

I Falconer; A Jackson; J Langley; M Runnegar. 1980. Liver Pathology of aToxin from the Bloom-Forming Blue-Green Alga Microcystis Aeruginosa.Proceedings of the Australian Biochemical Society, 13:41-41.


©2007 Thermo FisherScientific Inc. All rightsreserved. All trademarks arethe property of Thermo FisherScientific Inc. and its sub-sidiaries.

Specifications, terms andpricing are subject to change.Not all products are availablein all countries. Please con-sult your local sales repre-sentative for details.








www.thermo.com

AN62272_E 01/07S


Figure 3: Calibration Curves 0.1 ppb – ~1.0 ppm

Administrator

Text Box


t tDrug ResiduesAntibiotics

465: Analysis of (Fluoro)quinolones in Honey with Online Sample Extraction and LC-MS/MS

435: Analysis of Sulfonamides in RiverWater using Thermo Scientific EQuan,an Online Concentration Analysis System

397: Determination of SulfonamideAntibiotics in Wastewater by LiquidChromatography–Tandem MassSpectrometry

361: Determination of Trace Level Nitrofuran Metabolites in Crawfish Meat by Electrospray LC-MS/MS on the Thermo Scientific TSQ QuantumDiscovery MAX

358: Highly Selective Detection andIdentification of Nitrofurans Metabolites in Honey using LC-MS/MS

442: LC-MS/MS Analysis of Malachite Green, Leucomalachite Green, Ciprofloxacin,and Tetracycline in Food Samples using a TurboFlow Method

464: Multi-class Antibiotic Screening ofHoney Using Online Extraction with LC-MS/MS

354: On-line Enrichment HTLC/MS/MSAssay for Multiple Classes of Antibiotics in Environmental Water Sources

407: Simple and Rapid Analysis ofChloramphenicol in Milk by LC-MS/MS

tt

t

tt

tt

Analysis of (Fluoro)quinolones in Honey withOnline Sample Extraction and LC-MS/MSYves-Alexis Hammel, Nestle Research Center, Lausanne, Switzerland; Frans Schoutsen, Thermo Fisher Scientific, Breda, The Netherlands; Cláudia P. B. Martins, Thermo Fisher Scientific, Barcelona, Spain

IntroductionThe global food market has become more competitive andequally cost responsive. The need for analyticalprocedures that permit high sample throughput as well ashigher sensitivity allied to good reproducibility is growingby the day.1,2,3 A method using automated onlineextraction with tandem mass spectrometry is presented asan alternative to the commonly used, time-consumingsolid-phase extraction (SPE) method.

Quinolones, including fluoroquinolones, are a groupof synthetic antibacterial compounds used in the treatmentof several diseases. There has been a significant andprogressive increase in the use of quinolones in animalproduction, which has led to their residual presence infood. In the European Union, the maximum residue limits(MRLs) for several of these compounds are defined fordifferent food matrices of animal origin, but not forhoney.4 Furthermore, the presence of these compounds isan indication of unsafe practices of food production anddeficient methods in the production of honey.

The complexity of the matrix plays a fundamental roleon the adoption of the method of analysis. ThermoScientific TurboFlow technology enables the reduction ofsample preparation as well as the elimination ofinterferences from complex matrices such as honey.

GoalTo develop a sensitive and reproducible liquidchromatography tandem mass spectrometry (LC-MS/MS)method for the quantitation of 12 fluoroquinolones and 4quinolones in honey using automated extraction byTurboFlow™ technology.

Experimental

Sample Preparation

To a sample of 1 g of honey, 1 mL of water was addedand the mixture was homogenized. The sample was thenfiltered directly to the HPLC vial using a 0.22 µmpolyethersulfone membrane syringe filter.

Different concentration levels were achieved byspiking the sample with different concentration levels ofstandard stock solution.

The total sample preparation time was 40 minutes for12 samples.

TurboFlow Method Conditions:

System: Thermo Scientific Aria TLX-1 controlled by Aria™

software (Figure 1)Online Extraction: TurboFlow Cyclone 50 x 0.5 mmMobile Phase A: 0.1 % formic acid in waterMobile Phase B: 0.1 % formic acid in acetonitrileMobile Phase C: 10 mM ammonium formate in waterMobile Phase D: acetonitrile/isopropanol/acetone (4:3:3 v/v/v)Injection Volume: 90 µL

HPLC conditions:

Analytical Column: Thermo Scientific Hypersil GOLD 2.1 x 50 mm, 3 µm column at 40° C

Solvent A: 0.5 % formic acid in waterSolvent B: 0.5 % formic acid in methanol/acetonitrile (1:1 v/v)

Key Words

• TurboFlowTechnology

• Aria TLX-1


• Food Safety


Figure 1: Aria software with LC Method Editor

Administrator

Text Box


2

MS ConditionsMS analysis was carried out on a Thermo Scientific TSQQuantum Ultra AM triple stage quadrupole massspectrometer. The MS conditions were as follows:

Ion Source Polarity: Positive Ion ModeSpray Voltage: 3000 VVaporizer Temperature: 350 °CSheath Gas Pressure (N2): 40 unitsAuxiliary Gas Pressure (N2): 35 unitsCapillary Temperature: 325 °CCollision Gas (Ar): 1.5 mTorrQ1/Q3 Peak Resolution: 0.7 u (unit mass resolution)Scan Time: 0.025 sScan Width: 0.010 m/zData Acquisition Mode: SRM

The optimization of Selective Reaction Monitoring(SRM) parameters was performed by direct infusion ofstandards in the positive electrospray ionization mode.Collision induced dissociation (CID) mass spectra wererecorded for each analyte and the optimum collisionenergies were obtained for the selected ion transitions.Table 1 summarizes these parameters and Figure 2displays the MS method controlled by Thermo ScientificXcalibur software.

Table 1: Selected ion transitions (m/z), collision energy (CE) and tube lensvoltages (TL) for studied compounds

Precursor Ion Product Ion CE TLAnalyte (m/z) (m/z) (V) (V)

1. Nalidixic Acid 233.064 104.143 40 78215.020 15 78187.025 25 78

2. Oxolinic Acid 262.032 130.106 33 82244.012 19 82

3. Flumequine 262.050 199.998 34 61243.962 19 61

4. Cinoxacin 263.029 105.202 37 59189.014 29 59217.049 22 59245.011 16 59

5. Pipemidic Acid 304.062 189.000 29 82217.029 19 82286.075 20 82

6. Norfloxacin 320.096 276.058 17 70302.055 21 70

7. Enoxacin 321.083 206.012 29 65302.981 21 65

8. Ciprofloxacin 323.100 231.024 36 74314.018 22 74

9. Lomefloxacin 352.104 265.010 23 78308.067 17 78

10. Danofloxacin 358.120 82.215 39 75314.097 18 75340.089 24 75

11. Enrofloxacin 360.128 245.025 26 72315.958 19 72

12. Ofloxacin 362.107 261.041 27 109318.055 19 109

13. Marbofloxacin 363.066 70.067 34 6672.073 22 66

276.064 14 66320.022 14 66

14. Fleroxacin 370.094 269.023 27 112326.061 19 112

15. Sarafloxacin 386.095 298.979 28 105342.078 18 105367.878 22 105

16. Difloxacin 400.107 299.009 29 75356.017 20 75

Figure 2: MS method showing the SRM transitions and other conditions

Administrator

Text Box


3

Results and DiscussionThe analysis of food samples normally requires longpreparation times due to the complexity of the matrices.The Thermo Scientific Aria TLX-1 system powered byTurboFlow technology enables reduction of the samplepreparation time. It took only 40 minutes to prepare thebatch of samples for LC-MS/MS analysis, instead of anaverage time of 6 hours when using Solid PhaseExtraction (SPE). Even when dealing with complexmatrices, such as honey, the use of the TLX-1 systemenables the elimination of possible interferences andcreates less noisy chromatograms (Figure 3).

The results of a high-throughput, rapid, sensitive andlinear method for the determination of 16 quinolones,including 12 fluoroquinolones, by LC-MS/MS usingTurboFlow technology are presented (Table 2). The Limitof Detection (LOD) was calculated by using the statisticaldefinition LOD = YB + 3SB, where YB is the blank signaland SB is the standard deviation of the blank.

Figure 3: Representative SRM chromatogram (20 µg/kg) showing the selected ion transitions and retention times for the studied analyte

Administrator

Text Box








Africa-Other+27 11 570 1840Australia+61 2 8844 9500Austria+43 1 333 50 34 0Belgium+32 2 482 30 30Canada+1 800 530 8447China+86 10 8419 3588Denmark+45 70 23 62 60 Europe-Other+43 1 333 50 34 0Finland / Norway /Sweden+46 8 556 468 00France+33 1 60 92 48 00Germany+49 6103 408 1014India+91 22 6742 9434Italy+39 02 950 591Japan +81 45 453 9100Latin America+1 608 276 5659Middle East+43 1 333 50 34 0Netherlands+31 76 579 55 55South Africa+27 11 570 1840Spain+34 914 845 965Switzerland+41 61 716 77 00UK+44 1442 233555USA+1 800 532 4752

www.thermo.com

AN63070_E 06/09S





Table 2: Linearity, sensitivity and precision of the method

Range LOD RSD R2Analyte (µg/kg) (µg/kg) (%) (1/X)

1 1-50 0.8 1.3 - 4.7 0.99432 1-50 1.4 0.3 - 10.6 0.99093 1-100 0.9 1.7 - 8.9 0.99024 2-100 2.0 4.3 - 7.7 0.99185 1-100 0.9 1.5 - 10.1 0.99646 1-100 2.3 2.7 - 11.5 0.99257 1-100 1.9 2.1 - 11.7 0.99288 1-100 1.4 2.4 - 11.6 0.99679 1-100 0.5 0.2 - 13.7 0.995410 1-100 1.1 2.3 - 13.6 0.996111 1-100 0.8 1.5 - 16.9 0.990712 2-100 1.3 2.1 - 11.5 0.994513 1-100 2.6 2.4 - 13.9 0.993914 1-50 1.5 6.0 - 16.8 0.990315 1-100 1.1 1.1 - 11.2 0.996616 1-100 0.8 1.9 - 10.4 0.9947

The method proved to be linear in the range studied.Three replicates were used for each point of thecalibration levels, which, in addition to the relativestandard deviation values, demonstrate the precision ofthe method.

ConclusionA rapid, sensitive and reliable method for the quantitationof 16 quinolones, including 12 fluoroquinolones, wasdeveloped using a TurboFlow method in combinationwith a TSQ Quantum Ultra™ mass spectrometer. The useof TurboFlow technology enables a significant reductionof the sample preparation time. For 12 samples thepreparation time was reduced from 5 hours to 40 minutes.Preliminary trials indicate this online extraction coupledwith a TSQ Quantum Ultra is an excellent total solutionfor the quantification of a large number of compounds infood samples.

References and Acknowledgements1. Mottier, P.; Hammel, Y.-A.; Gremaud, E.; Guy, P. A., Quantitative High-

Throughput Analysis of 16 (Fluoro)Quinolones in Honey UsingAutomated Extraction by Turbulent Flow Chromatography Coupled toLiquid Chromatography – Tandem Mass Spectrometry. Journal ofAgricultural and Food Chemistry 2008, 56, 35-43.

2. Gunes, N.; Cibik, R.; Gunes, M. E.; Aydin, L., Erythromycin residue inhoney from the Southern Marmara region of Turkey. Food Additives andContaminants 2008, 25, 1313-1317.

3. http://www.cfsan.fda.gov/~comm/fluoroqu.html. U.S. Food and DrugAdministration. Preparation and LC/MS/MS analysis of honey forfluoroquinolones residues 2006.

4. EU Comission Regulation No. 2377/90. Laying down a communityprocedure for the establishment of maximum residue limits of veterinarymedicinal products in foodstuffs of animal origin. Off. J. Eur.Communities 1990, L224, 1-8.

Administrator

Text Box


Analysis of Sulfonamides in River Water using EQuan, an Online Concentration Analysis SystemYoko Yamagishi, Thermo Fisher Scientific, Yokohama, Japan

Introduction

Triple stage quadrupole LC-MS systems are often used forhighly sensitive quantitative analyses of environmentalpollutants. The performance of an LC-MS system is oneimportant factor that determines if it will efficiently andaccurately detect pollutants present at low concentrationsin environmental samples. Another factor is the process bywhich the system pre-treats the samples.

In its sample pretreatment process, the ThermoScientific EQuan system is able to quickly perform thetime-consuming concentration process online, dramaticallyreducing the time required for analysis. Consequently,EQuan™ is capable of measuring multiple samples moreefficiently than systems using off-line sample preparation,while reaching lower detection limits than are achievablewith a conventional LC-MS/MS analysis.

The following application note presents a quantitativeanalysis of sulfonamide antibiotics using EQuan. Thesecompounds are used widely as anti-inflammatory medicationsfor humans and livestock, and have recently becomecompounds of interest to regulatory agencies worldwide.Figure 2 shows the library spectra and chemical structuresfor the nine sulfonamide antibiotics used in this experiment.


Calibration standards were prepared using a mixedstandard solution of the nine LC target sulfonamides(Kanto Chemical Co., Ltd.). For the test samples, riverwater collected in Kanagawa Prefecture was passedthrough a 0.4 µm filter prior to analysis.

HPLC

Analytical Column: Thermo Scientific Hypersil GOLD C18 2.1 x 150 mm, 5 µm

Concentration Column: Hypersil GOLD™ C18 2.1 x 20 mm, 12 µm Mobile Phase A: 1 mM ammonium formate, 0.05% formic acid

in waterMobile Phase B: 1 mM ammonium formate, 0.05% formic

acid-methanolGradient (for analysis): 5% B (1.5 min) to 90% B (in 10 min) to 90% B

(5 min hold) Injection Volume: 0.5 mLFlow: 0.2 mL/minColumn Temperature: 40 °C

MS: Thermo Scientific TSQ Quantum

Ionization Mode: Positive ESISpray Voltage: 4500 VSheath Gas: 50AUX Gas: 20Sweep Gas: 0Capillary Temperature: 360 °CSkimmer Offset: 7 VScan Time: 0.1 sec/SRM transitionCollision Gas Pressure: Argon, 1.2 mTorrMass Resolution (FWHM): Q1 & Q3 0.7 Da

SRM Conditions

m/z 251.07 → 156.0 at 17 eV (sulfadiazine) m/z 254.07 → 156.0 at 16 eV (sulfamethoxazole) m/z 265.08 → 156.0 at 18 eV (sulfamerazine)m/z 268.08 → 156.0 at 14 eV (sulfisoxazole) m/z 279.10 → 186.0 at 19 eV (sulfadimidine)m/z 281.08 → 156.0 at 18 eV (sulfamethoxypyridazine) m/z 281.08 → 156.0 at 18 eV (sulfamonomethoxine) m/z 301.08 → 156.0 at 17 eV (sulfaquinoxaline) m/z 311.09 → 156.0 at 19 eV (sulfadimethoxine)

Key Words

• Antibiotics

• EQuan

• LC-MS/MS

• OnlineConcentration

• Sulfonamides


Figure 1: EQuan system schematic

Administrator

Text Box



Standard Sample Results

It was possible to detect all of the target compounds at aconcentration of 1.0 ppt using the EQuan system (Figure 3).Furthermore, linearity was obtained over the range of 0.5 to 100 ppt (Figure 4).

Figure 2: Product ion spectra (from MS/MS library)

Administrator

Text Box


Figure 3: SRM chromatogram of 1.0 ppt standard samples. Signal-to-noise (S/N) value shown above each peak. �–� on each chromatogram shows noiserange for calculating S/N.

Figure 4: Calibration curves

Administrator

Text Box


River Sample Assay Results

In the measurement of the river water, sulfamethoxazolewas detected at a concentration of 12.3 ppt and four othercomponents were detected at concentrations of about 1 pptor less (Figure 5).

The mixed standard sample was spiked in the riverwater samples at the concentration of 1.0 ppt (except forSulfamethoxazole, which was spiked at 10 ppt) and thesamples were analyzed. A good recovery rate of 70% to98% was obtained for each of the compounds. Furthermore,

reproducibility for allreplicates was 11% or lessfor the spiked samples (seeTable 1).

Conclusion

With the EQuan onlineconcentration analysissystem, it was possible tomeasure the sulfonamideantibiotics that were presentin the river water samples atlow concentrations quicklyand accurately.








www.thermo.com

AN62734_E 09/08M



Legal Notices©2008 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific Inc. and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scientific Inc. products. It is not intended to encourage use of these productsin any manners that might infringe the intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are available in all countries. Please consult your local sales representative for details.


Figure 5: SRM chromatogram of river samples. Signal-to-noise (S/N) value shown above each peak. �–� on eachchromatogram shows noise range for calculating S/N.

1.0 ppt spiked samples (n = 4)

Concentration in river Concentration inwater sample spiked samples Recovery

Compound (ppt) (ppt) rate (%) CV (%)

Sulfadiazine 0.35 1.19 84 10.4Sulfamerazine 0.39 1.13 73 5.1Sulfadimidine NF 0.98 98 11.3Sulfamethoxypyridazine NF 0.87 87 7.1Sulfamethoxazole 12.35 19.35* 70 2.8Sulfamonomethoxine 1.11 1.85 74 3.3Sulfisoxazole NF 0.95 95 8.5Sulfadimethoxine 1.42 2.19 77 1.4Sulfaquinoxaline NF 0.85 85 3.3* The spiking concentration was set at 10 ppt because sulfamethoxazole was detected in the river water samples at concentrations higher than 10 ppt.

Table 1: River water and spiked sample assay results and reproducibility

Administrator

Text Box


Determination of Sulfonamide AntibioticsinWastewater by Liquid Chromatography–Tandem Mass SpectrometryEleni Botitsi, Charalampia Frosyni, and Despina Tsipi; General Chemical State Laboratory, Athens, Greece

Key Words


• Surveyor HPLCSystem

• EnvironmentalApplication

• Sensitivity

• Solid PhaseExtraction (SPE)


Introduction

Antibiotics are widely used in human and veterinarymedicine for the prevention and treatment of bacterialinfectious diseases. An important but often disregardedaspect of antibiotic use is the fate of antibiotic residuesentering the environment.1 Pharmaceutical industrywastewater, improperly-disposed of unused antibiotics,and non-metabolized antibiotics excreted by humans canall enter the sewer system in low concentrations. Becausesewage treatment plants are rarely equipped to filter thesedrugs from wastewater, antibiotics are released into thewater system. Veterinary antibiotics used in livestockoperations are another major source of antibiotics inthe environment. Agricultural waste such as manureand water runoff can carry these antibiotics into thesoil and groundwater.

The effects of antibiotics in the environment are stillpoorly understood. One major concern is the developmentof antibiotic resistant strains of bacteria that could criticallydisturb the natural bacteria ecosystems and lead to aserious threat to human health. There are also concernsthat, exposure to environmental antibiotic residues mightlead to carcinogenic or allergic reactions in humans andcreate hazards to aquatic and soil organisms.2,3

Sulfonamides (Figure 1) are a common class ofsynthetic antimicrobials that are widely used in humanand in veterinary medicine and as feed additives to pro -mote growth in concentrated animal feeding opera tions.They are regarded as emerging contaminants that areintroduced into the environment predominantly in theUSA and Europe. There is no regulation of the levelsof these compounds in environmental matrices (water,sediment, soil). This is likely becauseof the limited knowledge of the input,fate, and effects of most pharma ceu -ticals in the environment. Therefore,sensitive and reliable analyticalmethods for detection of lowconcentrations (ng/L) of thesecompounds are needed.

Goal

To develop methods for the determination of sulfonamideantibiotics at trace levels in effluent wastewaters.


Sample Preparation Samples of secondary effluent were collected from sewagetreatment plants in Greece and then vacuumed filtered.Each 50 mL sample was diluted with 200 mL deionizedwater. After acidification to pH 4, 5 ng of the internalstandard d4-sulfamethoxazole (d4-SMX) was addedbefore enrichment to assess possible losses during theanalytical procedure. The effluent samples were enrichedby solid phase extraction (SPE). The diluted wastewatersamples were percolated through the cartridges at a flowrate of 5 mL/min. The cartridges were then washed with5 mL deionized water. Wastewater organics were elutedwith 2× 4 mL methanol. The solvents were evaporatedunder a stream of nitrogen gas and then the extractswere redissolved in 0.5 mL mobile phase A (0.1%formic acid in water).

HPLCHPLC analysis was performed using the Thermo ScientificSurveyor HPLC System. Each 20 µL sample was injecteddirectly onto a 150× 2.1 mm, 3.5 µm, C18 HPLC column.A gradient LC method used mobile phases A (0.1%formic acid in water) and B (0.1% formic acid inacetonitrile) at a flow rate of 0.2 mL/min.

NH2 S NH

O

O

N

NH2 S NH N O

O

O

CH3

NH2 S NH

O

ON

NCH3

NH2 S NH

O

ON

NCH3

CH3

NH2 S NH N

SO

O

NH2 S NH N

SO

O

Sulfamethazine (Mr = 278)Sulfamerazine (Mr = 264)

Sulfamethoxazole (Mr = 253)

Sulfapyridine (Mr = 249) Sulfathiazole (Mr = 255)

Figure 1: Chemical structures of some sulfonamide compounds

Administrator

Text Box


MSMS analysis was carried out on a Thermo Scientific TSQQuantum Ultra triple stage quadrupole mass spectrometerwith an electrospray ionization source. The MS conditionswere as follows:

Ion source polarity: Positive ion modeSheath gas pressure (N2): 40 unitsIon transfer tube temperature: 350°CCollision gas pressure (Ar): 1.0 mTorrQ1 resolution: 0.2 FWHM, Q3 resolution: 0.7 FWHMDwell time: 0.2 sScan Type: SRM

Table 1 summarizes the SRM transitions that weremonitored. MS detection of the target compounds wasdivided into three time segments on the basis of theirretention times during chromatography. The protonatedmolecular ion of the compound [M+H]+ was selectedas the precursor ion. Detection was performed in themultiple reaction monitoring mode using, usually, thetwo most intense and characteristic precursor/product-iontransitions obtained from the MS/MS optimizationprocedure. Identification criteria for the target compoundswere based on the LC retention time (tR) and on the ratioof the two monitored transitions for each compound.

Method accuracy and precision were evaluated byrecovery studies, using deionized water spiked withappropriate amounts of the sulfonamides at threeconcentrations (2 ng/L, 20 ng/L, and 200 ng/L).Calibration plots were obtained by analysis of standardsolutions at eight concentrations in the range 0.1µg/L–100 µg/L (2 pg–2000 pg injected).


The method validation data are summarized in Table 2.Linearity of the method was assumed because the r2 valueswere greater than 0.99 for the linear regression equations(1/x weighted) and the residuals were less than 20% foreach calibration point in the concentration range 0.1µg/L–100 µg/L. Quantification was performed on thebasis of external calibration plots using the peak areaof the most intense transition of the analyte. For SMX,quantification was performed using the ratios of thepeak areas of the most abundant monitored ion of theanalyte to that of the respective ion of the surrogatestandard d4-SMX.

The accuracy of the method was determined byrecovery studies conducted on spiked deionized watersamples at three concentration levels: 2 ng/L, 20 ng/L,and 200 ng/L. The precision of the method wasdetermined by repeated intra-day (n=3) and inter-dayanalysis (n=6) of samples spiked at the three concen -trations. High rates of recovery were achieved, usuallygreater than 72%, and relative standard deviations forinter-day analysis (n=6) ranged between 3.1% and 19.0%.The SRM chromato grams obtained from deionized waterspiked at a concentration of 2 ng/L are shown in Figure 2.

The method was applied to the wastewater samplesto investigate the occurrence of sulfonamide antibiotics.Sulfamethoxazole was detected in all of the samples.The median concentration was 150 ng/L. The SRMchromatograms of SMX in the wastewater effluent extractare shown in Figure 3.

Retention Precursor ion Product CECompound time (min) [M+H]+ m/z ions m/z (V)

Sulfapyridine, SPY 10.1 250 108, 156 20Sulfamethoxazole, SMX 26.2 254 108, 156 25Sulfathiazole, STZ 10.3 256 108, 156 17Sulfamerazine, SMR 12.0 265 108, 156 17Sulfamethazine, SMZ 18.6 279 186, 204 24D4-Sulfamethoxazole, D4-SMX 26.0 258 112, 160 25D4-Sulfathiazole, D4-STZ 10.0 260 112, 160 20

Table 1: Diagnostic ions of sulfonamide antibiotics

R2*

concentration range LOD Mean Recovery (± %RSD)

Compound 0.1–100 µg/L (µg/L) 2 ng/L H2O (n=6) 20 ng/L H2O (n=6) 200 ng/L H2O (n=6)

Sulfapyridine, SPY 0.997 0.053 77 (±7.8) 83 (±5.1) 120 (±8.4)Sulfamethoxazole, SMX 0.999 0.055 102 (±7.5) 99 (±5.5) 110 (±7.2)Sulfathiazole, STZ 0.999 0.054 79 (±14.0) 83 (±17.0) 106 (±4.2)Sulfamerazine, SMR 0.998 0.110 80 (±15.3) 87 (±3.1) 120 (±7.2)Sulfamethazine, SMZ 0.999 0.110 72 (±19.0) 77 (±14.0) 116 (±8.4)*linear fit calibration curves with 1/x weighting, (n=5 replicates)

Table 2: Validation data (linearity, accuracy, precision)

Table 1: Diagnostic ions of sulfonamide antibiotics

Administrator

Text Box


0 5 10

100

50

0100

50

0100

50

0100

50

0100

50

0100

50

0100

50

15 20 25 30 350

250 108, 156

254 108, 156

256 108, 156

258 112, 160

260 112, 160

265 108, 156

279 186, 204

SPY

SMX

STZ

D4-SMX (I.S.)

D4-STZ

SMR

SMZ

Time (min)

Rela

tive

Abun

danc

e

H2N S 2+

H2N O+

H2N SO2+

H2N O+

0 5 10 15 20 25 30 35 40

254 108, 156

254 108, 156

TICSample

TICStandard (5 µg/L)

SRM254 156

SRM254 108

SRM254 156

SRM254 108

Time (min)

Rela

tive

Abun

danc

e

m/z 156

m/z 108’

100

60

40

20

80

0100

60

40

20

80

0100

60

40

20

80

0100

60

40

20

80

0100

60

40

20

80

0100

60

40

20

80

0

Figure 2: LC-ESI(+)-MS/MS SRM chromatograms of a spiked (2 ng/L) deionized water extract

Figure 3: LC-ESI(+)-MS/MS chromatograms of sulfamethoxazole in wastewater sample extract and of a standard solutionof sulfamethoxazole

Administrator

Text Box









www.thermo.com

AN62480_E 10/07S





Conclusion

LC-ESI-MS/MS is a powerful analytical method for thesensitive determination of sulfonamide antibiotics inmunicipal wastewater at low ppt levels (ng/L). Thesolid phase extraction scheme for trace enrichment andseparation of sulfonamide compounds from wastewatersamples yielded high recovery rates and enabled theiraccurate quantification. Sulfamethoxazole (medianconcentration 150 ng/L) was detected in the wastewatereffluent samples, which indicated that it was notcompletely eliminated in the sewage treatment plants.The described method proved to be a valuable tool forthe detection of pharmaceuticals in wastewater effluentsbefore they reached the aquatic environment.

References1 Petrovic, M.; Hernando, M. D.; Diaz-Cruz, M. S.; Barceló, D. “Liquidchromatography-tandem mass spectrometry for the analysis of pharma ceu -tical residues in environmental samples: a review”; J, Chrom. A 2005,1067(1-2), 1-14.

2 Göbel, A.; McArdell, C. S.; Suter, M. J.-F.; Giger, W. “Trace determinationof macrolide and sulfonamide antimicrobials, a human sulfonamidemetabolite, and trimethoprim in wastewater using liquid chromatographycoupled to electrospray tandem mass spectrometry”; Anal. Chem. 2004,76(16), 4756-4764.

3 Yang, S.; Cha, K.; Carlson, K.“Quantitative determination of trace concen -trations of tetracycline and sulfonamide antibiotics in surface water usingsolid-phase extraction and liquid chromatography/ion trap tandem massspectrometry”; Rapid Commun. Mass Spectrom. 2004, 18(18), 2131-2145.

Administrator

Text Box


Determination of Trace Level NitrofuranMetabolites in Crawfish Meat by ElectrosprayLC-MS/MS on the TSQQuantumDiscovery MAXTao Ding,1 Jingzhong Xu,1 Chongyu Shen1 and Kefei Wang2

1 Food Laboratory, APFIC, Jiangsu Entry-Exit Inspection and Quarantine Bureau of People’s Republic of China, Nanjing, China2 Thermo Fisher Scientific, San Jose, CA, USA


Key Words


• Surveyor HPLC

• Food ResidueAnalysis

• SRM

• Veterinary Drugs

Introduction

Nitrofurans (furazolidione, furaltadone, nitrofurazoneand nitrofurantoin) are a group of veterinary antibioticsbanned in many countries because of human healthconcerns. The ban has stimulated significant interest indevelopment of analytical methods for detecting tracelevels of these drug residues in animal products.

Due to the rapid in vivo metabolism of the parentdrugs, detection of nitrofurans in meat products relieson determination of their corresponding tissue-boundmetabolites: 3-amino-2-oxazolidinone (AOZ), 3-amino-5-morpholinomethyl-2-oxazolidinone (AMOZ), semi -carbazide (SEM) and 1-aminohydantoin (AHD). Thesemetabolites were removed from tissues by hydrolysis inacidic solution and derivatized to nitrobenzyl- (NB-)

derivatives with 2-nitrobenzyladehyde (2-NBA). Figure 1illustrates the transformation. LC-MS/MS utilizing selectedreaction monitoring (SRM) of the corresponding fourmetabolite derivatives has become the method of choice.

In this note we describe a sensitive and selectiveLC-MS/MS method for detecting trace level nitrofuranmetabolites in crawfish using a Thermo Scientific TSQQuantum Discovery MAX triple quadrupole mass spec-trometer coupled to a Thermo Scientific Surveyor HPLCmodule. The limit of quantitation (LOQ) as low as<0.05 µg/kg has been clearly demonstrated in fortifiedcrawfish meat for all four nitrofuran metabolites. ThisLOQ represents twenty-fold better than the MinimumRequired Performance Limit (MRPL) established byEuropean Union (EU) in 2003.


Standards and Reagents

The following are a list of chemicals used in thiswork, and unless specified all chemicals are of atleast reagent grade.

AOZ and SEM•HCl (Sigma-Aldrich, St. Louis,MO, USA)

2-NBA (Sigma-Aldrich)DMSO (Sigma-Aldrich)d4-AMOZ and d5-AMOZ (Cambridge Isotope

Laboratory, MA, USA)1-Amino-imidazolidin-2,4-dione-[2,4,5-13C]

(WITEGA Laboratorien Berlin-Adlershof GmbH,Berlin, Germany)

Semicarbazide hydrochloride-13C, 15N2 (WITEGA)Ammonium Acetate (NH4Ac), K2HPO4, and

NaOH (Sigma-Aldrich)Methanol (HPLC grade, Thermo Fisher Scientific,

Pittsburgh, PA, USA)Water (in-house distilled water, filtered with a 0.45

µm filter)

O

N NO

O

O2N

N

O

OH2N

NO

O

N

NO2

O

NN

OO

O2N

N

O

N

OO

N

O

H2NN

O

O

NO

N

NO2

O NN

NH

OO2NH2N

N

NH

O

OO

NNH

O

N

NO2

O

O

N NH

O2N

O

H2N

H2N

HN O

NH2

NO2

N

HN O

NH2

Furazolidione AOZ NBAOZ

Furaltadone AMOZ NBAMOZ

Nitrofuranzone AHD NBAHD

Nitrofurantoin SEM NBSEM

Nitrofurans Metabolites NB-Derivatives

Figure 1: Nitrofurans, their metabolites, and 2-NBA derivatives

Administrator

Text Box


Analytical Equipment

HPLC: Surveyor HPLC module consisting of an ASAutosampler and MS PumpMass spectrometer: TSQ Quantum Discovery MAX

Analytical Standard Preparation

The primary analytical standard solutions of 1 mg/mLwere prepared by dissolving the corresponding solidstandards into methanol. The working standard solutionswere prepared by serial dilution of the primary standardsolution with 95:5 water:methanol.

Sample Preparation

Note: Nitrofuran metabolite derivatives are sensitive tolight; avoid prolonged exposure of sample to direct lightsources during sample preparation.

The extraction and derivatization of the nitrofuransfrom the crawfish were performed at Food Laboratory ofJiangsu Entry-Exit Inspectional and Quarantine Bureau atNanjing, China, following the published procedure withsome modification:

• To 2 g of homogenized crawfish meat inside a 50-mLglass tube, add 4 mL water, 0.5 mL of 0.5 M HClsolution, and 200 µL of freshly prepared 50 mM 2-NBAin DMSO. Vortex for one minute and store the samplein the dark at 37°C overnight (14-16 hours)

• After cooling the sample to room temperature,add 5 mL 0.1 M K2HPO4, adjust the pH of the mixtureto 7.0-7.5 with 0.4 M NaOH solution. Centrifuge themixture and collect the supernatant

• Extract the supernatant twice each time with 7 mLethyl acetate. Combine the ethyl acetate extracts andevaporate to dryness under N2 at 40°C

• Reconstitute the residues in 1.0 mL of water:methanol(95:5). Centrifuge the samples and filter the supernatantwith 0.2 µm syringe filter prior to injection to LC-MSsystem

Note that the sample preparation results in a two-foldconcentration that will be factored into the calculationof nitrofuran metabolite concentrations in meat samples.For fortified samples, the nitrofuran metabolites and theirinternal standards were added into the homogenized meatsample prior to the hydrolysis and derivatization. Forcalibration, the same procedures were followed exceptthat 2 mL of working standard solutions was used insteadof the meat samples.

Chromatography Conditions

Analytical column: Thermo Scientific Hypersil GOLD,5 µm, 100 × 2.1 mm

Eluent: A: 0.5 mM Ammonium Acetate in Water;B: Methanol

Gradient:Time (min) % A % B

0 80 208.5 50 509.5 50 5010 80 2015 80 20

Flow rate: 250 µL/minColumn temperature: Ambient (18-22°C)Injection volume: 20 µL (with loop)

Mass Spectrometry Conditions

The mass spectrometer was calibrated routinely with1,3,6-polytryosine, according to the standard operatingprocedures at the Nanjing laboratory. For methoddevelopment, a standard solution containing 1 µg/mL ofderivatized nitrofuran metabolites including the internalstandards was infused at 10 µL/min with 250 µL/min50:50 (A:B) mobile phase into the ESI source. First, thespray voltage, sheath gas, auxiliary gas and tube lens wereoptimized with the automated tune of Thermo ScientificXcalibur software. Second, the most abundant fragmentions and their optimized collision energy (CE) values werefound in MS/MS optimization. For known SRM transi-tions, parent and product ions can be input directly toobtain the optimized CE value for each SRM transition.Finally, the Source CID (skimmer offset voltage), collisiongas pressure, and ion transfer capillary temperature wereadjusted manually for best signal sensitivity. The finaloperation parameters are summarized as follows:

Ion source (polarity): ESI (+)Spray voltage: 5000 VSheath gas pressure: 30 unitsAuxiliary gas pressure: 8 unitsIon transfer capillary temperature: 300°CSource CID: 10 VScan type: SRMQ1 and Q3 peak width (FWHM): 0.7 DaCollision gas and pressure: Ar at 1.3 mTorr

For each parent ion, two SRM transitions were used,one for quantitation and one for confirmation, whichwould give 4 IP (identification points) to meet the EU’scriteria for residue analysis in food. Based on the elutionorder of the nitrofuran metabolite derivatives, thechromatography run was divided into three segments fordata acquisition. Table 1 lists SRM transitions and theirparameters in each segment.


Figure 2 shows representative chromatograms of a 0.050µg/kg fortified crawfish sample. As shown, all four nitro-furan metabolite derivatives were detected with excellentsignal quality as measured by signal-to-noise (S/N) ratio.

Administrator

Text Box


The LOQ values reported in literature, ranging from0.02 to 0.1 µg/kg for different nitrofuran metabolites inmeat samples, have mostly been obtained from extrapo -lation based on S/N =10 of the signals of analytes athigher concentrations in standards or fortified samples.In reality, however, these extrapolated LOQ’s often cannotbe achieved, because the S/N ratios of the signals deterio-rate more than as predicted by the dilution factor. Thecurrent data in Figure 2 demonstrates that all four nitro-furan metabolites can be detected in fortified crawfishsamples at 0.05 µg/kg level, far lower than the MRPLvalue of 1 µg/kg required by EU.

Figure 3 shows seven-point calibration curves con-structed from data from measuring standard solutionsat concentration levels of 0.025, 0.2, 0.5, 1, 2.5, 5.0 and

Time Parent Product CESegment (min) Analyte Ion (m/z) Ions (m/z) (V)

1 0–7.4 NBAOZ 236.045 104 19236.045 134* 22

d4-NBAOZ (IS) 240.037 134 14NBAHD 249.040 104 22

249.040 134* 1413C3-NBAHD (IS) 252.037 134 14

2 7.4–8.5 NBSEM 209.000 166* 11209.000 192 13

(13C, 15N2)-NBESEM (IS) 212.048 168 113 8.5–14.5 NBAMOZ 335.092 291* 12

335.092 262 19d5-NBAMOZ (IS) 340.134 296 12

TIC

249>134

249>104

252>134

TIC

236>134

236>104

240>134

TIC

335>291

335>262

340>296

TIC

209>166

209>192

212>168

5 6 7 8 9 10

Time (min)

0

100

0

100

0

100

0

100

RT: 6.99AA: 9031SN : 352

RT: 6.98AA: 6490SN : 502

RT: 7.00AA: 2542SN : 187

RT: 6.99AA: 5915SN : 69

NL: 1.70E3

NL: 1.21E3

NL: 5.62E2

NL: 1.07E3

5 6 7 8 9 10

Time (min)

0

100

0

100

0

100

0

100

RT: 7.82AA: 31604SN : 95

RT: 7.88AA: 14778SN : 104

RT: 7.82AA: 17214SN : 69

RT: 7.88AA: 13803SN : 95

NL: 4.71E3

NL: 2.20E3

NL: 2.77E3

NL: 2.48E3

5 6 7 8 9 10

Time (min)

0

100

0

100

0

100

0

100

RT: 7.13AA: 38204SN : 289

RT: 7.13AA: 26470SN : 239

RT: 7.13AA: 11734SN : 344

RT: 7.05AA: 39116SN : 869

NL: 5.74E3

NL: 3.94E3

NL: 1.80E3

NL: 5.81E3

5 6 7 8 9 10

Time (min)

0

100

0

100

0

100

0

100

RT: 9.23AA: 290684SN : 284

RT: 9.23AA: 55037SN : 1014

RT: 9.22AA: 234181SN : 227

RT: 9.13AA: 114339SN : 213

NL: 4.59E4

NL: 7.88E3

NL: 3.81E4

NL: 1.87E4

NBAHD

NBAOZ NBAMOZ

NBSEM

Figure 2: Chromatograms of shrimp meat sample containing 0.050 µg/kg fortified nitrofurans and internal standard in the shrimp meat.For each panel from the top: TIC (total ion current), SRM for quantitation (bold and red) and confirmation (green), and internal standard (italic and blue).RT: retention time, AA: peak area counts, SN: signal-to-noise ratio.

Table 1: Segments of chromatography separation and SRM transitionsNote: * SRM transition for quantitation; IS : internal standard; CE: Collision Energy.For each segment, Scan Time (s) = 0.1 and Scan Width (m/z) = 0.002.

Administrator

Text Box


10.0 ng/mL. Excellent linearity was obtained for all fournitrofuran metabolites within the calibration range, withthe correlation coefficient R2 > 0.995 (weight factor = 1/X).

The method accuracy and precision were evaluatedby performing triplicate preparation and analysis of onebatch of homogenized crawfish meat samples fortifiedwith nitrofuran metabolites at three different concentra-tion levels of 0.05, 0.5 and 2.5 µg/kg. The results aregiven in Table 2. As shown, recovery values in the rangeof 79%–110% were obtained with standard deviationsfrom 3 to 22%. It should be noted that standard devia-tions include the errors of both the sample preparation(major contributor) and analytical instrument.

Conclusions

With use of the TSQ Quantum Discovery MAX, a sensi-tive and reliable LC-MS/MS method using SRM has beendeveloped for detecting trace level nitrofuran metaboliteswith a quantitation limit of less than 0.050 µg/kg incrawfish. The sample preparation procedure is relativelystraightforward and setup of the instrument method iseasy and fast.

References

1. US FDA: Detection of Nitrofuran Metabolites in Shrimp,http://www.cfsan.fda.gov/~comm/methnf.html

2. A. Leitner et al., Journal of Chromatography A, 939 (2001) pp 49-58.

3. Seu-Ping Khong et al., J. Agric. Food Chem. 2004, 52 pp 5309-5315.

NBAHDY = 0 .0 85 02 68 +1 .06 85 7*X R ^2 = 0 .99 56 W: 1 /X

0 2 4 6 8 10pp b

NBAOZY = 0 .05 17 6 03 +0 .6 9 21 18 *X R ^2 = 0 .9 98 2 W: 1 /X

7

6

5

4

3

2

1

0

NBSEMY = 0 .0 5 6 2 9 1 2 + 1 .0 4 3 2 1 *X R ^2 = 0 .9 9 7 2 W : 1 /X

10

8

6

4

2

0

20

15

10

5

0

Area

Rat

io

10

8

6

4

2

0

Area

Rat

io

Area

Rat

ioAr

ea R

atio

NBAMOZY = 0 .1 9 2 4 3 5 + 1 .9 2 7 5 7 *X R ^2 = 0 .9 9 8 2 W : 1 /X

0 2 4 6 8 10 0 2 4 6 8 10

0 2 4 6 8 10

Figure 3: Seven-point calibration curves of nitrofuran metabolite, from 0.025 ng/mL (equivalent to 0.05 ng/mL after sample preparation) to 10 ng/mL

FortificationLevel (µg/kg) AHD AOZ SEM AMOZ

0.05 82 ± 13% 110 ± 22% 89 ± 15% 98 ± 14%0.5 88 ± 4% 100 ± 11% 100 ± 11% 95 ± 6%2.5 109 ± 3% 86 ± 18% 79 ± 19% 100 ± 3%

Table 2: Mean recovery values (n=3) of crawfish samples fortifiedat three levelsNote: values given after ± are standard deviations.

AN62492_E 11/07S











www.thermo.com


Administrator

Text Box


Highly Selective Detection and Identification ofNitrofuran Metabolites in Honey Using LC-MS/MSEduardo Matus,1 Jean-Jacques Dunyach,2 Alejandro Albornoz3;1Food Science Laboratories, Buenos Aires, Argentina; 2Thermo Fisher Scientific, San Jose, CA; 3Eidomet, Buenos Aires, Argentina


Key Words


• LC-MS/MS

• QuantitativeAnalysis

Introduction

Nitrofurans are broad-spectrum antibiotics used to treatbees and other animals with bacterial infections. As aresult of dosing bees with these antibiotics their metabo-lites are sometimes found in honey. Female rats givennitrofurans in both low and high doses have exhibitedincreased incidences of ovarian granulose cell tumors.In the same study, newborn mice showed an increasedincidence of pulmonary papillary adenomas.2 As a result,nitrofurans have been banned from use in food-producinganimals in Australia (1993), the European Union (EU)(1995), The Philippines (2001), the United States (2002),Brazil (2002), Thailand (2002), and other countries.

Several studies have shown that animals rapidlymetabolize nitrofurans within a few hours so detectionhas focused on the metabolites rather than the nativedrug.3 The metabolites accumulate in tissue where they arestable and can be analyzed long after the nitrofurans havebeen administered. The EU has established a harmonizedminimum required performance limit (MRPL) for thedetection of residues of nitrofurans at one part per billion(ppb). Some European laboratories have been working toa detection limit of 0.3 ppb for several of the nitrofuranmetabolites.4 The EU recently tightened its inspection policyfor food imports after nitrofuran residues were foundin shrimp, fish, and poultry imports. This significantlyreduced the volume of those imports.

As a result, food exporting countries are required todetect nitrofuran metabolites at very low levels. There areseveral challenges that must be overcome. The first is thathoney, as well as other food products, provides a complexmatrix which increases the difficulty of sample prepara-tion. Second, efficient chromatography is critical in orderto provide good separation of the various metabolitesfrom each other and any contaminants that might bepresent. The third and most important requirement is avery high level of sensitivity and linearity in the massspectrometer in order to achieve the required high levelsof accuracy in quantifying the metabolites. This notedescribes LC-MS methods developed on the ThermoScientific TSQ Quantum Discovery by the Food ScienceLaboratories and Eidomet in Argentina in cooperationwith Thermo Fisher Scientific. The method exceeds allcurrent detection limits as set by the EU.

Goal

To demonstrate the ability to accurately quantitate nitro -furan metabolites at levels as low as 0.3 ppb in a matrixconsisting of honey using the TSQ Quantum Discovery.


In this study, 2 grams of honey samples were treatedwith four nitrofuran metabolites, AOZ, AMOZ, SEM,and AHD.5 An aliquot of honey was dissolved in 125 mMHCl and derivatized with 2-nitrobenzaldehyde and themixture was shaken for 3 minutes. The slurry was thenincubated at 37°C in a water bath for 17 hours. Themixture was then cooled to room temperature and neu-tralized by adding potassium phosphate to adjust the pHto approximately 7.0. Ethyl acetate was added to theslurry and it was hand shaken for 2 minutes and centri -fuged for 15 minutes. The organic phase was collectedinto a tube, water added, and the mixture centrifuged.The supernatant was evaporated to dryness under astream of nitrogen. The dry residue was reconstitutedwith water and injected into a filter cartridge. The residuewas then washed with water and eluted with hexane, thenanalyzed by LC-MS/MS.

HPLC was performed on a Thermo Scientific SurveyorMS Pump with a Thermo Scientific Surveyor Autosampler.A 100 × 2.1 mm, 3 µm HPLC column was used. Themobile phase consisted of A (water containing 0.05%acetic acid) and B (methanol containing 0.05% aceticacid). The gradient program was as follows: 0-3.0 min.90% A 10% B; 3.0-5.0 min. 85% A 15% B; 5.0 to 10.0min. 75% A 25% B; 10.0-15.0 min. 70% A 30% B; 15.0-17 min. 65% A 35% B; 15-17 min. 65% A 35% B; 17.0-21.0 min. 60% A 40% B; and 21.0-25.0 min. 90% A10% B.

Sample analysis was performed on a TSQ QuantumDiscovery mass spectrometer. The 0-13.4 min segmenteluted AMOZ and d5-AMOZ while the 13.4 to 25 minsegment eluted AOZ, d4-AOZ, SEM, and AHD. Sampleswere analyzed using positive electrospray ionization (ESI)in SRM mode. The scan width was 0.002 m/z and thescan time was 0.1 second. A peak width of 0.7 FWHMwas used in both Q1 and Q3. Argon was used as thecollision gas at a pressure of 1.5 mTorr.

Administrator

Text Box


of 6

Table 1: Transition reactions for MS/MS

The nitrofuran metabolites were quantified with fivecalibration standards at nominal concentrations of 0.63ppb, 1.04 ppb, 2.09 ppb, 4.17 ppb, and 8.34 ppb. Thearea ratio of the analyte versus the quality control (QC)samples was plotted against the standard concentrationratio. The linearity of the MS response was determinedby calculating the relative standard deviation (RSD) of

the average of results from a series of injections at a singleconcentration. The method is generally considered to bevalidated if the RSD is less than 15%. The recovery ratiowas also calculated by injecting a known amount ofsample and comparing it with the calculated amountdelivered by the detector.


Tables 2 through 5 report the results obtained with thestandard and QC samples in honey for each metabolite.Note that for clarity purposes, all areas reported in thetables are divided by 1000. The concentrations of theQC samples were calculated by comparing the area to thestandards. Then the relative standard deviation for eachset of QC samples was calculated. The RSD for AOZranged from 6.7% at 0.3 ppb to 2.7% at 4 ppb. The RSDfor AMOZ ranged from 3.60% at 0.3 ppb to 2.50% at4 ppb. The RSD for AHD ranged from 9.0% at 0.3 ppbto 2.9% at 4 ppb. The RSD for SEM ranged from 8.3%at 0.3 ppb to 4.2% at 4 ppb.

IDENT. AREA AREA AREA SPECIFIED CALCULATED Diff RSD RECOVERYLEVEL AOZ ISTD RATI CONC. CONC. % % %

(d4-AOZ) (ppb) (ppb)

QC-4ppb 1609.6 356.5 4.52 4.172 4.204 0.77% 100.8QC-4ppb 1728.1 381.1 4.53 4.172 4.222 1.20% 101.2QC-4ppb 1849.5 392.6 4.71 4.172 4.253 1.94% 2.7% 101.9QC-4ppb 1743.8 378.7 4.60 4.172 4.155 -0.41% 99.6QC-4ppb 1919.4 389.7 4.93 4.172 4.451 6.69% 106.7QC-2ppb 864.7 372.9 2.32 2.086 2.111 1.20% 101.2QC-2ppb 912.9 370.1 2.47 2.086 2.252 7.96% 108.0QC-2ppb 789.1 358.8 2.20 2.086 1.938 -7.09% 5.5% 92.9QC-2ppb 924.2 377.7 2.45 2.086 2.166 3.84% 103.8QC-2ppb 912.2 375.5 2.43 2.086 2.150 3.07% 103.1QC-1ppb 436.6 356.7 1.22 1.043 1.068 2.40% 102.4QC-1ppb 466.2 390.9 1.19 1.043 1.038 -0.48% 99.5QC-1ppb 431.3 359.3 1.20 1.043 1.017 -2.49% 4.7% 97.5QC-1ppb 477.7 358.6 1.33 1.043 1.138 9.11% 109.1QC-1ppb 451.5 346.4 1.30 1.043 1.112 6.62% 106.6

QC-0.5ppb 258.3 376.1 0.69 0.521 0.556 6.72% 106.7QC-0.5ppb 271.1 388.1 0.70 0.521 0.567 8.83% 108.8QC-0.5ppb 260.2 366.4 0.71 0.521 0.565 8.45% 7.4% 108.4QC-0.5ppb 228.8 372.2 0.61 0.521 0.477 -8.45% 91.6QC-0.5ppb 249.0 380.6 0.65 0.521 0.513 -1.54% 98.5QC-0.3ppb 162.6 357.3 0.46 0.313 0.335 7.03% 107.0QC-0.3ppb 146.9 369.5 0.40 0.313 0.280 -10.54% 89.5QC-0.3ppb 145.5 341.0 0.43 0.313 0.304 -2.88% 6.7% 97.1QC-0.3ppb 171.5 412.6 0.42 0.313 0.293 -6.39% 93.6QC-0.3ppb 156.0 369.4 0.42 0.313 0.300 -4.15% 95.8

IDENT. NOMINAL AREA AREA AREALEVEL CONC. AOZ ISTD RATIO

(ppb) (d4-AOZ)Std 0,6 ppb 0.63 217.3 559 0.39Std 1 ppb 1.04 344.1 524.3 0.66Std 2 ppb 2.09 671.2 532.3 1.26Std 4 ppb 4.17 1286.3 567.2 2.27Std 8 ppb 8.34 2601.2 580.6 4.48

EquationY = 0.5424 X + 0.0973 R2 = 0.9998

Table 2: AOZ data

Precursor Product CollisionAnalyte Ion Ion Energy

AMOZ 335 262 15335 291 10

d5-AMOZ 340 296 10AOZ 236 78 15

236 134 8d4-AOZ 240 134 8SEM 209 166 6

209 192 5AHD 249 104 16

249 134 7

Administrator

Text Box


of 6

IDENT. AREA AREA AREA SPECIFIED CALCULATED Diff RSD RECOVERYLEVEL AMOZ ISTD RATIO CONC. CONC. % % %

(d5-AMOZ) (ppb) (ppb)

QC-4ppb 8814.0 1700.1 5.18 4.000 4.200 5.00% 105.0QC-4ppb 6866.9 1376.6 4.99 4.000 4.040 1.00% 101.0QC-4ppb 9431.2 1894.0 4.98 4.000 3.938 -1.55% 2.50% 98.5QC-4ppb 7438.6 1464.6 5.08 4.000 4.017 0.43% 100.4QC-4ppb 7235.0 1439.7 5.03 4.000 3.974 -0.65% 99.4QC-2ppb 3117.9 1246.4 2.50 2.000 2.000 0.00% 100.0QC-2ppb 3951.3 1511.5 2.61 2.000 2.100 5.00% 105.0QC-2ppb 3617.4 1477.9 2.45 2.000 1.914 -4.30% 3.45% 95.7QC-2ppb 3214.0 1263.0 2.54 2.000 1.991 -0.45% 99.6QC-2ppb 4164.8 1662.7 2.50 2.000 1.959 -2.05% 98.0QC-1ppb 1742.6 1357.0 1.28 1.000 1.010 1.00% 101.0QC-1ppb 2178.9 1687.2 1.29 1.000 1.020 2.00% 102.0QC-1ppb 1864.6 1476.0 1.26 1.000 0.967 -3.30% 2.90% 96.7QC-1ppb 2307.9 1853.7 1.25 1.000 0.952 -4.80% 95.2QC-1ppb 2281.8 1781.8 1.28 1.000 0.981 -1.90% 98.1

QC-0.5ppb 1153.4 1710.4 0.67 0.500 0.517 3.40% 103.4QC-0.5ppb 850.2 1345.0 0.63 0.500 0.482 -3.60% 96.4QC-0.5ppb 1067.5 1602.7 0.67 0.500 0.489 -2.20% 4.63% 97.8QC-0.5ppb 830.8 1331.6 0.62 0.500 0.455 -9.00% 91.0QC-0.5ppb 1067.2 1640.8 0.65 0.500 0.477 -4.60% 95.4QC-0.3ppb 548.5 1289.1 0.43 0.300 0.314 4.67% 104.7QC-0.3ppb 485.0 1139.9 0.43 0.300 0.314 4.67% 104.7QC-0.3ppb 559.5 1309.3 0.43 0.300 0.298 -0.67% 3.60% 99.3QC-0.3ppb 727.5 1719.1 0.42 0.300 0.295 -1.67% 98.3QC-0.3ppb 529.3 1264.3 0.42 0.300 0.291 -3.00% 97.0

IDENT. NOMINAL AREA AREA AREALEVEL CONC. AMOZ ISTD RATIO

(d5-AMOZ)Std 0,6 ppb 0.6 760.3 1799.0 0.40Std 1 ppb 1.0 1320.8 1936.8 0.69Std 2 ppb 2.0 2811.8 2170.0 1.37Std 4 ppb 4.0 5582.3 2166.6 2.68Std 8 ppb 8.0 11265.8 2231.2 5.32

EquationY = 0.6123 X + 0.0413 R2 = 1.0000

(ppb)

IDENT. AREA AREA AREA SPECIFIED CALCULATED Diff RSD RECOVERYLEVEL AHD ISTD RATIO CONC. CONC. % % %


QC-4ppb 436.6 356.5 1.22 4.063 3.585 -11.76% 88.2QC-4ppb 461.5 381.1 1.21 4.063 3.543 -12.80% 87.2QC-4ppb 397.0 392.6 1.01 4.063 3.559 -12.40% 2.9% 87.6QC-4ppb 408.3 378.7 1.08 4.063 3.799 -6.50% 93.5QC-4ppb 402.6 389.7 1.03 4.063 3.638 -10.46% 89.5QC-2ppb 222.6 372.9 0.60 2.034 1.683 -17.26% 82.7QC-2ppb 224.4 370.1 0.61 2.034 1.711 -15.88% 84.1QC-2ppb 179.1 358.8 0.50 2.034 1.726 -15.14% 2.6% 84.9QC-2ppb 194.8 377.7 0.52 2.034 1.785 -12.24% 87.8QC-2ppb 181.8 375.5 0.48 2.034 1.672 -17.80% 82.2QC-1ppb 123.0 356.7 0.34 1.017 0.919 -9.64% 90.4QC-1ppb 126.8 390.9 0.32 1.017 0.857 -15.73% 84.3QC-1ppb 100.6 359.3 0.28 1.017 0.941 -7.47% 5.5% 92.5QC-1ppb 90.6 358.6 0.25 1.017 0.843 -17.11% 82.9QC-1ppb 98.3 346.4 0.28 1.017 0.954 -6.19% 93.8

QC-0.5ppb 78.7 376.1 0.21 0.508 0.509 0.20% 100.2QC-0.5ppb 70.6 388.1 0.18 0.508 0.425 -16.34% 83.7QC-0.5ppb 57.7 366.4 0.16 0.508 0.502 -1.18% 9.2% 98.8QC-0.5ppb 50.8 372.2 0.14 0.508 0.427 -15.94% 84.1QC-0.5ppb 52.6 380.6 0.14 0.508 0.433 -14.76% 85.2QC-0.3ppb 44.5 357.3 0.12 0.305 0.251 -17.70% 82.3QC-0.3ppb 45.8 369.5 0.12 0.305 0.249 -18.36% 81.6QC-0.3ppb 35.4 341.0 0.10 0.305 0.309 1.31% 9.0% 101.3QC-0.3ppb 39.6 412.6 0.10 0.305 0.281 -7.87% 92.1QC-0.3ppb 35.1 369.4 0.10 0.305 0.278 -8.85% 91.1

IDENT. NOMINAL AREA AREA AREALEVEL CONC. AHD ISTD RATIO

( ppb ) (d4-AOZ)Std 0,6 ppb 0.61 70.9 559.0 0.127Std 1 ppb 1.02 112.2 524.3 0.210Std 2 ppb 2.03 200.3 532.3 0.380Std 4 ppb 4.06 416.4 567.2 0.730Std 8 ppb 8.13 797.2 580.6 1.370

EquationY = 0.1396 X + 0.0174 R2 = 0.9955

Table 3: AMOZ data

Table 4: AHD data

Administrator

Text Box


of 6

Figure 1 shows the chromatograms of a negative and apositive unknown sample set in which the AOZ metaboliteis clearly identified at the 0.6 ppb level. Table 6 summarizesthe average method results for the four metabolites overmultiple sample sets. The limits of detection (LODs) and

limits of quantification (LOQs) are reported and thedata shows good accuracy at the LOQ levels for all themetabolites. The LODs and LOQs achieved on the fournitrofuran metabolites are all at sub ppb levels.

IDENT. AREA AREA AREA SPECIFIED CALCULATED Diff RSD RECOVERYLEVEL AHD ISTD RATIO CONC. CONC. % % %


QC-4ppb 436.6 356.5 1.22 4.063 3.585 -11.76% 88.2QC-4ppb 461.5 381.1 1.21 4.063 3.543 -12.80% 87.2QC-4ppb 397.0 392.6 1.01 4.063 3.559 -12.40% 2.9% 87.6QC-4ppb 408.3 378.7 1.08 4.063 3.799 -6.50% 93.5QC-4ppb 402.6 389.7 1.03 4.063 3.638 -10.46% 89.5QC-2ppb 222.6 372.9 0.60 2.034 1.683 -17.26% 82.7QC-2ppb 224.4 370.1 0.61 2.034 1.711 -15.88% 84.1QC-2ppb 179.1 358.8 0.50 2.034 1.726 -15.14% 2.6% 84.9QC-2ppb 194.8 377.7 0.52 2.034 1.785 -12.24% 87.8QC-2ppb 181.8 375.5 0.48 2.034 1.672 -17.80% 82.2QC-1ppb 123.0 356.7 0.34 1.017 0.919 -9.64% 90.4QC-1ppb 126.8 390.9 0.32 1.017 0.857 -15.73% 84.3QC-1ppb 100.6 359.3 0.28 1.017 0.941 -7.47% 5.5% 92.5QC-1ppb 90.6 358.6 0.25 1.017 0.843 -17.11% 82.9QC-1ppb 98.3 346.4 0.28 1.017 0.954 -6.19% 93.8

QC-0.5ppb 78.7 376.1 0.21 0.508 0.509 0.20% 100.2QC-0.5ppb 70.6 388.1 0.18 0.508 0.425 -16.34% 83.7QC-0.5ppb 57.7 366.4 0.16 0.508 0.502 -1.18% 9.2% 98.8QC-0.5ppb 50.8 372.2 0.14 0.508 0.427 -15.94% 84.1QC-0.5ppb 52.6 380.6 0.14 0.508 0.433 -14.76% 85.2QC-0.3ppb 44.5 357.3 0.12 0.305 0.251 -17.70% 82.3QC-0.3ppb 45.8 369.5 0.12 0.305 0.249 -18.36% 81.6QC-0.3ppb 35.4 341.0 0.10 0.305 0.309 1.31% 9.0% 101.3QC-0.3ppb 39.6 412.6 0.10 0.305 0.281 -7.87% 92.1QC-0.3ppb 35.1 369.4 0.10 0.305 0.278 -8.85% 91.1

IDENT. NOMINAL AREA AREA AREALEVEL CONC. AHD ISTD RATIO

( ppb ) (d4-AOZ)Std 0,6 ppb 0.61 70.9 559.0 0.127Std 1 ppb 1.02 112.2 524.3 0.210Std 2 ppb 2.03 200.3 532.3 0.380Std 4 ppb 4.06 416.4 567.2 0.730Std 8 ppb 8.13 797.2 580.6 1.370

EquationY = 0.1396 X + 0.0174 R2 = 0.9955

Figure 1: Identification of AOZ metabolite in honey at 0.6 ppb level

Table 5: SEM data

Time (min) Time (min)

Administrator

Text Box


of 6

Conclusion

An LC-MS/MS assay to detect and identify nitrofuranmetabolites was developed using the TSQ Discovery. Theextraction method appears to be extremely robust andreliable with good recovery efficiency (better than 80%),allowing unambiguous routine identification and quantifi-cation of all nitrofuran metabolites in honey. The LC-MS/MS-based method described here provides high speed,excellent sensitivity, and specificity of detection. The assaydemonstrated the ability to easily meet the 0.3 ppb limitof quantitation that is required by the most stringentcurrent requirements of food monitoring applicationsoperating under FDA and EC regulations.

References1 Proceedings of the National Academy of Sciences of the United States of

America. Vol 73 No 10. pp 3386-3390. Nitrofurans, a Group of SyntheticAntibiotics, with a New Mode of Action: Discrimination of SpecificMessenger RNA Classes. Peter Herrlich, Manfred Schweiger

2 NTP (1988). Toxicology and carcinogenesis studies of nitrofurazone (CASNo. 59-87-0) in F344/N rats and B6C3F1 mice. Technical Report SeriesNo. 337. US Department of Health and Human Services, Public HealthService/National Institutes of Health National Toxicology Program.

3 Journal of Chromatography. 1997 Vol 691 pp 87-94. Determination of thefurazolidone metabolite, 3-amino-2-oxazolidinone in porcine tissues usingliquid chromatography-thermospray mass spectrometry and the occurrenceof residues in pigs produced in Northern Ireland. R.J. McCracken, D.G.Kennedy.

4 Food Standards Agency. June 30, 2003. Reporting limits for nitrofuran andchloramphenicol residues harmonized.

5 Determination of Nitrofurans Residues in Honey by LC-MS/MS, AnalysisMethod, Food Science Laboratories, 2005.

LOD LOQ ANALYTICAL RANGE % RECOVERY RANGE % REC CVANALYTE MATRIX(ppb) (ppb) (ppb) %

INCERTITUDE %

AMOZ Honey 0.04 0.09 0,04-4,0 99.3 88,8-109.8 3.5 7.1

AHD Honey 0.06 0.16 0,06-4,063 88.5 71.2-105.8 6.5 13.0

AOZ Honey 0.06 0.14 0,06-4,172 101.2 84.2-118.5 5.6 11.3

SEM Honey 0.08 0.18 0,08-4,064 94.3 70.6-117.8 8.3 16.6

Table 6: Summary of Method Results

Administrator

Text Box


Laboratory Solutions Backedby Worldwide Service and Support

Tap our expertise throughout the life of your instrument. As an industry

leader in analytical instruments, Thermo extends its support throughout

our worldwide network of Thermo-trained and certified engineers who

are experts in laboratory technologies and applications. Put our team

of experts to work for you in a range of disciplines, from system installa-

tion, training and technical support, to complete asset management and

regulatory compliance consulting. Improve your productivity and lower

the cost of instrument ownership through our product support services.

Maximize uptime while eliminating the uncontrollable cost of unplanned

maintenance and repairs. When it’s time to enhance your systems,Thermo

also offers certified parts and a range of accessories and consumables

suited to your application.


visit our Web site at www.thermo.com.

Laboratory Solutions Backedby Worldwide Service and Support

AN62490_E 10/07S










www.thermo.com



Administrator

Text Box


LC-MS/MS Analysis of Malachite Green,Leucomalachite Green, Ciprofloxacin, andTetracycline in Food Samples using aTurboFlow MethodCharles Yang and Dipankar Ghosh, Thermo Fisher Scientific, San Jose, CA, USA


Key Words

• Aria TLX System


• TSQ QuantumAccess

• TurboFlowTechnology

• Veterinary Drugs

Introduction

Accurate monitoring of chemical residue levels in foodand agriculture products is essential to assure the safety of the food supply and manage global health risks. Theanalysis of chemical residues requires techniques sensitiveenough to detect and quantify contaminants at or belowthe maximum residue limit (MRL) of the compound in a given sample matrix. In addition, because of increasedfood safety regulations and the growing numbers of samples to be analyzed, it is critical that the analyticaltechniques provide high sample throughput.

With the continuing rapid growth of the aquacultureindustry, there is increasing concern about the use ofunapproved drugs and unsafe chemicals in aquafarmingoperations. Malachite green (MG), a triphenylmethanedye, is an effective and inexpensive fungicide used inaquaculture, particularly in Asian countries (Figure 1).

During metabolism, malachite green reduces to leucomala-chite green (LMG), which has been shown to accumulate infatty fish tissues and can be found long after MG may nolonger be detected.1 Both MG and LMG have demonstratedputative carcinogenic activity, and thus MG has been bannedfor use as an aquaculture veterinary drug in many countriesincluding the United States and Canada, as well as theEuropean Union (EU). For substances that are bannedfrom use in food producing animals, EU legislation definesminimum required performance limits. For malachite green,an analytical test method must be able to determine thesum of MG and LMG residues in fish muscle at the minimum required performance limit of 2 µg/kg (ppb).2

Ciprofloxacin is a broad-spectrum antibiotic belongingto the fluoroquinolone group (Figure 2). Fluoroquinoloneshave been shown to be very effective in combating variousdiseases in animal husbandry and aquaculture and areused extensively worldwide. However, because of concernsthat fluoroquinolone residues in food products may lead

to the development ofantibacterial resistance to these drugs in humans,the FDA has prohibitedextra-label use of fluoro-quinolones in food animals.3 According to the EU legislation on veterinary drug residues, the maximumresidue limits for the sum of enrofloxacin and its metaboliteciprofloxacin are 100 µg/kg (ppb) in muscle for all foodproducing species and 200 µg/kg (ppb) in pork liver.4

Tetracycline is a polyketide antibiotic that is highly effective against a number of gram-positive and gram-negative bacteria (Figure 3). As with other veterinary antibiotics, when tetracycline is used in food animals, it hasthe potential to generate drugresidues in the animals and animal products which can lead to increases in microbialresistance. The MRLs for tetracycline are 100 µg/kg (ppb)in muscle and 300 µg/kg (ppb)in liver for all food producingspecies.5

Here we show a solution that combines the ThermoScientific Aria TLX system utilizing TurboFlow technologywith a Thermo Scientific TSQ Quantum Access mass spectrometer. Compared to traditional offline extractionmethods, this solution provides fast and reliable sampleanalysis of chemical residues in food by online sampleextraction followed by LC-MS/MS. The Aria TLX systemuses TurboFlow technology to retain small molecules and filter out proteins and larger materials by diffusion,size exclusion, and column chemistry. This enables usersto directly inject samples into the LC-MS system foranalysis, greatly simplifying sample preparation andincreasing throughput.

Goal

To demonstrate a reduction in overall analytical time compared to traditional methods, such as liquid-liquid or solid phase extraction, while also minimizing ion suppression and matrix interference in food samples.

Figure 1: Structures of malachite green and leucomalachite green

Figure 2: Structure of Ciprofloxacin

Figure 3: Structure of Tetracycline

Administrator

Text Box



Sample Preparation

Samples were prepared for analysis using a simple extractionprocedure with acetonitrile. Individual samples, weighing3 to 4 g, of shrimp (with head attached), tilapia, and pigliver were homogenized with approximately 30 mL of ace-tonitrile and then left at room temperature for 10 minutes.The liquid phase was then aspirated from the sample intoa 0.22 µm pore centrifuge tube and spun at 10,000 rpmfor approximately 10 minutes. The supernatant was aspirated into a 50 mL scintillation vial. This portion of the sample preparation took 25 minutes.

A stock mix solution of malachite green, leucomala-chite green, ciprofloxacin, and tetracycline was preparedat a concentration of 1 mg/L. All four analytes weremixed in one vial at 0.1 mg/mL (1000 µg/mL) in methanol.Calibration solutions in the concentration range 10 µg/kgto 5 ng/kg were prepared by serial dilution of the stocksolution into the three sample matrices. The total samplepreparation time was approximately 30 to 40 minutes.

TurboFlow Method Conditions

The samples were processed on the Thermo Scientific AriaTLX-1 System. The Multiple Column Module (MCM),which allows 6 loading columns and 6 analytical columnsto be tested at once, was used to facilitate method devel-opment. First, the loading column that gave the bestrecovery of each analyte was selected. Because all of theanalytes were in one stock solution, the run time was min-imized. Then the analytical column that gave the best per-formance was selected. The final TurboFlow method con-ditions were as follows:

Loading Column: Thermo Scientific TurboFlow XL C18 column Analytical Column: Thermo Scientific Hypersil GOLD

50 x 3 mm 5 µm columnMobile Phase A: 0.1% formic acid in waterMobile Phase B: 0.1% formic acid in acetonitrileMobile Phase D: 20% acetone/40% methanol/40% acetonitrileAutosampler Injection Size: 10 µLSample Extraction Solution: 50:50 (A/B)

The Aria TLX TurboFlow method is shown in Figure 4.The analytical run was completed in less than 6 minutes.

Figure 4: Aria TLX method

Administrator

Text Box


MS Conditions

MS analysis was carried out on a Thermo Scientific TSQ QuantumAccess triple stage quadrupole mass spectrometer. The MS conditionswere as follows:

Ion Source Polarity: Positive ion modeSpray Voltage: 3500 VVaporizer Temperature: 472 °CSheath Gas Pressure (N2): 40 arbAuxiliary Gas Pressure (N2): 50 arbIon Transfer Tube Temperature: 270 °CSkimmer Offset: 5 VCollision Gas (Ar): 1.5 mTorrQ1/Q3 Peak Resolution: 0.7 u (unit mass resolution)Scan Mode: Selected Reaction Monitoring

The MS method is shown in Figure 5.


As the international food trade continues to grow, so does the needfor careful monitoring of the food supply to ensure that levels of drug residues and other chemical contaminants are below establishedstandards. In this study, ciprofloxacin, tetracycline, and malachitegreen/leucomalachite green were examined in several food matrices.

LC-MS/MS assays of chemical residues in food matrices typicallyrequire extensive sample preparation prior to analysis, which can betime consuming and expensive. The Aria TLX system with TurboFlowtechnology eliminates the need for lengthy sample preparation steps.In this study, each sample was centrifuged once to clean out anyfloating particles and then immediately injected onto the column.Most of the time-consuming steps in the sample preparation processwere removed, which increased sample throughput and helped tominimize errors and variability.

The Aria TLX system software allows both HPLC and TurboFlowmethods to be run on a single system, injection to injection. To evaluate the differences in performance between a standard HPLC

Figure 5: MS method showing the SRM transitions that were monitored

Administrator

Text Box


method and a TurboFlow method, each sample was analyzedby both methods. In the standard HPLC method, only theanalytical HPLC column was used. In the TurboFlowmethod, the TurboFlow column and the analytical columnwere used. Figure 6 compares representative standardHPLC and TurboFlow method chromatograms of 500 ng/kg(parts per trillion) tetracycline in the fish matrix. The

TurboFlow method chromatogram shows that interferencespresent in the standard HPLC chromatogram have beenremoved. The turbulent flow properties successfullyremove matrix interferences that cause ion suppression.

Similar results were observed in the analysis of drugsin other food matrices. Figure 7 compares representativestandard HPLC and TurboFlow method chromatograms

Figure 6: Chromatogram comparison of tetracycline at 500 ng/kg in fish (tilapia) matrix in standard HPLC and TurboFlow method

Figure 7: Chromatogram comparison of LMG at 500 ng/kg and 50 ng/kg in shrimp matrix in standard HPLC and TurboFlow method

Administrator

Text Box


for leucomalachite green at both 500 ng/kg (ppt) and 50 ng/kg (ppt) in the shrimp matrix. The TurboFlowmethod chromatograms show improved signal-to-noiseratios and significant reduction in ion suppression. The most dramatic results are shown in the analysis of tetracycline at 500 ng/kg (ppt) in pig liver (Figure 8).

Peak areas were used for quantitation and the resultantlinearity of responses is plotted in Figure 9 for both thestandard HPLC and the TurboFlow methods for cipro -floxacin in pig liver. Excellent R2 values were observed for the TurboFlow method. A data summary showing the improvement in the TLX system results over those of

the standard HPLC results is shown in Table 1. The R2

values for the TLX system results were all greater than0.99 for the linear regression equations (1/x weighted) inthe concentration ranges tested.

Table 1 shows the results of the assay for ciprofloxacin,MG, LMG, and tetracycline in fish, shrimp, and pork liverextracts. The limits of quantitation (LOQs) achieved forall four analytes using online extraction followed by LC-MS/MS were significantly better compared to thatachieved by standard HPLC. This indicates the removalof endogenous tissue by the Aria TLX system, thus reducingion suppression effects and increasing detection limits.

Table 1: Data summary showing the improvement in the TurboFlow method results over those of the standard HPLC results. Note that results are onlyshown for a compound in the matrix in which it would be found; for example, MG and LMG would be found in fish but not in pig liver.

Fish

Standard HPLC TurboFlow MethodLOD %RSD LOQ %RSD LOD %RSD LOQ %RSD

(µg/kg) n = 3 (µg/kg) n = 3 R2(µg/kg) n = 3 (µg/kg) n = 3 R2

Ciprofloxacin 0.5 32.0 1.0 8.0 0.9875 0.1 15.9 0.5 7.1 0.9968MG 0.1 23.0 0.5 1.4 0.9988 0.1 6.8 0.1 8.2 0.9984LMG 0.5 7.4 0.5 7.4 0.9990 0.1 12.0 0.1 12.0 0.9983

Shrimp

Ciprofloxacin 5.0 10.5 5.0 10.5 0.9580 0.5 16.0 1.0 2.0 0.9906MG 0.1 21.4 0.5 7.2 0.9991 0.05 7.9 0.05 7.9 0.9990LMG 0.1 20.5 0.5 11.0 0.9975 0.05 13.5 0.1 11.2 0.9988

Pig Liver

Ciprofloxacin 0.5 38.8 1.0 10.4 0.9707 0.1 29.0 0.5 8.6 0.9969Tetracycline 0.5 14.8 1.0 10.5 0.9932 0.1 11.5 0.5 11.3 0.9953

Figure 9: Ciprofloxacin calibration 1/x on standardHPLC vs. TurboFlow method in pig liver matrix

Figure 8: Chromatogram comparison of tetracycline at 500 ng/kg in pig liver matrix in standardHPLC and TurboFlow method

Administrator

Text Box


Conclusion

A rapid, sensitive and reliable method for the quantitationof veterinary drugs in food matrices was developed usingthe Aria TLX-1 system with the TSQ Quantum Accessmass spectrometer. Minimal sample preparation wasrequired because the TurboFlow method allows directinjection of samples into the system. The overall processingtime for analysis was significantly shortened compared tomethods using offline sample preparation. In addition, theAria TLX system reduced ion suppression and matrixeffects compared to standard HPLC runs.

References1. Plakas, S. M.; Doerge, D. R.; Turnipseed, S. B. Disposition and

Metabolism of Malachite Green and Other Therapeutic Dyes in Fish. InXenobiotics in Fish; Smith, D. J., Gingerich, W. H., Beconi-Barker, M. G.,Eds.; Plenum Press: New York City, 1999; p. 149-166.

2. European Commission, Commission Decision 2002/657/EC of 12 August2002 implementing Council Directive 96/23/EC concerning the perform-ance of analytical methods and the interpretation of results, as amendedby Decision 2003/181/EC(4), (Official Journal of the EuropeanCommunities L 221, 17.08.2002, p. 8-36).

3. U.S. Food and Drug Administration. Federal Register: May 22, 1997(Volume 62, Number 99).

4. European Parliament and Council, Regulation (EC) 1181/2002 of 1 July2002, amending Annex I of Council Regulation (EEC) No 2377/90 layingdown a Community procedure for the establishment of maximum residuelimits of veterinary medicinal products in foodstuffs of animal origin food(Official Journal of the European Communities L 172, 2.07.2002).

5. European Parliament and Council, Regulation (EC) 508/1999 of 4 March1999, amending Annex I of Council Regulation (EEC) No 2377/90 layingdown a Community procedure for the establishment of maximum residuelimits of veterinary medicinal products in foodstuffs of animal origin food(Official Journal of the European Communities L 60, 9.3.1999).








www.thermo.com

AN62883_E 10/08M



Legal Notices©2008 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific Inc. and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scientific Inc. products. It is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are available in all countries. Please consult your local sales representative for details.


Administrator

Text Box


Multi-class Antibiotic Screening of Honey Using Online Extraction with LC-MS/MSCatherine Lafontaine, Yang Shi, Francois A. Espourteille, Thermo Fisher Scientific, Franklin, MA, USA

Introduction

Antibiotics are commonly used in bee hives to controlbacterial disease in honey bees. Use of these antibioticsrequires caution to prevent persistent residues fromoccurring in food-grade honey. If antibiotic residues arepresent in sufficient quantities, allergic reactions andbacterial resistance can develop.

Many countries now monitor antibiotic residues inhoney. LC-MS/MS is currently a common analyticalapproach for the quantification of antibiotic contaminationin honey. Sample preparation for LC-MS/MS analysis can betime and labor intensive, often involving pH modification,hydrolysis, liquid-liquid extraction, solid phase extraction,solvent evaporation, and pre-concentration. A quick,comprehensive, online screening liquid chromatography (LC)method using a Thermo Scientific Aria TLX system poweredby Thermo Scientific TurboFlow technology has beendeveloped here to monitor several classes of antibiotics.

Goal

To develop a broad, generic, automated LC-MS/MSmethod for screening multi-class antibiotics in honey.

Experimental

Method Information

Residues representative of several classes of antibiotics(macrolides, sulfonamides, aminoglycosides, andtetracyclines) were extracted from wildflower honey usingbuffer containing ethylenediaminetetraacetic acid (EDTA).The extract cleanup was accomplished using a TurboFlow™

method involving two TurboFlow columns placed intandem, a mixed mode anion exchange column and apolar polymer-based column. Simple sugars were un-retainedand moved to waste during the loading step while theanalytes of interest were retained on the extractioncolumn set. This was followed by organic elution to anend-capped silica-based mixed mode reversed phaseanalytical column (Thermo Scientific BETASIL Phenyl/Hexyl)and gradient elution to a Thermo Scientific TSQ QuantumUltra triple stage quadrupole mass spectrometer with aHeated Electrospray Ionization (H-ESI) source operatingin positive selective reaction monitoring (SRM) mode. Thetotal LC-MS/MS method run time was less than 18 minutes.Positive SRM transitions and other MS parameters forindividual analytes are shown in Table 1.

Sample Preparation

A McIlvaine/0.1 M EDTA buffer was used as a 1:1 w:v(gram weight honey: milliliter volume buffer) diluent forwildflower honey, the testing matrix in this study.1 A stocksolution was prepared for sulfapyridine, sulfathiazole,tilmicosin, tylosin, oxytetracycline, and erythromycin in3:1 methanol:water at 100 µg/mL. Additionally, one wasprepared for doxycycline, demeclocycline, streptomycin,and dihydrostreptomycin in water at 100 µg/mL. Thesestocks were each spiked into 1:1 honey:buffer matrix andused as a spiking stock to make a set of calibrationstandards and quality controls (QCs). All blanks, standards,and QCs were prepared and analyzed in polypropylenevials. Injection volumes were 0.050 mL.

Aria™ TLX-1: TurboFlow Method Parameters

TurboFlow Cyclone MAX and TurboFlow Cyclone-P columns (0.5 × 50 mm), in-tandem

BETASIL Phenyl/Hexyl column, 100 × 3 mm, 3 µmAria operating system 1.6.2 software

Loading Pump Mobile Phases

Mobile Phase A: 1.0% formic acid in waterMobile Phase B: 0.1% formic acid in acetonitrileMobile Phase C: 10 mM ammonium acetate in water, pH 9Mobile Phase D: 50 mM ammonium acetate in methanol with

0.1% formic acid

Elution Pump Mobile Phases

Mobile Phase A: 1 mM NFPA*, 0.5% formic acid, 0.04% TFA** in waterMobile Phase B: 0.5% formic acid, 0.04% TFA in 1:1 methanol:acetonitrile

*NFPA is nonafluoropentanoic acid.

**TFA is trifluoroacetic acid.

Key Words

• Aria TLX-1System


• Food Testing

• TurboFlowMethod


Administrator

Text Box


TSQ Quantum Ultra™ Mass Spectrometer (MS) Parameters

Ion Polarity: Positive Ionization Source: H-ESISpray Voltage: 4000 VVaporizer Temperature: 400 °CCapillary Temperature: 370 °CSheath Gas Pressure (N2): 30 arb unitsAuxiliary Gas Pressure (N2): 60 arb unitsIon Sweep Gas Pressure (N2): 0.0 arb unitsSkimmer Offset: 5 V (for streptomycin),

0 V (for all others)Collision Pressure: 1.2 mTorrChrom Filter Peak Width: 8.0 sScan Type: SRMScan Time: 0.020 s Scan Width: 0.100 m/zPeak Width Q1 Da. (FWHM): 0.700Peak Width Q3 Da. (FWHM): 0.700


Results were packaged using Thermo Scientific LCQUAN

2.5.6 data quantitation software and included subtractionof background due to the presence of a few endogenousanalytes in the store-bought honey. Figure 1 shows arepresentative chromatogram of the 10 analytes at 100 ng/mL in 1:1 honey/buffer. Matrix-matched calibrationstandards showed linear response of two orders ofmagnitude (r2 > 0.99) for all of the analytes investigated(Table 2). All %CVs (n=3) were less than 19% for thelower limit of quantifications (LLOQ) and less than 8% forall other points of the curves. Figure 2 shows an LCQUAN™

representative linear regression using oxytetracycline as anexample. QC sample variability was determined byprocessing and analyzing three replicates of each of fourQC samples (2, 50, 100, and 500 ng/mL). All % RSDswere lower than 7% (except for erythromycin which wasbelow 15%). Data was not used for any QC level that fellbelow the analyte’s determined LLOQ.

Structural Class Analyte Precursor Ion Product Ions

Sulfonamides Sulfapyridine 250.1 156.0 (Q), 108.1 (C), 92.1 (C)Sulfathiazole 256.1 156.1 (Q), 92.0 (C), 108.1 (C)

Tetracyclines Doxycycline 445.3 154.0 (Q), 428.5 (C)Oxytetracycline 461.2 426.4Demeclocycline 465.2 448.4 (Q), 430.4 (C)

Aminoglycosides Streptomycin 582.3 263.0 (Q), 246.0 (C), 203.9 (C), 221.0 (C)Dihydrostreptomycin 584.3 262.9 (Q), 245.9 (C)

Macrolides Erythromycin 734.5 576.2 Tilmicosin 869.6 696.3Tylosin 916.5 772.3

NOTE: (Q)=Quantification Ion; (C)=Confirmation Ion.

Table 1: The 10 analytes and their positive SRM transition ions

Administrator

Text Box


Figure 2: LCQUAN view of oxytetracycline calibration curve and LLOQ (left window) vs. ULOQ (right window) chromatograms

Figure 1: Example chromatogram of 100 ng/mL calibration standard in 1:1 honey/buffer

Administrator

Text Box


Conclusion

During the honey quality monitoring process, it is alwaysan analytical challenge to deal with a large number ofantibiotics belonging to different classes. This oftenrequires multiple LC-MS methods. In this study, a novelapplication was introduced using dual online TurboFlowextraction columns with different chemistries. The resultsreveal that this design facilitates the separation andquantification of all of the representative compounds inthe complex honey matrix. Sample preparation time wasminimal, requiring only the addition of a buffer to reducesample viscosity. These factors enabled a broad screeningfor antibiotic contaminants to be performed quickly for agiven sample, thus increasing sample throughput.

Additionally, multiplexing with an Aria TLX-4 systemwould further reduce total LC-MS/MS run time four-foldand enable screening of 12 samples per hour. Future workcould involve screening a larger range of antibiotic andenvironmental contaminants and lowering detection limitsfor all analytes thus combining a screening method withaccurate quantification.

References1. Qualitative Identification of Tetracyclines in Tissues, CLG SOP No: CLG-

TET2.01, Rev 01, p. 6, United States Department of Agriculture, FoodSafety and Inspection Service, Office of Public Health Science, 9/25/03.








www.thermo.com

AN63043_E 05/09M





Dynamic Limit of PercentR2 Range* Detection Carryover

Analyte (1/x weighting) (ng/mL)** (ng/mL) (%)

Sulfapyridine 0.9980 50-500 10.0 8.95Sulfathiazole 0.9988 50-500 10.0 5.46Doxycycline 0.9990 10-500 5.0 10.80Oxytetracycline 0.9990 5-500 2.0 11.70Demeclocycline 0.9996 10-500 5.0 18.70Streptomycin 0.9960 50-500 10.0 11.60Dihydrostreptomycin 0.9980 50-500 10.0 6.47Erythromycin 0.9877 50-500 10.0 1.16Tilmicosin 0.9917 2-50 0.5 16.80Tylosin 0.9958 10-100 5.0 13.70

*Based on analysis using 8 point standard curve (ng/mL): 0.500, 2.00, 5.00, 10.0, 50.0, 100, 200, & 500.**The level of carryover was included in the determination of dynamic range (kept to 20% or less).

Table 2: Calibration curve statistics of the 10 analytes

Administrator

Text Box


On-line Enrichment HTLC/MS/MSAssayfor Multiple Classes of Antibiotics inEnvironmental Water SourcesKevin J. McHale,1 Chris Esposito,2 and Francois Espourteille2

1Thermo Fisher Scientific, Somerset, NJ, USA; 2Thermo Fisher Scientific, Franklin, MA, USA


Key Words


• Aria TLX-2

• H-SRM

• Turbulent-flowExtraction

Introduction

There is a growing concern over the presence of antibioticsin environmental sources of water. This has causedenvironmental and government labs to develop LC/MSmethods to monitor water supplies for the presence ofantibiotics.1-4 However, the low-level concentration ofantibiotics in environmental water sources often requiresextensive sample preconcentration and cleanup. Prepara -tion of water samples (100-1000 mL) prior to LC/MSanalysis, even with an “unlimited” sample volume, is timeconsuming and reduces sample throughput. This reportpresents a method that significantly decreases sampleprepa ration time by applying on-line preconcentrationand extraction in conjunction with detection using theThermo Scientific TSQ Quantum Ultra in highly-selectivereaction monitoring (H-SRM) mode for assaying anti -biotics at low pg/mL concentrations.

Goal

Develop a method to screen for antibiotics in surfacewater by applying on-line preconcentration and analyteextraction with LC/MS/MS detection.


The antibiotics assayed in this method (Table 1) werepurchased from Sigma (St. Louis, MO) and used withoutfurther purification. Stock solutions of the antibiotic stan-dards were prepared at 1.0 mg/mL in methanol and storedin amber polypropylene vials at -20°C until needed. Priorto High-Throughput HPLC (HTLC/MS/MS) analysis,water samples were prepared in 2 µg/mL Na2EDTA (aq)to inhibit binding of the tetracycline antibiotics to glasssurfaces and to metal ions in solution.1 Using the Thermo

Scientific Aria TLX-2 system, water samples in 1 mLvolumes were injected onto a TurboFlow® column withoutany further sample preparation. Targeted antibiotics werefocused and concentrated on the turbulent-flow extractioncolumn, then transferred to the analytical column. Analyteseparation was accomplished using a reverse-phasegradient prior to detection with the TSQ Quantum Ultrain highly-selective reaction monitoring (H-SRM) mode.

On-line TurboFlow ExtractionAria TLX-2

• TurboFlow Column: 0.5× 50 mm Thermo ScientificCyclone MAX

• Autosampler: CTC PAL (Leap Technologies)

• Injection Volume: 1.0 mL

• Loading Pump Mobile Phase:(A) 10 mM ammonium bicarbonate,(B) 0.5% HAc + 0.04% TFA,(C) ACN + 0.1% HCOOH

• Flow Rate: 2.0 mL/min

Liquid Chromatography• Analytical Column: 4.6× 100 mm, 3 µm

Thermo Scientific Hypersil GOLD

• Flow Rate: 1.2 mL/min

• Eluting Pump Mobile Phase:(A) 0.5% HAc + 0.04% TFA,(B) ACN + 0.5% HAc + 0.04% TFA

• Flow Split: post-column, 0.5 mL/min to ESI source

Mass Spectrometry• TSQ Quantum Ultra

• Ionization mode: Positive ion ESI

• Ion Transfer Tube Temperature: 375°C

Selective Reaction Monitoring (SRM) Parameters• Q2 Pressure: 1.5 mTorr argon

• SRM Transitions: see Table 1

• SRM Scan Time: 40 ms per transition

• Q1 Resolution: Unit (0.7 Da FWHM)and H-SRM (0.15 Da FWHM)

• Q3 Resolution: Unit (0.7 Da FWHM)

Compound Precursor m/z Product m/z ResolutionSulfamethoxazole 254 108, 156 H-SRM

Sulfamerazine 265 156, 172 H-SRMSulfamethizole 271 108, 156 H-SRMSulfamethazine 279 156, 186 H-SRM

Lincomycin 407 126, 359 H-SRMTetracycline 445 154, 410 H-SRMDoxycycline 445 321, 428 H-SRM

Chlortetracycline 479 444, 462 H-SRMDehydroerthromycin 716 158, 558 H-SRM

Erythromycin 734 158, 576 Unit H-SRMRoxithromycin 837 158, 679 Unit H-SRM

Tylosin 916 174, 772 Unit H-SRM

Table 1: List of antibiotics and SRM transitions for the HTLC/MS/MS assay

Administrator

Text Box



Details on HTLC/MS/MS Method

The 13 antibiotics studied (Table 1) were preconcentratedon a new mixed-mode TurboFlow extraction column, theCyclone™ MAX, which has reverse-phase and anionexchange characteristics. After sample loading andflushing the TurboFlow column at 2.0 mL/min to removematrix interferences (see Figure 1), valve 1 is switched toallow the extraction solvent plug (50% ACN + 0.1%HCOOH) to transfer the antibiotics to the analyticalcolumn. The organic content of the extraction solventplug, which contains the antibiotic analytes, is diluted bythe highly aqueous mobile phase of the eluting pumps at1.2 mL/min. This dynamic mixing occurs before the ana-lytical column so that the antibiotics can be effectivelyrefocused prior to the reverse-phase separation step.During the LC/MS/MS data acquisition, the TurboFlowcolumn is reconditioned for the next sample injection.

For unit SRM and H-SRM detection on the TSQQuantum Ultra, data acquisition was sectioned into fivetime segments, whereby eight SRM transitions per segmentwere employed. Two SRM transitions were monitored foreach antibiotic compound (see Table 1) for confirmationpurposes and to improve ion statistics.

Sensitivity and Calibration Data for HTLC/MS/MS Assay

Figure 2 presents the HTLC/MS/MS chromatograms ofthe 13 antibiotic standards and dehydroerythromycin attheir limits of quantitation (LOQs), which ranged from0.5-5 pg/mL, using 1 mL injections. Calibration data weregenerated from the LOQ to 100 pg/mL for the 13 antibi-otic standards in deionized water. Linear fit calibrationcurves with 1/x weighting were used for 11 of 13 anti -biotics. Sulfamethoxazole and sulfamethizole providedthe best results by employing quadratic fit calibrationcurves. All sample standard regressions yielded R2 ≥0.990(n = 4 replicates).

Spike of Antibiotics into Surface Water Sample

Figure 3 shows the results for the HTLC/MS/MS assayfor a surface water sample that was spiked at 25 pg/mLwith the antibiotic standards. Prior to this experiment,the surface water sample was screened using the describedmethod, and it was found to be devoid of the target anti -biotics. Figure 4 presents a comparative HTLC/MS/MSassay for a neat standard solution at 25 pg/mL. The sul-fonamide class showed a minor signal suppression in thesurface water sample, while the response for the macrolideswere slightly enhanced. The tetracyclines, however, showeda significant difference in response in the surface watersample vis-à-vis the neat standard. This difference may beattributed to binding of these tetracycline anti biotics toresidual metals in the water sample.

Figure 1: Schematic of the Aria TLX-2 system coupled to the TSQ Quantum Ultra

oc ul nm

A B C D

LoadingPumps

Auto-sampler

ElutingPumps

HTLCTurboFlow®

Valve 1

Valve 2

TSQQuantum Ultra

AnalyticalColumn

Waste

TransferLoop

analytes retainedmatrix to waste

TurboFlow Column

Figure 2: Chromatograms for antibiotics at their LOQs using HTLC/MS/MS

6 7 8 9 10Time (min) Time (min)

Inte

nsity

Inte

nsity

Sulfamethazine0.5 pg/mL

Sulfamethizole0.5 pg/mL

Oxytetracycline5 pg/mL

Chlortetracycline5 pg/mL

8 9 10 11 12

Tetracycline2.5 pg/mL

10000

20000

15000

10000

30000

20000

10000

30000

20000

10000

30000

40000

20000

10000

5000

15000

10000

5000

1500

1000

500

15000

10000

5000

5000

10000

5000

10000

5000

10000

5000

6000

4000

2000

60008000

40002000

60008000

40002000

AA: 76974

AA: 53467

AA: 45434

AA: 120532

AA: 75164

AA: 42810

AA: 32106

AA: 33060

Sulfamerazine0.5 pg/mL

Doxycycline2.5 pg/mL

Lincomycin1 pg/mL

Sulfadimethoxine1 pg/mL

Erythromycin0.5 pg/mL

Tylosin2.5 pg/mL

Roxithromycin2.5 pg/mL

Dehydroerythomycin

AA: 34513

AA: 115192

AA: 10458

AA: 51026

AA: 5359

AA: 3185

AA: 51325

Sulfamethoxazole1 pg/mL

Administrator

Text Box


Figure 3: 25 pg/mL antibiotics spiked into a surface water sample

6 7 8 9 10Time (min) Time (min)

8 9 10 11 12

Sulfamethazine

Sulfamethizole

Oxytetracycline

Chlortetracycline

Tetracycline

AA: 2259422 AA: 77977

AA: 75565

AA: 399786

AA: 3193035

AA: 1501694

AA: 18567

AA: 27617

AA: 25406

SulfamerazineDoxycycline

Lincomycin

Sulfadimethoxine

Erythromycin

Tylosin

Roxithromycin

Dehydroerythomycin

AA: 408175

AA: 1748334

AA: 816634

AA: 959990

AA: 166426

AA: 134268

AA: 2440704

Sulfamethoxazole

Inte

nsity

Inte

nsity

20000

40000

20000

20000

10000

400000

200000

20000

10000150000200000

10000050000

100000

200000

30000

40000

10000

20000

400000

600000

200000

10000

100000

50000

8000

6000

4000

2000

60000

40000

20000

3000

4000

2000

1000

10000

5000

Figure 4: Neat 25 pg/mL antibiotics standard

6 7 8 9 10Time (min)

Sulfamethazine

Sulfamethizole

Oxytetracycline

Chlortetracycline

Tetracycline

AA: 2930945 AA: 597641

AA: 1036150

AA: 4017893

AA: 2886119

AA: 207609

AA: 325303

AA: 145222

Sulfamerazine Doxycycline

Lincomycin

Sulfadimethoxine

Erythromycin

Tylosin

Roxithromycin

Dehydroerythomycin

AA: 763241

AA: 2674376

AA: 541236

AA: 663101

AA: 73328AA: 66932

AA: 1001733

Sulfamethoxazole

Inte

nsity

Inte

nsity

400000

300000

200000

100000

600000

400000

200000

20000

10000

400000

200000

20000

10000

400000

200000

100000

200000

30000

40000

10000

20000

400000

600000

200000

100000

50000

150000

100000

50000

30000

40000

20000

10000

30000

40000

60000

40000

20000

20000

10000

Time (min)8 9 10 11 12

Administrator

Text Box


HTLC/MS/MS Screening for Antibioticsin Environmental Water Samples

Surface water samples from multiple locations inCalifornia, Florida and Ontario were screened for anti -biotics using the described HTLC/MS/MS method. Ofthe water samples screened only the Lake Ontario sampleshowed measurable levels of the targeted antibiotics(Figure 5). Insets in Figure 5 show chromatographic tracesfor the two monitored SRM transitions for the observedantibiotics, providing additional confirmation and higherconfidence in these results.

Conclusions

A method for assaying antibiotics in water samples at thelow pg/mL level using on-line sample clean-up and pre-concentration has been demonstrated. The capability ofon-line turbulent-flow extraction of large sample volumes(1 mL) significantly reduces sample analysis time from amatter of hours to a matter of minutes. Detection usinghighly-selective reaction monitoring (H-SRM) provides anadditional level of selectivity and confidence over unit res-olution SRM. This HTLC/MS/MS method, when appliedto screening surface water samples, was able to detect andquantitate the observed antibiotics at the low pg/mL level.

Figure 5: Chromatograms of the detected antibiotics in a Lake Ontario water sample

Time (min)

Inte

nsity

20000

15000

10000

10000

6000

2000

10000

6000

2000

10000

6000

2000

5000

AA: 78525

AA: 41832

AA: 40234

AA: 17829

AA: 37424

AA: 13955

AA: 5150

.9 0 .9 2 .9 4 .9 6 .9 8 .01 0Time (min)

Time (min)

Time (min)

Time (min)

8000

4000

14000

7000

Inte

nsity

Inte

nsity

254 108

254 156

734 158

734 756

916 174

916 772

716 158

716 558

.9 6 .9 8 .01 0 .01 2 .01 4 .01 6

.9 8 .01 0 .01 2 .01 4 .01 6 .01 8

.01 3 .01 5 .01 7 .01 9 .11 1

SulfamethoxazoleCalc Conc = 2.3 pg/mL

ErythromycinCalc Conc = 1.9 pg/mL

Dehydroerythromycin

TylosinCalc Conc = 2.1 pg/mL

4500

2000

7000

3500

Inte

nsity

Inte

nsity

7000

3000

3500

1500

Inte

nsity

Inte

nsity

3500

1500

8000

4000

Inte

nsity

Inte

nsity

8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0

References

1 Lindsey, M.E.; Meyer, M.; Thurman, E.M. Anal Chem., 2001, 73,4640-4646.

2 Yang, S; Cha, J.; Carlson, K. Rapid Commun. Mass Spectrom., 2004, 18,2131-2145.

AN62485_E 11/07S




3 Gobel, A.; McArdell, C.S.; Suter, M.J.-F.; Giger, W. Anal. Chem., 2004, 76,4756-4764.

4 Shang, D.; Dyck, M.; Jia, X.; DiCicco, A.; Alleyne, C.; Nicolidakis, H.;Mori, B. Proc. 48th ASMS Conf. on Mass Spectrom. and Allied Topics,TPH #264.








www.thermo.com


Administrator

Text Box


Simple and Rapid Analysis of Chloramphenicolin Milk by LC-MS/MSTing Liu1, Peter Wang1, and Kefei Wang2

1Thermo Fisher Scientific, Shanghai, China; 2Thermo Fisher Scientific, San Jose, CA, USA

Key Words


• Accela HighSpeed LC System

• Antibiotic


• SRM


IntroductionChloramphenicol (CAP) is a broad-spectrum antibioticwith historical veterinary uses in all major food-producinganimals (see Figure 1 for structure). It has serious sideeffects on humans that may cause aplastic anemia, and thesuspected carcinogen effect is also thought to be doseindependent. Consequently, chloramphenicol has beenbanned for use in all food-producing animals by theEuropean Union (EU), USA and Canada. A minimumrequired performance limit (MRPL) for chloramphenicoldetermi nation was recently set by the EU at 0.3 µg/kg (ppb)in all foods of animal origin, such as meat, seafood, egg,milk, honey, etc. However, residues of CAP at unacceptablelevels continue to be found in food imports, as a result ofillegal use in some countries to mask the poor hygieneconditions of animal-raising farm and to augment animalgrowth. The growing food safety concerns call for inten -sive surveillance of chloram phenicol in food products.

Analysis of residual of chloramphenicol in foodstuffis challenging because of the complicated sample matricesand stringent requirements of both low quantitation limit(<0.3 ppb) and method validation. The technique of liquidchromatography separation followed by tandem massspectrometry detection, LC-MS/MS, is the technology ofchoice because of its sensitivity and specificity. A samplecleanup process is generally required to remove the samplematrix prior to the LC-MS/MS run. Typically, this involvesthe costly and labor-intensive solid phase extraction (SPE)and/or liquid-liquid extraction (LLE) procedures.

In this work, we report a simple sample preparationprocedure involving only the acetonitrile protein precipita -tion and dilution to extract the CAP from milk, followedby a high-speed LC separation and detection by a triplequadrupole mass spectrometer operated in selected reactionmonitoring (SRM) mode. The sample preparation is simple,fast, and inexpensive, and the method exceeds the sensitivityand specificity requirements for both screening and confir -matory assays. Validation according to the European Com -mission Decision 2002/657/EC has also been performed.

Goal

To develop a simple, rapid, and sensitive LC-MS/MSmethod for analyzing chloramphenicol in milk. Themethod should be suitable for both screening andconfirmatory purposes.


Sample Preparation Standards and Regents: Chloramphenicol (98%) waspurchased from Sigma-Aldrich (St. Louis, MO) and d5-chloramphenicol (100 µg/mL in acetonitrile) as internalstandard from Cambridge Iosotope Lab (Andover, MA).Regent grade water, acetonitrile and methanol were fromThermo Fisher Scientific (Pittsburgh, PA).

Procedures:

Chromatography ConditionsHPLC Module: Thermo Scientific Accela High Speed LC

SystemColumn: Thermo Scientific Hypersil GOLD 50 mm× 2.1

mm and 1.9 µm particle size Column Temperature: AmbientMobile Phase: A: Methanol B: WaterGradient: Time (min) A%

0.0-0.6 5%2.3 100%2.35-3.0 5%

Flow Rate: 500 µL/minInjection Volume: 20 µL (with loop)

Figure 1: Structure of chloramphenicol

0.5 g Milk + d5–CAP (0.3 ppb) as IS

+ 0.75 mL CH3CN, vortex 1 min,Centrifuge @ 14000 rpm for 10 min

Take 0.7 mL Supernatant + 0.3 mL Water, store at 4°C for ≥ 1 hr

Pipette 0.8 mL upper solution forLC-MS/MS Analysis

Administrator

Text Box


Mass Spectrometer ConditionsMass Spectrometer: Thermo Scientific TSQ Quantum

Access triple stage quadrupole massspectrometer

Source: ESI-, 3000 VSheath Gas: 45 unitAuxiliary Gas: 10 unitCapillary Temperature: 300°CSource CID: -7 VQ1 and Q3 Peak Width (FWHM): 0.7 DaScan Time: 0.1 sCollision Gas: Ar (1.5 mTorr)SRM Transitions: 3 SRMs for CAP, 1 SRM for d5-CAP(see Table 1)


Sample Preparation: A major goal for the methoddevelopment in this study is to avoid using the laborintensive and time-consuming SPE or LLE procedures asin literatures. In curret work, the proteins from milk wereremoved with acetonitrile precipitation at ratio of 1.5:1(v/v Acetonitrile:Milk), followed by dilution with water,which is necessary for gradient chromatographic separa -tion. At such ratio, protein removal was not complete,trace amount of precipitates of proteins appeared afterthe sample was stored at 4°C for some time. Thus, thesupernatant was taken for LC-MS/MS analysis after thesample was stored at 4°C for ≥1 hr.

Choice of Quantitation and Qualification Ions: Threeproduct ions were chosen to give an Identification Points(IPs) of 5.5 to meet the requirement of ≥ 4.0 IPs by theDecision 2002/657/EC for confirmatory assay of the pro -hibited substances such as CAP. The m/z 152 was chosenas quantitation ion, the m/z 257 and 194 as confirmationions, consisting with those reported in literatures.

The results of relative ion abundance measured atvarious concentrations are given Table 2. Both relativeion abundance ratios of 257/152 and 194/152 meet therequirements set by Decision 2002/657/EC.

Note that we found the 321>257 transition is morelikely subjected to matrix interferences in many othercases of different matrices, thus if two SRM transitionsneed to be selected (4.0 IPs) for the method, 321>152and 321>194 are preferred.

Method Performance: Figure 2 shows representative SRMchromatograms for a blank and 0.05 µg/kg spiked milksamples. As shown, with high-speed LC, each chromato -graphic run is only 3 min, allowing high throughput forscreening assay. All three SRM traces for CAP at 0.05µg/kg spiked samples can be well quantified. Note thatthe 0.05 µg/kg spiked in milk is equivalent to 0.46 pginjected on column by assuming a full recovery.

It should also be noted that with the high-speed LCseparation of only 3 min for each chromatographic run,the CAP peak width (at 10% above baseline) is as narrowas 6 s. Under current MS acquisition conditions, there are13-14 points across each peak, enough for maintaininga well-defined peak shape for accurate integration.

A representative calibration curve from standardsprepared in milk is shown in Figure 3. Good linearityfrom 0.05 to 1.0 µg/kg with correlation coefficient ofR2= 0.9954 (Weighting factor W = 1/X) was obtained.

Table 3 shows excellent recovery and within-labora -tory reproducibility of the method (at four different days).

Decision Limit (CCα) and Detection Capability (CCβ):According to Decision 2002/657/EC, the Decision LimitCCα is the minimum CAP concentration at which asample is really non-compliant with an error probabilityof 1% (α =0.01), and the Detection Capability (CCβ) isthe minimum amount of CAP that can be quantified andconfirmed with an error probability of 5% (β =0.05).

Two methods can be used for calculating the CCαaccording to the Decision. One is to use the S/N ratio of 3:1of blank samples, similar to those for estimation of limit ofdetection. The other is to use the intercept of calibrationcurve at low levels and the within-laboratory reproducibility.The former method does not work well for LC-MS/MSbecause the very low background (noise count ~0) ofSRM chromatogram often yields unrealisti cally low valuesfor CCα. Thus we use the latter approach by using cali -

157 (17)*326.93d5-CAP (M -H -)

152 (17)*257 (15)194 (16)

320.93CAP (M -H -)

Product Ion(Collision Energy)

Precursor Ion

157 (17)*326.93d5-CAP (M -H -)

152 (17)*257 (15)194 (16)

320.93CAP (M -H -)

Product Ion(Collision Energy)

Precursor Ion

* Product ion used for quantitation

Table 1: SRM transitions for CAP and d5-CAP (IS)

20%

Tolerance by Decision

2002/657/EC

3.4%

15%

7.6%

16%

%RSD

90%

93%

92%

96%

Meann=6 n=6 n=6 n=6

Relative Ion Abundance of 257/152

0.50

0.30

0.15

0.05

CAP Spiked Level

(µg/kg)

17%31%

15%31%

25%28% 25%21%26%

Tolerance by Decision

2002/657/EC

%RSDMean

Relative Ion Abundance of 194/152

Note: Relative ion abundance values were calculated by relative peak area ratios

Table 2: Relative ion abundances at various CAP concentrations in milk and the associated tolerances required by Decision 2002/657/EC

Administrator

Text Box


bra tion data of (0.05-0.15-0.30 µg/kg) to obtain theY-intercept and its standard deviation, SDY-intercept,

CCα=Y-intercept + 2.33*SDY-intercept

Similarly, the CCβ can be calculated from CCα andthe standard deviation of 20 measurement of samplesspiked at CCα level. Here the latter term is approximated

with the within-laboratory reproducibility data of 0.15µg/kg spiking level, thus,

CCβ =CCα + 1.64*SD0.15 µg/kg

Where SD0.15 µg/kg is the within-laboratory repro -ducibility (in standard deviation) of the 0.15 µg/kg inTable 3. The calculated values of CCα and CCβ are0.087 µg/kg and 0.12 µg/kg, respectively.

Figure 2: SRM chromatograms for milk blank and 0.050 µg/kg spiked milk samples

CAPY = 0.148836 + 2.55752*X R 2̂ = 0.9954 W: 1/X

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1

µg/kg

Are

a Ra

tio

Figure 3: Calibration of CAP in milk

8.0%0.04294%0.50

11%0.037104%0.30

13%0.020101%0.15

14%0.006597%0.05

%RSDSD (µg/kg)Mean (%)

CAP Spiking Level Within-laboratory Reproducibility

(n= 20)(µg/kg)

Table 3: Recovery and Reproducibility Data

Administrator

Text Box


AN62530_E 11/07S




Conclusions

A simple, rapid and sensitive method for analysis of CAPin milk by LC-MS-MS has been developed and validated.The sample preparation by protein precipitation anddilution is very simple to perform and avoids the use ofSPE or LLE. With the high-speed Accela LC coupled toa triple quadruple TSQ Quantum Access, each analyticalrun is as short as 3 min. The method can be used for thepurposes of both high-throughput screening and rapidconfirmatory assays.

For screening assay, the method can detect < 0.050µg/kg CAP in milk. For confirmatory assay, the methodvalidated according to Decision 2002/657/EC gives aCCα =0.087 µg/kg and CCβ = 0.12 µg/kg, both belowthe MRPL of 0.3 µg/kg.

References

1. Commission Decision 2002/657/EC of 12 August 2002 implementingCouncil Directive 96/23/ECD concerning the performance of analyticalmethods and the interpretation of results, Official Journal of the EuropeanCommunities, L 221, 2002, 8-36.

2. Bogusz, M.J. et al. “Rapid determination of chloramphenicol and itsglucuronide in food products by liquid chromatography–electrospraynegative ionization tandem mass spectrometry”; J, Chrom. B 2004,807(2), 343-356.

3. Tao, D. et al. “Effects of sample preparation and high resolution SRM onLC-MS-MS determination of chloramphenicol in various food products”;Poster Presentation, 53rd ASMS Conference, San Antonio, TX, USA, June5-9, 2005.

4. Gallo, P. et al. “Development of a liquid chromatography/electrospraytandem mass spectrometry method for confirmation of chloramphenicolresidues in milk after alfa-1-acid glycoprotein affinity chromatography”;Rapid Commun. Mass Spectrom. 2005, 19(4), 574-579.

5. Vinci, F. et al. “In-house validation of liquid chromatography electrospraytandem mass spectrometry method for confirmation of chloramphenicolresidues in muscles according to Decision 2002/567/EC”; Rapid Commun.Mass Spectrom. 2005, 19(22), 3349-3355.








www.thermo.com


Administrator

Text Box


t

Drug ResiduesAntibiotics & Antimicrobials

372: LC/MS/MS Analysis of Anti-InfectivesIn Raw and Treated Sewage

LC-MS/MS Analysis of Anti-InfectivesIn Raw and Treated SewageP.A. Segura1, A.Garcia Ac1, A.Lajeunesse2, D.Ghosh3, C. Gagnon2 and S. Sauvé1

1 Département de Chimie, Université de Montréal, C.P. 6128, succursale Centre-ville, Montréal, QC, Canada H3C 3J72 Centre Saint-Laurent, Environnement Canada, 105, rue McGill, Montréal, QC, Canada H2Y 2E73 Thermo Fisher Scientific, San Jose, CA, USA


Key Words


• Surveyor HPLC

• Antibiotics

• SPE

Introduction

“Anti-infectives” is a general term that refers to severalclasses of biologically active compounds used to treator prevent infections. Therapeutic agents such as anti -microbials (synthetic) and antibiotics (natural orsemi-natural) are examples of anti-infectives.

The widespread utilization of anti-infectives in urbancenters as well as their resistance to biodegradation orelimination in wastewater treatment plants (WWTPs) hasled to their appearance in effluents and surface waters[1-3].In the last few years there has been a growing concernabout the environmental fate and the possible effects ofthese agents on the aquatic environment[4,5].

The first report on the occurrence of anti-infectivetraces in the aquatic environment was published as earlyas 1983[6]. A later study[7] acknowledged that pharmaceu-ticals would enter the water cycle mainly via a “domesticroute” (i.e. by the excreta of individuals taking medicationat homes, hospitals or clinics). It is therefore important toknow the amounts of these substances released in theaquatic environment to be able to evaluate potential effects.

A sensitive and robust method was developed for thedetermination of some of the most prescribed anti-infec-tives in trace amounts (lower nanogram-per-liter range) inraw and treated wastewaters.

Goals• Quantify several anti-infectives at the lower nanogram-

per-liter level in raw and treated wastewaters.

• Apply two specific single reaction monitoring mode(SRM) transitions and their peak ratio to avoid thepresence of false positives.

Method

Raw sewage (north and south influent) was collected andtreated (effluent) 24-h composite samples at the municipalwastewater treatment plant of the City of Montréal(Québec, Canada). This plant has physico-chemical treat-ments only and its effluent is one of the largest in NorthAmerica. We analyzed six of the most prescribed com-pounds (sulfamethoxazole, trimethoprim, ciprofloxacin,levofloxacin, clarithromycin and azithromycin) (Figure 1),by using solid phase extraction (SPE) and liquid chro-matography-tandem mass spectrometry (LC-MS/MS).The compounds were selected based on drugstore sales.

N H 2

SO

OH N

NO

O

O

ON

N N H 2

N H 2

N

O

O HF

N

H N

O

N

O

O HF

N

N

O

O

N

NN H 2

N H 2

C l

O

O

O

OH OO H

OO

O HN

O O H

OO

O

N

O

O HH OO H

OO

O HN

O O H

OO

Sulfamethoxazole Trimethoprim

Ciprofloxacin Levofloxacin Pyrimethamine (IS)

Clarithromycin Azithromycin

O

ON

N NH2

NH2

N

O

O HF

N

HN

O

F

Diaveridine (IS)

Lomefloxacin (IS)

Josamycin (IS)

O

O

HO O

O

O

OO O O

OHN

OO

OH

O

a. b. c.

Figure 1: Molecular structures of the anti-infectives studied (a), the surrogate standard (b), and the internal standards (c).

Administrator

Text Box


Sample Preparation

Wastewater samples were filtered using 1.2 µm pore-sizefiber glass filters and then 0.45 µm pore-size mixed cellu-lose membranes. 50 mM of formic acid and 1 mL of a5% Na2EDTA (w/v) solution were added to 250 mL ofwastewater and the pH adjusted to 3 with NaOH 1.0 M.Pyrimethamine was used as a surrogate standard andspiked at a concentration of 500 ng L-1.

Analytes were pre-concentrated and extracted using a200 mg reversed phase polymeric SPE cartridge on top ofa 200 mg mixed mode polymeric SPE cartridge. Retainedanalytes were eluted from the cartridges using 2 × 2.5 mLACN: MeOH 1:1 (reversed phase) and 2 × 2.5 mL 5%NH3 in ACN: MeOH 1:1 (mixed mode). The eluates wererecovered from both cartridges and were collected on thesame conical-bottom centrifuge tube and then evaporatedto dryness with N2(g). Extracted analytes were reconsti-tuted to 250 µL with 0.1% formic acid in 90% H2O/5%MeOH/5% ACN solution containing the internal stan-dards (diaveridine, lomefloxacin and josamycin).

LC-MS/MS Conditions

HPLC separation was done with a Thermo ScientificSurveyor HPLC system. Detection and quantification ofthe analytes was performed with a Thermo Scientific TSQQuantum Ultra using the single reaction monitoring mode(SRM) (Table 1). Two specific single reaction monitoring(SRM) transitions were used for each compound as wellas their peak area ratios to reliably confirm the presenceof the targeted anti-infectives. This reduced the possibilityof false positives given that some interfering matrix com-ponents areco-extracted with the analytes and could havethe same SRM transition.[8]


MS/MS in the SRM mode proved to be highly selective.Instrument response was linear (r2 ≥ 0.99) in the dynamicrange (25–1000 ng L-1) in spite of the presence of highconcentrations of organic as well as inorganic interfer-ences in the matrix. Limits of detection ranged from

Table 1: Instrument ParametersHPLC MSColumn Thermo Scientific BetaBasic C18

(50 × 2.1 mm, 3 µm) Ionization mode ESI+Column temperature 30°C Spray voltage 3500 VMobile phase A 0.1 % formic acid/H2O Ion transfer capillary temperature 350 ºCMobile phase B 0.1% formic acid/MeOH:ACN 1:1 Sheath gas pressure 21 mTorrInjection volume 20 µL Auxiliary gas pressure 4 mTorrFlow rate 200 µLmin-1 Collision gas pressure 1.5 mTorrGradient t=0 min, A=90%, B=10% Source CID –12 V

t=2 min, A=80%, B=20%t=15 min, A=75%, B=25%t=17 min, A=50%, B=50%t=20 min, A=5%, B=95%t=25 min, A=5%, B=95%t=30 min, A=90%, B=10%

Table 2: SRM transitions used for detection and quantification (SRM #1) and confirmation (SRM #2)Compound SRM #1 CE (V) SRM #2 CE (V) Tube Lens

Pyrimethamine 249.10 177.07 40Sulfamethoxazole† 254.08 92.11 36 254.08 108.10 37 70

Diaveridine 261.15 123.11 34Trimethoprim† 291.16 123.10 33 291.16 230.17 34 91Ciprofloxacin‡ 332.16 231.07 49 332.16 288.15 27 82Lomefloxacin 352.17 265.13 34Levofloxacin‡ 362.17 261.12 35 362.17 221.05 43 92

Clarithromycin* 748.55 590.36 19 748.55 115.99 35 96Azithromycin* 375.33 82.96 25 749.54 158.04 38 74/112

Josamycin 828.53 108.87 46 828.53 173.96 47 126†Quantified using diaveridine as the internal standard, ‡Quantified using lomefloxacin as the internal standard, *Quantified using josamycin as the internal standard

Table 3: Analytical method parametersLimit of Detection Standard SRM Sample SRM SRM ratio

Compound r 2 matrix* (ngL-1) ratio±SD† ratio±SD‡ difference^Sulfamethoxazole 0.9995 22 1.53 ± 0.03 1.6 ± 0.2 -2.6

Trimethroprim 0.9998 7 4.2 ± 0.1 4.39 ± 0.07 -3.3Ciprofloxacin 0.9996 21 5.5 ± 0.8 6.59 ± 0.05 -18.9Levofloxacin 0.9996 4 3.65 ± 0.07 3.83 ± 0.06 -5.0

Clarithromycin 0.9997 0.3 1.67 ± 0.04 1.59 ± 0.09 4.3Azithromycin 0.9900 12 1.2 ± 0.1 0.44 ± 0.1 6.4

*Determination coefficient of the calibration curve made using the WWTP effluent diluted by a factor of 10; **Calculated from the effluent data based on a S/N=3;†Standards spiked WWTP effluent diluted by a factor of 10, n=4; ‡WWTP effluent, n=3; ^Percentage difference between the standard and sample SRM ratio.

Administrator

Text Box


0.3 to 22 ng L-1 (Table 3). As suggested byHernandez[8], the use of two SRM transi-tions in the analytical method (Figure 3)as well as their peak ratios effectively andunambiguously confirmed the presence ofthe studied anti-infectives in all the samples.SRM peak ratios were reproducible (RSD<10%) and differences with SRM peakratios of spiked standards were not higherthan 20% except for AZI (64%).

The tandem-SPE approach utilized topre-concentrate and extract the analytesfrom untreated and treated sewage improvedthe recovery on all six analytes (Figure 2).

The combination of reversed-phase andion-exchange surface chemistry proved tobe a suitable way to extract compoundshaving different chemical properties suchas pKa and pKow.

All targeted anti-infectives were foundin the wastewater samples in concentrationsranging from 39±1 to 276±7 ng L-1 (Figure 4).

Anti-infective daily mass flows in theSt. Lawrence River were estimated usingthe flow of the sampling day (35 m3 s-1)(Table 4). These results show that whileanti-infective concentration in urban waste-waters are typically in the low nanogram-per-liter range, their daily discharged inputsin surface waters can be substantial.

0

20

40

60

80

100

Sulfam

ethoxa

zole

Trimeth

oprim

Ciprofl

oxacin

Levofl

oxacin

Clarith

romyci

n

Azithrom

ycin

Reco

very

/%

Reversed phase

Reversed phase+ mixed mode

Figure 2: Analytes mean percentage recovery (spiked in the effluent at 500 ng L-1, n=2)

100

50

100

50

100

50

100

50

100

50

100

50

0 2 4 6 8 10 12 14 16 18 20 22 24 26Time (min)

Figures 3a-b: Chromatograms showingtwo SRM transitions of the studiedcompounds in treated wastewater.Peaks due to interferences are markedby asterisks(*).

100

50

100

50

100

50

100

50

100

50

100

50

0 2 4 6 8 10 12 14 16 18 20 22 24 26Time (min)

Sulfamethoxazole254.08 → 92.11

Sulfamethoxazole254.08 → 108.10

Trimethoprim291.16 → 123.70

Trimethoprim291.16 → 230.17

Ciprofloxacin332.76 → 288.15

Ciprofloxacin332.16 → 231.07

Levofloxacin362.17 → 261.12

Levofloxacin362.17 → 221.05

Clarithromycin748.55 → 590.36

Clarithromycin748.55 → 115.99

Azithromycin375.33 → 82.96

Azithromycin749.54 → 158.04

a

b

9.21

6.96

9.16

6.99

9.61

9.67

8.77

8.75

20.23

20.23

20.31

20.30

*

*

*

Administrator

Text Box


Conclusions

The developed analytical method allowed the extraction,detection and quantification of six of the most usedanti-infectives in untreated and treated sewage. Detectionlimits ranged from 0.3 to 22 ng L-1 and instrumentresponse was linear (r2 ≥ 0.99) in the dynamic range(25–1000 ng L-1). The use of two specific SRM transitionsand their peak area ratios proved to be a reliable andeffective way to reduce false positives and confirm thepresence of targeted substances. All the studied anti-infectives were found in the wastewater samples in con-centrations ranging from 39 to 276 ng L-1. More studiesare necessary to elucidate the fate of these anti-infectivesafter they are discharged into the St. Lawrence River aswell as their effects on aquatic biota and the environment.

0

100

200

300

Sulfam

ethoxa

zole

Trimeth

oprim

Ciprofl

oxacin

Levofl

oxacin

Clarith

romyci

n

Azithrom

ycin

Conc

entr

atio

n (n

g/L)

North Influent

South Influent

Effluent

Figure 4: Occurrence of the studied anti-infectives in the dissolved phase of raw and treatedsewage of the City of Montréal (n=3)

Table 4: Removal efficiency of the Montréal wastewatertreatement plant and average mass flow of the studiedanti-infectives.

Mean mass flow inthe St. Lawrence

Compound River (g day-1)

Sulfamethoxazole 340 ± 30Trimethroprim 310 ± 20Ciprofloxacin 320 ± 10Levofloxacin 118 ± 2

Clarithromycin 830 ± 60Azithromycin 310 ± 20

References1 Kolpin D.W., Furlong E.T., Meyer M.T., Thurman E.M., Zaugg S.D.,

Barber L.B., and Buxton H.T. (2002) Environmental Science & Technology36:1202-1211.

2 Hirsch R., Ternes T., Haberer K., and Kratz K.L. (1999) Science of theTotal Environment 225:109-118.

3 Metcalfe C.D., Koenig B.G., Bennie D.T., Servos M., Ternes T.A., andHirsch R. (2003) Environmental Toxicology & Chemistry 22:2872-2880.

4 Wilson B.A., Smith V.H., Denoyelles F., and Larive C.K. (2003)Environmental Science & Technology 37:1713-1719.

5 Richards S.M., Wilson C.J., Johnson D.J., Castle D.M., Lam M., MaburyS.A., Sibley P.K., and Solomon K.R. (2004) Environmental Toxicology andChemistry 23:1035-1042.

6 Watts CD, Crathorne B, Fielding M, and Steel CP (1983) Analysis ofOrganic Micropollutants in Water. D. Reidel Publishing Company,Dordrecht.

7 Richardson M.L. and Bowron J.M. (1985) Journal of Pharmacy &Pharmacology 37:1-12.

8 Hernandez F., Ibanez M., Sancho J.V., and Pozo S.J. (2004) AnalyticalChemistry 76:4349-4357.

AN62489_E 11/07S











www.thermo.com


Administrator

Text Box


t

Drug ResiduesPharmaceuticals

30162: The Thermo Scientific Exactive Benchtop LC/MS Orbitrap Mass Spectrometer

The Thermo Scientific Exactive BenchtopLC/MS Orbitrap Mass SpectrometerAndreas Wieghaus, Alexander Makarov, Ulf Froehlich, Markus Kellmann, Eduard Denisov, Oliver Lange, Thermo Fisher Scientific, Bremen, Germany

OverviewReview of a new benchtop mass spectrometer based on a stand-alone Thermo Scientific Orbitrap mass analyzer.Key features of the instrument layout, analyticalparameters and typical applications are described.

IntroductionOver the past three years the combination of Orbitraptechnology with a linear ion trap has become anestablished platform for high resolution, accurate massLC/MSn analysis. The high resolving power, mass accuracyand dynamic range of the Orbitrap analyzer allowrigorous characterization of complex mixtures even in theabsence of precursor ion mass selection. We now describethe development of a non-hybrid mass spectrometercomprising of an atmospheric-pressure ion source (API)and a standalone Orbitrap™ mass analyzer.

MethodsAll experiments were performed on a prototype of thenew Thermo Scientific Exactive mass spectrometer usingan electrospray ionization (ESI) source.

Instrument Layout OverviewFigure 1 shows the schematic layout of the instrument.Samples can be introduced introduced into the API sourceby a variety of methods includingdirect infusion or an U-HPLC system (Thermo Scientific Accela). The source is similar to the commercial source of theThermo Scientific TSQ Quantum Ultra.

Ions are transferred from the source through fourstages of differential pumping using RF-only multipolesinto a curved RF-only trapping quadrupole (the C-trap).In the C-trap ions are accumulated and their energydampened using a bath gas (nitrogen). Ions are theninjected through three further stages of differentialpumping using a curved lens system into the Orbitrapanalyzer where mass spectra are acquired via imagecurrent detection. The vacuum inside the Orbitrap massanalyzer is maintained below 1E-09 mBar.

Automatic Gain Control (AGC)Automatic control of the number of ions in the Orbitrapis performed by measuring the total ion charge using a pre-scan and by calculating the ion injection time for theanalytical scan from this. For very high scan rates, theprevious analytical scan is used as a prescan to optimizethe scan cycle time without compromising automatic gaincontrol. Ion gating is performed using a fast split lenssetup that ensures the precise determination of the ioninjection time.

Higher Energy Collision Induced Dissociation (HCD)In a HCD experiment ions are passed through the C-trapinto a multipole collision cell where they are fragmented.After that, the HCD cell voltages are ramped and ions aretransferred back into the C-trap from where they areinjected into the Orbitrap for detection.

Key Words

• Exactive

• Accurate Mass

• High Resolution

• Polarity Switching

• Scan Speed


Figure 1: Schematic layout of the instrument.

Administrator

Text Box


Automatic Gain Control (AGC)A requirement of any ion trap device is the ability tocontrol the ion population within the trap. When the ionpopulation is not accurately maintained it can result inlarge variations in the quality of data. The correct AGCfunctionality of the Exactive instrument is exemplified inFigure 2 by two mass spectra acquired in the middle andat the end of an eluting LC peak of Buspirone.

In both cases the mass resolution, mass accuracy andsignal-to-noise ratio are excellent. The AGC feature incombination with the precise determination of the ioninjection time allows the instrument to be used foraccurate quantitative analyses.

Scan SpeedThe use of a single mass analyzer with very high trans-mission characteristics in combination with the use of fastdigital and analog electronics allow high resolutionmass spectra to be detected, processed and recorded at

high scan rates of up to 10 Hz. This is compatible withthe narrow peak widths observed in fast chromatographyanalyses (Figure 2).

Mass ResolutionAt a scan rate of 10 Hz the resolving power of theinstrument is > 10,000 at m/z 200. Increasing the transientdetection time by a factor of 10 (corresponding to a scanrate of 1 Hz) the mass resolution can be increased beyond 100,000.

To demonstrate the resolving power of the instrumenta pesticide mixture was measured showing well resolvedisobaric peaks of Dimethon (m/z 231.0273) and Asulam(m/z 231.0434) within a full scan spectrum (Figure 3).

231.02 231.03 231.04 231.05m/z

0102030405060708090

100

231.02757R=104600C6H16O3PS2

C8H11O4N2S1.18536 ppm

231.04359R=106400

0.82379 ppm

231.02 231.03 231.04 231.050

102030405060708090

100

Rela

tive

Abun

danc

e

150 200 250 300 350 400 450 500 550 600

m/z

0

20

40

60

80

100

Rela

tive

Abun

danc

e

291.07571R=94800

313.05768R=89900

276.06339R=97300

199.05530R=114600

245.04321R=101600

169.00880R=124400

445.11548R=75400

417.08408R=77900 551.11932

R=68300

603.12616R=64200

Figure 2: LC peak and mass spectra of Buspirone acquired at a scan rate of 10 scans per second.

Figure 3: Full scan spectrum of a pesticide mixture demonstrating a resolving power of up to 100,000.

Results

Administrator

Text Box


Mass Accuracy and StabilityUsing fully automated AGC and mass calibrationprocedures, mass spectra with high mass accuracy arerecorded. The mass accuracy, precision and stability isequally as good as that obtained in ion trap based hybridinstruments, i.e. Thermo Scientific LTQ Orbitrap or LTQ FT Ultra™.

Figure 4 shows the mass accuracy and its stabilityover time for different molecular ions of an ESIcalibration mixture. The full scan spectra were acquired ata resolution setting of 100,000 in an infusion experimentapplying an external calibration, i.e. no lock masses wereused.

m/z 1722 RMS error: 0.63 ppm

-3-2

-10

12

3

devi

atio

n (p

pm)


-3-2

-1012

3

devi

atio

n (p

pm)


-3-2

-10

12

3

0 30 60 90 120 150time (min)

0 30 60 90 120 150time (min)

0 30 60 90 120 150time (min)

0 30 60 90 120 150time (min)

devi

atio

n (p

pm)


-3-2

-10

12

3

devi

atio

n (p

pm)

Figure 4: Mass accuracy and stability of ions at different m/z values acquired in an infusion experiment without using lock masses.

Figure 5: Mass deviations of m/z 524 (positive ions) and m/z 514 (negative ions) observed in polarity switching experiments.

Fast Polarity SwitchingDue to the use of a novel power supply design it ispossible to perform fast polarity switching withoutsacrificing mass accuracy in any scans. Figure 5demonstrates this feature by means of two experiments. In the first experiment the polarity was changed from scanto scan to check mass accuracy at fast alternating polarity

switching corresponding to a full cycle of 1 positive and 1 negative scan within 1 second. In the second experimentthe polarity was switched every 5 minutes to check forpotential drift effects. In both cases full scan spectra wereacquired at a resolution setting of 30,000 in an infusionexperiment using an ESI calibration solution applying anexternal calibration, i.e. no lock masses were used.

Administrator

Text Box


Dynamic RangeThe dynamic range of the instrument varies by sample andwith the instrument settings but it is typically about 3 to 4 orders of magnitude. Figure 6 shows that it is possibleto acquire full scan spectra with an in-scan dynamic rangeof more than 13,000. The spectrum was acquired in aninfusion experiment using a mix of Buspirone (m/z 386)and Caffeine (m/z 195).

The ratio of the Buspirone signal to the Caffeinesignal is greater than 13,000. Both peaks show massaccuracies of less than 1 ppm. Thus this spectrumdemonstrates not only the high in-scan dynamic range interms of signal but also the high dynamic range in termsof mass accuracy of this instrument – analogous to theperformance of a hybrid LTQ Orbitrap mass spectrometer.

All Ion Fragmentation (HCD)The instrument design allows high efficiency “All IonFragmentation“ experiments by means of Higher EnergyCollision Induced Dissociation (HCD).

As an example, Figure 7 shows full scan spectra ofVerapamil with and without HCD fragmentation anddemonstrates the high fragmentation efficiency and theexcellent mass accuracy of the HCD fragments.

1.3 ppm

1.4 ppm

1.2 ppm

2.1 ppm

150 200 250 300 350 400 450m/z

0

20

40

60

80

100

0

20

40

60

80

100

455.2890

167.0126

239.1614 311.2189

165.0912

303.2071

260.1642 455.2895

MH+

O

O+

+

N

O

O

+N

N

O

O

HCD off

HCD on

Figure 7: Full scan spectra of Verapamil with and without HCD fragmentation.

Figure 6: Spectrum of a mixture of Buspirone (m/z 386) and Caffeine (m/z 195) showing an in-scan dynamic range of > 13,000 and sub-ppm mass accuracies.

Administrator

Text Box


ApplicationsAs a result of the described performance characteristics of this new benchtop Orbitrap mass spectrometer severalkey applications are ideally suited to the use of theExactive mass spectrometer. Some of these are:

1. Exact mass measurements of organic compounds

2. Early drug discovery metabolism and pharmacokinetics (DMPK)

3. General unknown screening

4. Multiple residue analysis (Pesticides, Mycotoxins, veterinary drugs)

5. Metabolomics

For all of these applications high resolution, accuratemass measurements together with high dynamic range isrequired for unequivocal results in full MS mode. Where itis needed, additional information, can be provided by useof by high resolution/high mass accuracy MS/MSexperiments in an “All Ion Fragmentation“ mode. Figure8 shows an extracted ion chromatogram of 116 pesticidesand mycotoxins at a level of 50 ppb in a very complexmatrix of horse feed extract at a mass resolution of50,000. This exemplifies the high selectivity and sensitivityof the instrument working in full scan mode, which is a prerequisite for a successful screening approach, sinceresolving matrix interferences from the target analytes isessential.

ConclusionsA new benchtop mass spectrometer has been developedbased on an API ion source combined with a stand-aloneOrbitrap mass analyzer. The key performance featuresare as follows:

• Mass resolutions of up to 100,000

• Scan speeds of up to 10 Hz

• High in-scan dynamic range (4 orders of magnitude)

• Mass accuracies of better than 2 ppm in full scan and“All Ion Fragmentation“ mode

• Fast polarity switching (full cycle of 1 positive and 1negative scan within 1 second)

• High efficiency “All Ion Fragmentation“ Higher EnergyCollision Induced Dissociation (HCD)

The instrument is very easy to operate and with itsperformance characteristics are ideally suited for discoverywork, screening applications, quantitative analyses andelemental composition determinations.

AcknowledgementsWe would like to thank the other Exactive project teammembers Frank Czemper, Florian Grosse-Coosmann,Thomas Heise, Oliver Hengelbrock, SebastianKanngiesser, Alexander Kholomeev, Sascha Moehring,Uwe Rickens, Ronald Seedorf and Stevan Horning.

Figure 8: Extracted ion chromatogram of 116 pesticides and mycotoxins at a level of 50 ppb in a complexmatrix of horse feed extract.

Administrator

Text Box


AN30162_E 07/08C



View additional Thermo Scientific LC/MS application notes at: www.thermo.com/appnotesThermo Fisher Scientific(Bremen) GmbH is certifiedDIN EN ISO 9001:2000









enhance your system, we also offer certified parts and a range of accessories and

consumables suited to your application.



Laboratory Solutions Backed by Worldwide Service and Support In addition to these







www.thermo.com

Administrator

Text Box


t

Drug ResiduesPharmaceuticals, Personal CareProducts, and Pesticides

466: Detection of Pharmaceuticals, PersonalCare Products, and Pesticides in WaterResources by APCI-LC-MS/MS

Detection of Pharmaceuticals, Personal CareProducts, and Pesticides in Water Resources byAPCI-LC-MS/MSLiza Viglino1, Khadija Aboulfald1, Michèle Prévost2, and Sébastien Sauvé1

1Department of Chemistry, Université de Montréal, Montréal, QC, Canada; 2Département of Civil, Geological, and MiningEngineering, École Polytechnique de Montréal, Montréal, QC, Canada

IntroductionPharmaceuticals (PhACs), personal care productcompounds (PCPs), and endocrine disruptors (EDCs),such as pesticides, detected in surface and drinking watersare an issue of increasing international attention due topotential environmental impacts1,2. These compounds aredistributed widely in surface waters from human andanimal urine, as well as improper disposal, posing apotential health concern to humans via the consumptionof drinking water. This presents a major challenge towater treatment facilities.

Collectively referred to as organic wastewatercontaminants (OWCs), the distribution of these emergingcontaminants near sewage treatment plants (STP) iscurrently an area of investigation in Canada andelsewhere3,4. More specifically, some of these compoundshave been detected in most effluent-receiving rivers ofOntario and Québec5,6. However, it is not clear whethercontamination is localized to areas a few meters from STPdischarges or whether these compounds are distributedwidely in surface waters, potentially contaminatingsources of drinking water.

A research project at the University of Montreal’sChemistry Department and Civil, Geological, and MiningEngineering Department was undertaken to establish theoccurrence and identify the major sources of thesecompounds in drinking water intakes in surface waters inthe Montreal region. The identification and quantificationof PhACs, PCPs, and EDCs is critical to determine theneed for advanced processes such as ozonation andadsorption in treatment upgrades.

The establishment of occurrence data is challengingbecause of: (1) the large number and chemical diversity ofthe compounds of interest; (2) the need to quantify lowlevels in an organic matrix; and (3) the complexity ofsample concentration techniques. To address these issues,scientists traditionally use a solid phase extraction (SPE)method to concentrate the analytes and remove matrixcomponents.

After extraction, several different analytical techniquesmay perform the actual detection such as GC-MS/MS andmore recently, LC-MS/MS7,8. Another analytical challengeresides in the different physicochemical characteristics andwide polarity range of organic compounds – makingsimultaneous preconcentration, chromatographyseparation, and determination difficult. Analytical

methods capable of detecting multiple classes of emergingcontaminants would be very useful to any environmentalmonitoring program. However, up to now, it has oftenbeen a necessity to employ a combination of multipleanalytical techniques in order to cover a wide range oftrace contaminants9. This can add significant costs toanalyses, including equipment, labor, and timeinvestments.

GoalsTo develop a simple method for the simultaneousdetermination of trace levels of compounds from a diversegroup of pharmaceuticals, pesticides, and personal careproducts using SPE and liquid chromatography-tandemmass spectrometry (LC-MS/MS).

Determine which selected substances are present insignificant quantities in the water resources around theMontreal region.

Materials and Method

Analyte selection

Compounds were selected from a list of the most-frequently encountered OWCs in Canada4-6 (Figure 1).

Sample collection

Raw water samples were taken from the Mille Iles, desPrairies, and St-Laurent rivers. Three samples werecollected at the same time from each river in pre-cleaned,four-liter glass bottles and kept on ice while beingtransported to the laboratory. These water sources varywidely due to wastewater contamination and seweroverflow discharges.

All samples were acidified with H2SO4 for samplepreservation and stored in the dark at 4 °C. Immediatelybefore analysis, samples were filtered using 0.7 µm pore-size fiberglass filters followed by 0.45 µm pore size mixed-cellulose membranes (Millipore, MA, USA). Samples wereextracted within 24 hours of collection.

Key Words


• Water Analysis

• Solid PhaseExtraction


Administrator

Text Box


Personal Care Products

Caffeine (CAF)MW: 194.19, pKa =10.4, Stimulant

Triclocarban (TCC)MW: 315.19,Anti-bacterial agent

Pharmaceuticals

Carbamazepine (CBZ)MW: 236.27, pKa=13.9,Anticonvulsant

Clofibric acidMW: 214.65 Metabolite lipid regulator

Naproxen (NAPRO)MW: 230.26, pKa=4.15 Analgesic

Gemfibrozil (GEM)MW: 250.33,Anti-cholesterol

Salicylic acidMW: 138.12Metabolite of acetylsalicylic acid (aspirin)

Trimethoprim (TRI)MW: 290.30, pKa=7.12 Anti-infective

H3C

O OH

CH3O

CH3

CH3 H3C CH3OH

O

O

Hormones

EstroneMW: 270.4 Estrogen

EstriolMW: 288.4 Estrogen

EstradiolMW: 272.4Estrogen

ProgesteroneMW: 314.15Progestogen

17-α-EthinylestradiolMW: 296.4Synthetic estrogen

Pesticide

Atrazine (ATRA)MW: 215.68, pKa =1.7, Herbicide

Figure 1: Molecular structures of selected compounds

O

H

H

HO

O

H

H

HO

Concentration and Extraction ProcedureThe solid phase extraction procedure is illustrated inFigure 2. Briefly, analytes were concentrated and extractedusing a 200 mg C18-like analytical cartridge. Retainedanalytes were eluted from the cartridges using 3 mLMTBE:MeOH 90/10 and 3 mL MeOH. They were thencollected on the conical-bottom centrifuge tube forevaporation to dryness with N2 (g). Extracted analyteswere reconstituted to 200 µL with 90% water/formic acid0.1% and 5% MeOH solution containing the internalstandards.

Administrator

Text Box


LC-MS/MS conditions

HPLC separation was done with a Thermo ScientificSurveyor HPLC system. Separation conditions are given inTable 1. Detection and quantification of the analytes wereperformed with a Thermo Scientific TSQ Quantum Ultratriple stage quadrupole mass spectrometer using selectivereaction monitoring (SRM) (Table 2). Preliminaryexperiments were performed with two atmosphericpressure ionization (API) sources – ESI and APCI – todetect all compounds. Although some compounds showeda slightly higher intensity with the ESI source (i.e.atrazine), APCI was selected because of the highersensitivity provided for steroids. This endocrine disruptionclass is an important analytical challenge due to the lowdetection limits (1 ng/L) required for the determination ofthese compounds. These compounds are known to affectthe living organisms at very low concentrations. Giventhat the aim was to develop a simple analytical method todetect as wide a range of compounds as possible, weselected the APCI source. The small loss in sensitivity forsome easily measured molecules was more thancompensated by the gain in sensitivity for othercompounds that could not have been detected using ESI.Moreover, APCI ionization is known in some cases to beless susceptible to matrix interferences than ESIionization10. Lastly, some authors demonstrated signalsuppression for analysis of various organic wastecompounds in water samples using ESI-LC-MS/MS11.

The identification of analytes was confirmed by theLC retention time12,13. Instrument control and dataacquisition were performed with Thermo ScientificXcalibur software.

➛➛

➛➛

➛➛

➛

Solid Phase Extraction(200 mg, C18-like SPE cartridges)

1 Conditioning:

2 Load:

3 Washing:

4 Drying:

5 Elution:

6 Evaporate:

7 Detection:

MTBE (3 mL)MeOH (3 mL)

Reagent water (3 mL)

1 L filtered sample

Reagent water3 mL

Nitrogen40 min

10/90 MeOH/MTBE (3 mL)MeOH (3 mL)

Nitrogen (Vf = 200 µL)

LC-APCI-MS/MSAnalysis

Surrogate[13C3]-caffeine

Internal standard[13C3]-atrazine and

[13C3]-estradiol

Figure 2: SPE enrichment procedure

Table 1: Instrument Parameters

HP LC MSColumn: Thermo Scientific Hypersil GOLD Ionization mode: APCI+ APCI-

(50 x 2.1 mm, 3 µm)Column temperature: 30 °C Discharge current: 3 µA 4 µAMobile phase A: 0.1% Formic acid/H2O Vaporizer temperature: 500 °C 500 °CMobile phase B: MeOH Capillary temperature: 250 °C 250 °CInjection volume: 20 µL Sheath gas pressure: 40 arb units 30 arb unitsFlow rate: 500 µL/min Aux. gas pressure: 20 arb units 15 arb unitsGradient: T=0, A=90%, B=10% Collision gas pressure: 1.5 mTorr 1.5 mTorr

T=1, A=90%, B=10% Source CID: -10 V 15 VT=15, A=1%, B=99%T=16.5, A=1%, B=99%T=17, A=90%, B=10%T=22, A=90%, B=10%

Administrator

Text Box


Table 2: SRM transitions used for detection and quantification

Compound Precursor ion (m/z) Product ion (m/z) CE (eV) Tube lens (V)Trimethoprim 291.16 230.16 22 90Caffeine 195.10 138.10 18 77Estriol 271.24 157.10 18 80Carbamazepine 237.11 194.10 20 80Atrazine 216.11 174.10 34 97Naproxen 231.11 185.10 13 10117-α-Ethinylestradiol 279.16 133.10 31 86Estradiol 255.16 159.10 17 79Estrone 271.24 157.10 18 80Progesterone 315.26 109.10 38 118TCC 316.99 127.04 32 99Gemfibrozil 251.09 129.10 20 118Salicylic acid* 137.04 93.10 31 72Clofibric acid* 213.17 127.10 32 102

*APCI-

Results and DiscussionReproducibility (%RSD), ranging from 3% to 11% for allanalytes, was very good. Accuracy (recovery percentages),ranging between 72% to 94% for all compounds inspiked matrix, was satisfactory and indicated highperformance of our method. Results are shown in Table 3.

Matrix effects are very important when developing anLC-MS/MS method and can affect reproducibility andaccuracy14. This phenomenon was evaluated by comparingrecovery percentages in Milli-Q® water and surface watersamples (Mille Iles River) spiked at 50 ng/L (n = 6). Wecan consider a very low matrix effect in surface waterssince signal suppression varies from 1% to 13%, exceptfor atrazine and TCC showing an enhancement signal of 6% and 2%, respectively (Figure 3).

Good linearity in surface water samples was observedover a concentration range from <LOD to 100 ng/L withcorrelation coefficients greater than 0.99 for allcompounds. Detection limits in surface water were in therange of 0.03 to 2 ng/L (Table 3).

The compounds of interest were investigated usingsamples from various surface waters. Figure 4 showsrepresentative LC-MS/MS chromatograms of selectedcompounds in surface water. The concentrations areillustrated in Figure 5. The selected compounds weredetected in all river samples at various concentrationsdepending on sampling locations (Figure 5 a and b). The highest concentrations were found for caffeine (16-24 ng/L), atrazine (1.5-39 ng/L), salycilic acid (10-33 ng/L) and gemfibrozil (4-14 ng/L). The lowestconcentrations were found for carbamazepine (3-5 ng/L),clofibric acid, and two hormones (progesterone andestradiol). Trimethoprim, triclocarban and other selectedhormones were detected at trace levels (Trace ≤ limit ofdetection).

Overall, concentrations of most of the compoundsanalysed were similar to those reported from other areasin Canada and Europe3,4.

Figure 3: Mean recoveries for the extraction of selected compounds using C18-like cartridges (spiked inMilli-Q water and Mille Iles River water at 50 ng/L, n=6)

Administrator

Text Box


Table 3: Retention time, limit of detection (LOD), linearity, recoveries and RSD (%) data for each detected compounds in tap water.

Compound Retention time (min) LOD* (ng/L) R2** Recovery***(%) RSD (%)Trimethoprim 5.46 0.50 0.9998 91 7Caffeine 5.79 0.07 0.9995 87 9Estriol 10.14 0.30 0.9981 84 9Carbamazepine 10.76 0.09 0.9999 86 5Atrazine 11.41 0.03 0.9995 86 3Naproxen 12.62 2.00 0.9996 85 917-α-Ethinylestradiol 12.85 0.50 0.9931 73 10Estradiol 12.88 0.10 0.9979 72 6Estrone 12.94 0.60 0.9989 79 9Progesterone 14.44 0.08 0.9994 94 4TCC 15.10 0.20 0.9970 81 10Gemfibrozil 15.17 2.00 0.9991 84 6Salicylic acid 8.82 0.90 0.9993 77 6Clofibric acid 12.00 0.60 0.9989 83 11

*LOD in surface water (Mille Iles River)**Value for calibration line in river water (0-100 ng/L)***Recoveries over the total method (surface samples spiked at 50 ng/L, n = 6).

Figure 4: Representative SRM chromatograms of some selected compounds detected in water matrix (Mille Iles River). Peak due to interferences are markedby asterisks (*)

Caffeine195.10 → 138.10

Salicylic acid137.04 → 91.10

Carbamazepine237.11 → 194.10

Atrazine216.11 → 174.10

Clofibric acid213.17 → 127.10

Naproxen231.11 → 185.10

Estradiol255.16 → 159.10

Gemfibrozil251.09 → 129.10

* *

Administrator

Text Box









www.thermo.com

AN63041_E 06/09S



Legal Notices©2009 Thermo Fisher Scientific Inc. All rights reserved. Milli-Q is a registered trademark of Millipore Corporation. All other trademarks are the property ofThermo Fisher Scientific Inc. and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scientific Inc. products. It isnot intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Specifications, terms and pricingare subject to change. Not all products are available in all countries. Please consult your local sales representative for details.


ConclusionWe developed and successfullyapplied an APCI-LC-MS/MS methodfor quantifying a wide range ofcompounds from a diverse group ofpharmaceuticals, pesticides, andpersonal care products atconcentration in the low ng/L rangein surface waters with good precisionand accuracy. Results confirmed thepresence of pharmaceuticals, personalcare products, and endocrinedisruptors in all water resourcesaround the region of Montreal. Theconcentrations of compoundsfluctuated with sampling locationsdue to the variation of these sources,wastewater contamination andcombined sewer overflow discharges.

References1. T. A. Ternes and A. Joss, Eds., Human

Pharmaceuticals, Hormones and Fragrances:The Challenge of Micropollutants in UrbanWater Management, IWA, 2006, p 443.

2. A. Kortenkamp, Env. Health Perspective,2007, pp 1-42.

3. C. Hao, L. Lissemore, B. Nguyen, S. Kleywegt,P. Yang, K. Solomon, Anal. Bioanal. Chem.,2006, 384, 505-513.

4. C. D. Metcalfe, X-S Mia, W. Hua, R. Letcher,M. Servos, Pharmaceuticals in the CanadianEnvironment. In Pharmaceuticals in theEnvironment: Sources, Fate, Effects and Risks,second edition; K. Kummerer, Ed.; Springer-Verlag, 2004, 67-87.

5. F. Gagnè, C. Blaise, C. Andrè, Ecotoxicologyand Environmental Safety 2006, 64, 329-336.

6. P. A. Segura, A. Garcia-Ac, A. Lajeunesse, D.Ghosh, C. Gagnon and S. Sauvè, J. Environ.Monitor., 2007, 9, 307-313.

7. R. A. Trenholm, B. J. Vanderford, J. C. Holady,D. J. Rexing and S. A. Snyder, Chemosphere,2006, 65(11), 1990-1998.

8. C. Zwiener and F. H. Frimmel, Anal. Bioanal.Chem, 2004, 378, 862-874.

9. T. A. Ternes, Trends Anal. Chem., 2001, 20(8),419-434.

10. S. Souverain, S. Rudaz, J-L. Venthey, J.Chrom. A, 2004, 1058, 61-66.

Figure 5: (a) The highest mean concentrations of selected compounds in water samplescollected from Mille-Iles River, des Prairies River and St-Laurent River (n = 6). (b) Thelowest mean concentrations of selected compounds in water samples collected fromMille Iles River, des Prairies River and St-Laurent River (n = 6).

11. B. J. Vanderford, R. A. Pearson, D. J. Rexingand S. A. Snyder, Anal. Chem., 2003, 6265-6274.

12. J. B. Schilling, S. P. Cepa, S. D. Menacherry,L. T. Barda, B. M. Heard and B. L. Stockwell,Anal. Chem., 1996, 68, 1905.

13. L. Y. T. Li, D. A. Campbell, P. K. Bennett andJ. D. Henion, Anal. Chem., 1996, 68, 3397.

14. T. Reemtsa, Trends Anal. Chem., 2001, 20,533-542.

Administrator

Text Box


t

Natural CompoundsFlavanoids

420: Metabolomic Analysis of Green and Black Tea Extracts Using the Thermo Scientific LTQ Orbitrap XL Hybrid Linear Ion Trap Mass Spectrometer

Metabolomic Analysis of Green and Black TeaExtracts Using an LTQ Orbitrap XL HybridLinear Ion Trap Mass SpectrometerDonna L. Wilson and Charles Yang; Thermo Fisher Scientific, San Jose, CA, USA

Key Words

• Accela HighSpeed UHPLC

• SIEVE Software

• Higher EnergyCollisionalDissociation(HCD)

• Mass FrontierSoftware

• Natural ProductAnalysis


Overview

Purpose: To show a complete analytical metabolomicworkflow including (1) data acquisition using a highresolution accurate mass instrument that is equipped witha Higher Energy Collisional Dissociation (HCD) cell andcoupled to a high pressure LC (Figure 1), (2) metabolitedifferential abundance analysis, and (3) structuralelucidation of relevant metabolites using accurate massand HCD fragmentation information to highlight thecomponent differences between green and black tea.

Methods: Green and black tea extracts were analyzedusing the Thermo Scientific LTQ Orbitrap XL with anHCD cell. Chroma tography was performed using theThermo Scientific Accela High Speed LC equipped with a2.1 mm ID Thermo Scientific Hypersil GOLD columnpacked with 1.9 µm particles. Data Dependent analysiswas performed using an LTQ Orbitrap XL™ with full scandata acquired at a resolving power of 30,000 and MSn

data acquired at a resolving power of 7500 followingHCD fragmentation.

Results: The study included a comparative analysis ofgreen and black tea using differential analysis softwareto identify compositional variations between the two teasamples. Using a UHPLC coupled with a small particlecolumn afforded a fast analysis time while maintainingvery high chromatographic resolution. The high massaccuracy data (better than 3 ppm with external calibra -tion) was used to determine elemental composition andfor tentative identification of compounds via databasesearching. HCD fragmen ta tion facilitated structuralidentification and confir mation. This was demonstratedwith the example of epigallocatechin gallate (EGCG).

Introduction

Metabolomics, the comprehensive and quantitative analysisof wide arrays of metabolites in biological samples, markspromising new research territory. The numerous analytesin these samples have diverse chemistries and polarities.In addition, metabolites occur at a range of concentrationswithin a particular sample. Consequently, comprehensivemetabolomics investigations create many analyticalchallenges that can be addressed using LC-MS/MS.

Tea contains a wide range of components includingvitamins, amino acids, and polyphenols, many of whichare structurally similar and may differ only in the typeand location of a side chain. The use of high resolutionchroma tography is essential for the separation of sucha complex mixture. Furthermore, acquisition of accuratemass data in both full scan and MSn modes enablescomplete structural characterization.

Here, we highlight an untargeted metabolomic work -flow from data acquisition through metabolite ID. Thestudy included differential and structural characterizationof polyphenolic catechin (flavan-3-ol) derivatives andtheaflavin components of green tea and black tea.

Methods

Samples Green tea and black tea aqueous extracts were examinedwithout any pre-treatment. Each sample was analyzed inquadruplicate.

Figure 1: Schematic of the LTQ Orbitrap XL mass spectrometer with an HCD collision cell. The LTQ Orbitrap XL features an HCD collision cell to provide additionalflexibility for low level components in complex mixtures. Ions can be selected in the linear ion trap and fragmented either in the ion trap (CID) or in the HCDcollision cell. For HCD, ions are passed through the C-trap into the gas-filled collision cell, providing high sensitivity and high signal-to-noise fragmentation.

Administrator

Text Box


Chromatography ConditionsChromatographic separation was performed using theThermo Scientific Accela UHPLC system. The LCconditions were as follows:

Column: Thermo Scientific Hypersil GOLD, 100 × 2.1mm, 1.9 µm particle size

Mobile phase: (A) water with 0.1% formic acid;(B) acetonitrile with 0.1% formic acid

Flow rate: 500 µL/min

Injection volume: 10 µL

Gradient: Linear gradient of 100%–1% A over 20 minutes

Mass Spectrometry Conditions MS analysis was carried out using the Thermo ScientificLTQ Orbitrap XL mass spectrometer. The MS conditionswere as follows:

Positive electrospray ion source voltage: 5.0 kV

All methods: Full scan MS in the Orbitrap with a massresolution of 30,000. Data Dependent MS/MS in theOrbitrap on the top three most intense ions from the fullscan at a mass resolution of 7500.

Results

Considerable interest has developed in the potential healthbenefits of teas, particularly in the antioxidant and otherproperties of some of the polyphenolic catechins andtheaflavins (Figure 2). The analysis described here focusedon detection, relative quantitation, and identification ofthese low molecular weight components in green andblack tea samples.

The HPLC separation of tea samples shows excellentpeak separation and low noise, with most componentseluting in less than 10 min. High resolution full scan spectrawere acquired at a mass accuracy of better than 3 ppm.

After data acquisition, SIEVE software was used todetermine statistically significant differences in the metabo -lite profiles of green and black tea samples (Figure 3).By comparing peak intensities between the two chromato -graph ically aligned samples, metabolites present at differentlevels in the two teas were identified.

A

BO

O

OH

OH

R2

OH

OH

R1C

O

O

OH

OH

R2

OH

OH

R1

Epicatechin (EC) Structures

Catechin (C) Structures

O

OH

OH

OH

Galloyl (gal) Unit

458442306290MW (amu)

galgalHHR2

OHHOHHR1

EGCG or GCGECG or CGEGC or GCEC or C

R1 = H or OH

R2 = H or gal

O

OH

OH

O

OH

OH

O

OH

OOH

OHO

R1

R2

Theaflavin Structures

868716716564MW (amu)

galHgalHR2

galgalHHR1

Theaflavin3,3’-

digallate

Theaflavin 3’-monogallate

Theaflavin 3-monogallate

Theaflavin

OH OH

OH

A

OH

OH

R1

O R2

H

H

B

m/z 139, “A-ring” ion “B-ring” ion

C,EC; R1=R2-H; m/z 152

GC,EGC; R1=OH, R2=H; m/z 168

ECG; R1=H, R2=gal; m/z 304

EGCG; R1=OH, R2=gal; m/z 320

A and B-ring ions formed by retro-Diels-Alder (RDA)mechanism in catechin polyphenols.

OOH

OH

H+ H+

+

RDA fragmentation mechanism for theaflavin polyphenols.

O

O

O

OH

OH

OHOH

OHOH

OH

OH

O

R1=

For theaflavin derivatives theproton is retained by thebenzotropolone part (or “B-ring” part) presenting thelarger aromaticity than the“A-ring part”.

R1H

OH R2

OH

OH

O R1

O R2

Figure 2: Three types of tea are produced from Camellia sinensis leaves: green tea (nonfermented), oolong tea (semi-fermented), and black tea (fermented).Catechins are polyphenolic antioxidant plant metabolites of the class flavanoids called flavan-3-ols and are highly present in tea plants. Fermentation inducesenzymatic oxidation of flavan-3-ols and leads to the formation of two major pigments in black tea, theaflavins (TFs) and thearubigns (TRs). Catechins areexpected to be more abundant in green tea and theaflavins more abundant in black tea. The proposed fragmentation pathway for these compounds proceedsvia a Retro-Diels-Alder rearrangement as outlined here.

Administrator

Text Box


After differentially abundant metabolites of interestwere detected, the accurate mass and the MSn data wereused for structural identifica tion. The elemental formula,as determined by the accurate mass data, and the accuratemass value itself were used for metabolite database search -ing. The EGCG metabolite was tentatively assigned usingthis combination of tools.

Further metabolite characterization was accomplishedusing MSn spectra and Mass Frontier software. MassFrontier allowed confident metabolite identification usingits comprehensive spectral library and predictive frag -mentation algorithms to facilitate structural elucidation(Figure 4). The compounds in Figure 3 were identifiedusing this workflow.

CatechinsEluting Theaflavins

Eluting

Green Tea – BlueBlack Tea – Red

Figure 3: Differential metabolite abundance analysis with SIEVE software. A) Chromatographic alignment of the various LC sample traces is the first step inthe SIEVE process. Differences between the green tea samples (Blue) and the black tea samples (Red) can be identified. The Accela UHPLC provided highlyresolved chromatographic peaks and high signal-to-noise ratios. B) After alignment, the corresponding peak intensities are compared for green tea (Blue) andblack tea (Red). The relative abundances of several compounds of interest are shown with their abundance ratios. These metabolites were identified usinga combination of accurate mass database searching and MSn spectra interpretation via Mass Frontier software.

A)

EC or CRatio = 1.97

EGC or GCRatio = 4.53

EGCG or GCGRatio = 2.91

ECG or CGRatio = 0.94

CatechinsTheaflavin

Ratio = 0.07Theaflavin 3-monogallate

Ratio = 0.03Theaflavin 3,3’-digallate

Ratio = 0.01Caffeine

Ratio = 0.56

Theaflavins Caffeine

B)

Administrator

Text Box









www.thermo.com

AN62740_E 04/08S



Legal Notices©2008 Thermo Fisher Scientific Inc. All rights reserved. Mass Frontier is a trademark of HighChem, Ltd. All other trademarks are the property of ThermoFisher Scientific Inc. and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scientific Inc. products. It is notintended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Specifications, terms and pricing aresubject to change. Not all products are available in all countries. Please consult your local sales representative for details.


Conclusions

The analytical metabolomic workflow described hereencompasses data acquisition, discovery of differentiallyabundant metabolites, and metabolite identification.The LTQ Orbitrap XL coupled to an Accela U-HPLCsystem afforded fast analysis times while maintaining highchromatographic resolution. Accurate mass measurementsincreased the confidence in elemental composition assign -ments and ultimately metabolite identification. ThermoScientific SIEVE differential analysis software enabled

large-scale evalua tion of multiple complex LC-MS dataand com parison of metabolite profiles between green andblack tea samples. The spectral database andfragmentation algorithms of Mass Frontier softwarefacilitated structural assignments of metabolites of interestutilizing MSn fragmentation spectra.

References

Menet, MC., Sang, S. Yang, C.S., Ho, CT, and Rosen, R.T., Analysis ofTheaflavins and Thearubigins from Black Tea Extract by MALDI-TOF MassSpectrometry. J. Agric. Food Chem. 2004, 52, 2455-2461.

Figure 4: A metabolite of interest from Figure 3 was chosen for further characterization. This ion was present at levels ~3 times that of black tea and wasidentified as the potent antioxidant, EGCG. Accurate mass was used to identify the metabolite via database searching, and Mass Frontier software was usedto confirm the EGCG identification by using the fragmentation spectra of the parent ion.

Administrator

Text Box


t

Natural CompoundsFlavonoids

30173: Direct Analysis of Red Wine Using Ultra-Fast Chromatography and High Resolution Mass Spectrometry

Direct Analysis of Red Wine Using Ultra-FastChromatography and High Resolution Mass SpectrometryEugen Damoc, Michaela Scigelova, Anastassios E. Giannakopulos, Thomas Moehring, Thermo Fisher Scientific, Hanna-Kunath-Str. 11, 28199 Bremen, GermanyFrantisek Pehal, NNH Hospital Na Homolce, Roentgenova 2, 150 30 Prague, Czech RepublicMartin Hornshaw, Thermo Fisher Scientific, 1 Boundary Park, Hemel Hempstead HP2 7GE, UK

Overview Red wine is a very complex mixture and a rich sourceof beneficial anti-oxidants. Identification and quantitationof these natural products is challenging. Ultra HighPressure Liquid Chromatography (U-HPLC) coupled tothe Thermo Scientific LTQ Orbitrap XL massspectrometer was used for analysis of French red wine,which enabled simultaneous detection and relativequantitation of the wine's anti-oxidant constituents. The phenolic compounds (such as quercetin) responsiblefor most of the health benefits associated with theconsumption of red wine were identified and their variablecontent across two different harvest years was observed.Direct wine analysis approach was then applied tomonitor the progressive changes in red wine after itsexposure to air. This work demonstrated the feasibility of analyzing complex mixtures without any prior samplepreparation by making use of the high resolving power ofboth U-HPLC and the Orbitrap™ mass analyzer detector.

IntroductionFree radicals derived from molecular oxygen areconsidered major causative agents of tissue damage.1,2

Both recent and historical evidence suggests that regulardrinking of wine in moderation has a positive impact on human health thanks to its high content of anti-oxidants.3,4 Red wine in particular contains a complexmixture of phenolic compounds which are importantcontributors to the organoleptic quality of wines as wellas essential components in the evolution of wine.Quercetin is of special interest for its commercial use as ananti-oxidant food supplement with a proven record ofpromoting vascular relaxation, inhibiting human plateletaggregation in vitro, and modulating eicosanoid synthesistowards a pattern likely to be protective against coronaryheart disease.5

Reversed-phase HPLC is well established for theanalysis of flavonoids in red wine, including quantitativeanalysis.6,7,8 Coupling reversed-phase HPLC to a massspectrometer adds considerable benefits such as the ability to:

1) analyze complex mixtures without much sample fractionation

2) monitor hundreds of compounds in a single analysis over a wide dynamic range of concentrations

3) provide an unambiguous identification and structural characterization of the compounds based on accurate mass measurement and informative fragmentation spectra.

Recent advances in both HPLC and mass spectro-metry techniques are having a significant impact on theanalyses of complex mixtures such as those represented by food and agricultural products. First, the use of smallparticles (< 2 µm) in HPLC columns can provideremarkable increase in speed of analysis while maintainingor even improving the separation efficiency. Second, thenew generation of powerful but easy-to-use hybrid massspectrometers, like the LTQ Orbitrap XL, combinesextremely high mass accuracy and resolution with thecapability of multiple levels of fragmentation.9 Thecombination of these powerful techniques provides arobust and confident means of profiling complex mixturesas well as successful identification and advanced structuralcharacterization of detected compounds. As a result, weare seeing rapidly growing interest in the area ofmetabolomic analysis being applied in nutrition andhealth research.10,11

We investigated the potential of a direct analysis of red wine using U-HPLC coupled to a linear ion trap –Orbitrap hybrid mass spectrometer. Of particular interestwas the ability of the designed workflow to pinpointstatistically significant differences between individualharvest years for wines of a specific origin (area, label). In addition to that, we used the developed methodology tomonitor the trend in oxidative changes of red wine afterexposure to air.

Key Words

• LTQ Orbitrap

• U-HPLC

• Flavonoids

• Red Wine


Administrator

Text Box


MethodsTwo bottles of French red wine Les Charmes de Kirwan,Margaux (cuvee, Bordeaux region, France), years 2003and 2005, were obtained from a specialized winemerchant. The wine was stored at room temperature inthe dark until analyzed. Immediately after opening thebottle, a glass vial (20 mL) was filled with the wine to thevery top, quickly closed to ensure minimum oxidation,and stored at 4°C in the dark. This sample was collectedjust in case there was a need for repeated analysis of theprofiling experiments or structural elucidation studies. A second 20 mL aliquot of wine was poured from theoriginal bottle into a glass beaker. From this beaker asample vial was immediately filled to the rim and placedin the chilled (4° C) Thermo Scientific Accela autosamplertray, awaiting analysis. For a wine oxidation trendanalysis, further samples were taken from this openbeaker 1, 5 and 24 hours after the bottle opening.

Chromatography was performed using an Accela U-HPLC injecting 20 µL sample from a cooled tray (4°C)directly onto a Thermo Scientific Hypersil GOLD column(2.1 mm x 100 mm, 1.9 µm particles, equilibrated in 95% solvent A (0.1% aqueous solution of formic acid), 5% solvent B (acetonitrile containing 0.1% formic acid). The compounds were eluted using flow rate 300 µL/minby linearly increasing solvent B concentration from 5% tofinal 40% over 15 min, and from 40% to 95% over 1 min. The column was then washed with 95% solvent B (2 min) and re-equilibrated in 95% solvent A, 5% solvent B. The total run time, including column wash and equilibration, was 20 min.

A Thermo Scientific LTQ Orbitrap XL massspectrometer was operated in positive ion mode at 30,000resolving power (defined as FWHM @ m/z 400) for fullscan analysis (mass range 150 – 1500 u) followed by datadependent MS/MS on the most intense ion from the fullscan at 7,500 resolving power (~0.3 sec per scan). The measurements were done in triplicate with externalcalibration. The settings for the higher energy collisionaldissociation (HCD) fragmentation mode were 65%normalized collision energy, isolation width 3 u.

Thermo Scientific SIEVE 1.2 software was used forcomparative and trend analyses. The software allows forprocessing a large number of samples, presenting thestatistically significant differences between populationsand various time points. Data were normalized on totalspectral ion current. Results were filtered using pValue < 0.001 and at the same time requiring a minimum 2-foldchange in peak height.

The results from SIEVE™ were further subjected tomultivariate analysis with SIMCA P+™, version 11(Umetrics, Umea, Sweden).

Mass Frontier™ (HighChem, Slovakia) software wasused to confirm a suggested compound identity andstructure based on observed fragmentation patterns.

ResultsDue to the large number and the chemical complexity ofphenolic compounds in wine matrix, analytical methods inthe past involved sometimes difficult and complicatedtraditional chromatographic techniques. One of the major problems underlying separation of the phenoliccompounds is their similarity in chemical characteristics.As many phenolics show similar UV spectra with maximain a narrow range of 280-320 nm, extensive fractionationsteps might be needed prior to HPLC analysis. Ratherlarge initial volumes required and variable losses occuringdue to incomplete extraction or oxidation can be an issue.The use of modern chromatographic techniques coupledto mass spectrometric detection can alleviate theseproblems.

Our approach avoids entirely the sample fractionationstep: red wine is injected directly on the reverse phasecolumn. Moreover, the use of small particles (< 2 µm) andrelatively high flow rates (300 µL/min) enable swiftanalysis with excellent chromatographic resolution. Theobserved peak width for individual compounds was, onaverage, 7 sec, back pressure not exceeding 350 bar. With 20 min total cycle time per injection, this setupallows for high throughput analysis while the total sampleconsumption remains negligible (20 µL per injection). U-HPLC coupled to the LTQ Orbitrap XL proved to be very robust, allowing for an uninterrupted analysis of 24 untreated red wine samples which corresponds to an 8-hour continuous analysis without any requirementfor a system cleanup or column change.

Figure 1: Overview of differences between harvest years 2003 and 2005. The result from differential analysis software (SIEVE 1.2) highlights thecompounds having at least two-fold higher concentration in year 2005compared to year 2003 (■ blue shaded area) and compounds whoseconcentration in year 2005 was less than a half of that in year 2003 (■ red shaded area). The purple horizontal line represents 1:1 ratio betweenconcentrations in the year 2005 and 2003. The grayed area covers thefeatures with less pronounced concentration difference and those with low statistical significance, i.e. pValue > 0.001.

Administrator

Text Box


Variables like wine varietals, soil composition, andharvest year will play an important role by providing thebasic pool of compounds for these biotransformations.With accurate mass acting as a highly selective filter wecould monitor hundreds of compounds across multiplesamples, enabling advanced comparative studies and trendanalyses. Initially, we were interested in comparing thewine of the same origin (area, label) but harvested indifferent years.

Our differential analysis of the Les Charmes deKirwan, Margaux, contrasted wine from production years2003 and 2005 using SIEVE software. The features(peaks) were filtered for their statistical significance(pValue < 0.001) and significant change defined as aminimum 2-fold concentration difference between the twoharvest years (Figure 1). We observed 75 individualcompounds which showed at least 2-fold higher content inyear 2005 compared to year 2003 (blue shaded area inFigure 1). Kaempferol and quercetin concentrationincreased 25- and 8-fold, respectively, in year 2005compared to 2003 (Figure 2). On the other hand, therewere 36 other compounds whose concentration in the2005 sample was significantly less than in the 2003sample (red shaded area in Figure 1). Some flavonoids(myricetin) showed no change in concentration betweenthe two harvest years.

Total anti-oxidant status refers to overall antioxidantproperties of wine, and can be largely ascribed to a groupof compounds comprising vanillic acid, trans-polydatin,catechin, m-coumaric acid, epicatechin, quercetin, cis-polydatin and trans-resveratrol.12 In our analysis wedetected vanillic acid, (epi)catechin, coumaric acid, andquercetin. When compared to wine produced in 2003, thewine produced in 2005 contained 50, 40 and 20% lesscoumaric acid, vanillic acid and (epi)catechin, respectively,while the amount of quercetin increased 8-fold (Table 1).

Calc m/z Formula MW Name Change 24h/0h Change 2003/2005

165.0546 C9H8O3 Coumaric acid 0.32 0.51169.0495 C8H8O4 Vanillic acid 0.40 0.61199.0601 C9H10O5 Syringic acid 0.59 0.92391.1387 C20H22O8 Polydatin Not found Not found229.0859 C14H12O3 Resveratrol Not found Not found291.0863 C15H14O6 (Epi)catechin 0.66 0.82303.0499 C15H10O7 Quercetin 0.78 7.96319.0448 C15H10O8 Myricetin 0.69 1.00287.0550 C15H10O6 Kaempferol 1.00 20.96

Table 1: Overview of some compounds of interest and the changes in theircontent between year 2003 and 2005 (column Change 2003/2005), and after24 hours following exposure to air (column Change 24h/0h). The compounds highlighted are the major contributors to the total anti-oxidant status.12

The remarkable difference in the content of quercetinbetween the two harvest years is interesting. Quercetin isone of the most abundant natural flavonoids found infruits, vegetables and wine.

Figure 2: Extracted ion chromatogram for kaempferol and quercetin (left and right pane, respectively; 3 injections each) shows remarkable difference inconcentration of these compound in wine harvested in year 2005 (blue trace) and 2003 (red trace). The mass deviation did not exceed 0.7 ppm forkaempherol (calculated m/z = 287.0550) and 1.3 ppm for quercetin (calculated m/z = 303.0499). Note the reproducibility of the retention time (RT) values and peak height calculations (AH) for 3 replicate injections.

Administrator

Text Box


At present, labeling requirements for red wine are farfrom comprehensive, basically limited to listing the total alcohol content and the comment that it contains sulfites.Including more specific information about compoundswith strong anti-oxidant properties would improve a general public awareness and be helpful in the currentclimate of debate on healthy balanced diet. A fast buthighly informative analysis of wines as described hereincan thus help maintain consistency and quality, andprovide useful information about product’s nutritionalvalue.

Reliable accurate mass measurements over a broaddynamic range of concentration are helpful forunambiguous identification of compounds of interest. The mass deviation of our measurements did not exceed 2 ppm using external calibration. Such an accuracysupported by reliably measured isotope abundancies in the LTQ Orbitrap XL enabled a confident assignment of elemental composition to individual peaks.

For confident identification of a compound, theelemental composition suggestions based on massaccuracy need to be complemented with the evidence fromthe fragmentation spectra. Our method was set up tocollect higher energy collision dissociation (HCD) spectra.On average 700 such spectra were collected during each20-minute LC-MS run. The MS/MS spectrum acquired inthe multipole collision cell of the LTQ Orbitrap XL servesfor confirming identity of a known compound or evendetermining identity of an unknown. Such an approachwas demonstrated for the analysis of antioxidantcompounds in olive oil.13 Rich fragmentation, accuratemass measurement of both parent and fragment ions, andspectrum interpretation provided by Mass Frontier soft-ware were all crucial for this challenging task (Figure 3).

The anti-oxidant properties of wine are clearlybeneficial to a consumer. On the other hand, wines withhigher polyphenolic concentration are more susceptible to oxidation. We were interested to observe a trend ofchanges in the wine samples over the period of 24 hoursafter opening the bottle. The groups of samples from timepoints 0, 1, 5, and 24 hours (triplicate injections) wereprocessed with SIEVE and further subjected to principalcomponent analysis. The progressive changes caused byexposure to air are well observable and statisticallysignificant (Figure 4).

Figure 4: Wine sampled in triplicate at 0, 1, 5, and 24 hours after exposureto air. The sample groups are easily separated by the first two principalcomponents.

Figure 3: Confirmation and structural characterization of quercetin. Assignment of fragments in HCD spectrum usingthe Mass Frontier software relying on its extensive database of fragmentation mechanisms.

Administrator

Text Box


At this point, a potential effect of evaporation of morevolatile constituents of wine has to be considered. In general, the partial pressure for compounds withmolecular weight 300 and higher is considered negligible –such compounds should not be lost to evaporation atroom temperature. Kaempferol (MW 286), (epi)catechin (MW 290), myricetin (MW 318) and quercetin (MW 302)would fall into such a category. Kaempferol showed nochange over this period. Thus the decrease of 20% forquercetin and 30% for (epi)catechin and myricetinobserved over the period of 24 hours following the bottleopening could be confidently ascribed to oxidation. Forcoumaric (MW 164), vanillic (MW 168) and syringic(MW 198) acids we observed a more pronounced drop inconcentration (60, 70 and 40%, respectively) after 24 hours following the exposure to air (Figure 5). It mightprove difficult, however, to distinguish between the effectof evaporation and oxidation under the employedexperimental conditions.

Figure 5: Changes over a 24-hour period of air exposure. The amount of a given compound at time 0 hdefined as 100%. Panel A shows decrease in content of higher molecular weight compounds such as(epi)catechin (MW 290), myricetin (MW 318) and quercetin (MW 302). Lower molecular weightcomponents including coumaric (MW 164), vanillic (MW 168) and syringic (MW 198) acid display much higher rate of disappearance (panel B). Error bars show the standard deviation for three repetitiveanalyses of samples at each time point.

Administrator

Text Box



ConclusionsAs consumers are becoming increasingly aware of theharmful as well as helpful content of what they eat anddrink, modern powerful analytical tools will undoubtedlyplay a crucial role to supply that information moreaccurately and quickly. Albeit a very complex mixture, red wine is perfectly suitable for mass spectrometricsupported by SIEVE differential expression software. Such ‘fingerprinting’ analysis can be applied in qualitycontrol and process monitoring, and for highlightingrelevant nutritional value to consumers.

• U-HPLC affords fast analysis times while maintainingvery high chromatographic resolution (peak width 7 seconds at half height).

• The mass deviation of the LTQ Orbitrap XLmeasurements was always smaller than 2 ppm usingexternal calibration up to one day old.

• Higher collision energy dissociation MS/MS spectraconfirm the identity and structure of compounds incomplex mixtures.

• Accurate mass measurements also significantly improvethe precision of quantitation by eliminating nearlyisobaric interferences. This is a particularly importantaspect for complex mixture analyses, which red wineundoubtedly is.

• The methodology described here is extremely robust,allowing for an uninterrupted analysis of 24 untreatedred wine samples (continued analysis over an 8-hourperiod).

AcknowledgementsThe authors are grateful to David Kusel for insightfulcomments to the manuscript.

References1 Packer L., Interactions among antioxidants in health and disease:

Vitamin E and its redox cycle. Proc. Soc. Exp. Biol. Med. 1992, 200, 271-276.

2 Mehra M. R., Lavie C. J., Ventura H.O., Milani R. V., Prevention of atherosclerosis: the potential role of antioxidants. Postgrad. Med. 1995, 98, 175- 184.

3 Maxwell S., Cruikshank A., Thorpe G., Red wine and antioxidant activity in serum. Lancet 1994, 344, 193-194.

4 Whitehead T. P., Robinson D., Allaway S., Syms J., Hale A., Effect of red wine ingestion on the antioxidant capacity of serum. Clin. Chem. 1995, 41, 32-35.

5 Tomera J. F., Current knowledge of the health benefits and disadvantages of wine consumption. Trends Food Sci. Technol., 1999, 10, 129-138.

6 Hyoung S. L., HPLC Analysis of Phenolic Compounds. In Food Analysis by HPLC; Nollet, L.M.L. Ed., CRC Press, 2000, 796.

7 Stecher G., Huck C. W., Stoeggl. W. M., Bonn G. K., Phytoanalysis: a challenge in phytomics. TrAC (Trends in Analytical Chemistry). 2003, 22, 1-14.

8 Goldberg D. M., Tsang E., Karumanchiri A., Diamandis E. P., Soleas G. J.,Ng E., A method to assay the concentrations of phenolic constituents of biological interest in wines. Anal. Chem. 1996, 68, 1688-1694.

9 Makarov A., Denisov E., Kholomeev A., Balschun W., Lange O., Strupat K., Horning S., Performance evaluation of a hybrid linear ion trap/orbitrap mass spectrometer. Anal. Chem. 2006, 78, 2113-2120.

10 Breitling R., Pitt A. R., Barrett M.P., Precision mapping of the metabolome. Trends in Biotech. 2006, 24, 543-548.

11 Kussmann M., Rezziand S., Daniel H., Profiling techniques in nutrition and health research. Curr. Opin. Biotech. 2008, 19, 83-99.

12 Soleas G. J., Tomlinson G., Diamandis E. P., Goldberg D. M., The relative contributions of polyphenolic constituents to the antioxidant status of wines: development of a predictive model. J. Agric. Food Chem.1997, 45, 3995-4003.

13 Dourtoglou V., Mamalos A., Makris D. P., Kefalas P. Storage of olives (Olea europaea L.) under CO2 atmosphere: liquid chromatography-mass spectrometry characterization of indices related to changes in polyphenolic metabolism. J. Agric. Food Chem. 2006, 54, 2211−2217.








www.thermo.com

AN30173_E 02/09C


Thermo Fisher Scientific(Bremen) GmbH is certifiedDIN EN ISO 9001:2000

Thermo Fisher Scientific(Bremen) GmbH is certifiedDIN EN ISO 9001:2000

©2009 Thermo Fisher Scientific Inc. All rights reserved. Mass Frontier is a trademark of HighChem; SIMCA is a trademark of Umetrics AB, Umea, Sweden; Charmes des Kirwan is a trade name of Maison Schröder & Schyler, Bordeaux, France; Margaux is a protected designation of origin. All other trademarks are the property of Thermo Fisher Scientific Inc. and its subsidiaries. It is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are available in all countries. Please consult your local sales representative for details.

Administrator

Text Box


t tPesticidesPesticides

454: Analysis of Haloacetic Acids in Drinking Water by IC-MS/MS

391: Analysis of Regulated Pesticides inDrinking Water Using Thermo ScientificAccela and EQuan

434: Analysis of Triazine Pesticides inDrinking Water Using LC-MS/MS (EPA Method 536.0)

425: Determination of Different Classes of Pesticide Residues in Processed Fruits and Vegetables by LC-MS Using the Thermo Scientific TSQ Quantum UltraAccording to EU Directive 91/414 EEC

437: LC-MS/MS Analysis of Herbicides inDrinking Water at Femtogram Levels Using20 mL Thermo Scientific EQuan DirectInjection Techniques

395: LC-MS/MS Determination of MalachiteGreen and Leucomalachite Green in FishProducts

387: Multi-residue Analysis of Pesticides in Food using GC/MS/MS with the Thermo Scientific TSQ Quantum GC

378: Quantitation-Enhanced Data-Dependent(QED) Scanning of Drinking Water SamplesUsing the Thermo Scientific EQuan Systemfor Pesticide Analysis on a Triple StageQuadrupole

323: Simultaneous Detection of 88Pesticides on the Thermo Scientific TSQ Quantum Discovery using a NovelLC/MS/MS Method

373: Testing LC-MS System Robustness withAutomated Sample Cleanup Using Red Wineas a Matrix

453: UHPLC Separation of TriazineHerbicides at Elevated Temperature

355: Utility of H-SRM to Reduce MatrixInterference in Food Residue Analysis ofPesticides by LC/MS/MS using the Thermo Scientific TSQ Quantum Discovery

351: Zero Cross-talk on the ThermoScientific TSQ Quantum

tt

tt

t

tt

tt

t

t

Pixie.Hranilovich

Sticky Note

Marked set by Pixie.Hranilovich

Pixie.Hranilovich

Sticky Note

Marked set by Pixie.Hranilovich

Introduction

Haloacetic acids (HAAs) are formed as disinfection by-products when water is chlorinated to remove microbialcontent. The chlorine reacts with naturally occurringorganic and inorganic matter in the water, such asdecaying vegetation, to produce by-products that includeHAAs. Of the nine species of HAAs, five are currentlyregulated by the EPA (HAA5): monochloroacetic acid(MCAA), dichloroacetic acid (DCAA), trichloroacetic acid(TCAA), monobromoacetic acid (MBAA), anddibromoacetic acid (DBAA). The remaining four HAAsare unregulated: bromochloroacetic acid (BCAA),bromodichloroacetic acid (BDCAA), dibromochloroaceticacid (DBCAA), and tribromoacetic acid (TBAA).

According to the U.S. Environmental ProtectionAgency (EPA), there might be an increased risk of cancerassociated with long-term consumption of watercontaining levels of HAAs that exceed 0.6 mg/L.1 EPAMethods 552.1, 552.2, and 552.3, are used to determinethe level of all nine HAAs in drinking water.2,3,4 Thesemethods require derivatization and multiple extractionsteps followed by gas chromatography (GC) with electroncapture detection (ECD).

In comparison to the conventional EPA methods usingGC with ECD, the combination of ion chromatographyand mass spectrometry (IC-MS and IC-MS/MS) offerssensitive and rapid detection without the need for samplepre-treatment. Ion chromatography is a form of liquidchromatography that uses ion-exchange resins to separateatomic and molecular ions. The retention time in thecolumn is predominantly controlled by the interactions ofthe ions of the solute with the resin. Coupling IC with thehighly selective detection of a triple quadrupole massspectrometer allows unambiguous identification ofsubstance peaks. Matrix interference effects are greatlyreduced, which improves the sensitivity and lowers thedetection limits.

In the method described here, water samples can beinjected directly into an ion chromatography system thatis coupled to a Thermo Scientific TSQ Quantum Accesstriple stage quadrupole mass spectrometer. The separationof all nine HAAs addressed in the EPA methods isachieved with an anion-exchange column using anelectrolytically formed hydroxide gradient.

Goal

To develop a simple, rapid, and sensitive IC-MS/MSmethod for analyzing haloacetic acids in water.


Ion Chromatography

IC analysis was performed on a Dionex ICS 3000 system(Dionex Corporation, Sunnyvale, CA). Samples weredirectly injected and no sample pre-treatment wasrequired. The IC conditions used are shown in Table 1.

Column Set: Dionex IonPac® AG24 (2 × 50 mm), IonPac AS24 (2 × 250 mm)

Suppressor: ASRS® 300, 2 mm Column Temperature: 15 °CInjection Volume: 100 µLFlow Rate: 0.3 mL/min KOH gradient, electrolytically generated

(Table 2)

Table 1. Ion chromatography system conditions

Retention Time (min) [KOH] mM

0.00 7.015.1 7.030.8 18.031.0 60.046.8 60.047.0 7.0

Table 2. Electrolytically formed hydroxide gradient details

The separation performed on the IonPac AS24 columnused a hydroxide gradient. It is known that hydroxide isnot a recommended eluent for mass spectrometers. Theaddition of an ASRS 300 anion self-regeneratingsuppressor is critical. This suppressor is placed in lineafter the column and electrolytically converts thehydroxide into water, making the separation compatiblewith mass spectrometric detection. See Figure 1.

Analysis of Haloacetic Acids in Drinking Waterby IC-MS/MSCharles Yang1 and Stacy Henday2

1Thermo Fisher Scientific, San Jose, CA; 2Dionex Corporation, Sunnyvale, CA

Key Words


• EPA

• Ionchromatography

• Water analysis


Administrator

Text Box


In addition, a matrix diversion valve was placed inline prior to the mass spectrometer. This valve functions todivert the high sample matrix waste from the MS source,prolonging the time in between cleanings. Acetonitrile wasteed into the eluent stream after the matrix diversionvalve. The acetonitrile had two main purposes: to assist inthe desolvation of the mobile phase and to act as a make-up flow when the IC eluent was diverted to waste.

Mass Spectrometry

MS analysis was carried out on a TSQ Quantum Access™

triple stage quadrupole mass spectrometer with a heatedelectrospray ionization (H-ESI) probe. The MS conditionsused are shown in Table 3.

Figure 1: Flow schematic of the IC-MS/MS system

Skimmer Q1 Q3 CE Tube Lens Offset Scan Time

Analyte (m/z) (m/z) (V) (V) (V) (s)MCAA 93.01 35.60 10 26 0 1.25MBAA 136.99 79.09 12 33 0 1.25DCAA 127.02 83.20 11 26 0 1.25DBAA 214.80 79.20 24 33 0 1.25BCAA 171.00 79.20 35 44 0 1.25TCAA 161.06 117.10 10 69 0 1.60

BDCAA 79.00 79.00 15 30 0 1.60DBCAA 206.74 79.13 15 30 0 2.50TBAA 250.70 79.10 25 26 0 2.50

MCAA-ISTD 94.01 35.60 10 26 0 1.25MBAA-ISTD 138.00 79.09 12 33 0 1.25DCAA-ISTD 128.01 84.20 11 26 0 1.25TCAA-ISTD 162.06 118.10 10 69 0 1.60

Table 4. MS conditions for the various HAAs and internal standards

Ion source polarity: Positive ion modeSpray voltage: 4000 VSheath gas pressure: 40 unitsAuxiliary gas pressure: 15 unitsCapillary temperature: 270 °C

Table 3. Mass spectrometer conditions

Individual standards were infused into the massspectrometer to determine optimum tube lens settings andcollision energies for the product ions. Table 4 describesthe MS conditions for specific HAAs and internalstandards.

Administrator

Text Box


Figure 2: These chromatograms show the progress of the pump pressure and front end detector data along with the TSQQuantum Access MS data. The left side of the screen shows the status of the ion chromatography system and the status ofthe TSQ Quantum Access.

The status of the ion chromatography system wasmonitored at the same time as the MS data acquisition, asshown in Figure 2.


The separation of the nine HAAs is shown in Figure 3.The selectivity of the IC-MS/MS system allows separationof the HAAs from common inorganic matrix ions. Thisallows matrix peaks of chloride, sulfate, nitrate, andbicarbonate to be diverted to waste during the analyticalrun and avoids premature fouling of the ESI-MS/MSinstrument source.

An internal standard mixture of 13C labeled MCAA,MBAA, DCAA, and TCAA was spiked into each sampleat 3 ppb. The calibration curves were generated usinginternal standard calibrations for all of the HAAcompounds in water. Excellent linearity results wereobserved for all compounds as shown in Figures 4, 5, and6. Analytes were run at levels of 250 ppt to 20 ppb. Itmust be noted that the TCAA analyte could not be

detected at levels below 2.5 ppb. TCAA sensitivity is verystrongly correlated with the source temperature of themass spectrometer. To improve the TCAA detection, thetemperature was lowered. However, lowering thetemperature impacted the detection of the other eightanalytes. This phenomenon of TCAA temperaturesensitivity has been reported in studies with other MSinstrumentation configurations.5

To test the recoveries of all nine HAAs, spiked matrixsamples were run in a matrix of 250 mg/L of each ofchloride and sulfate, 150 mg/L of bicarbonate, 30 mg/L of

TSQ Quantum Access™

Administrator

Text Box


1. Monochloroacetic acid (MCAA)2. Monobromoacetic acid (MBAA)3. Dichloroacetic acid (DCAA)4. Bromochloroacetic acid (BCAA)5. Dibromoacetic acid (DBAA)

6. Trichloroacetic acid (TCAA)7. Bromodichloroacetic acid (BDCAA)8. Dibromochloroacetic acid (DBCAA)9. Tribromoacetic acid (TBAA)

Figure 3: Separation and detection of the nine haloacetic acids using an ISC-3000 system with an IonPac AS24 column,coupled to a TSQ Quantum Access MS/MS system. Analyte levels were 2.5 ppb each in deionized water with aninjection volume of 100 µL.

Figure 4: Calibration curve overlay of the HAA compounds in water by IC-MS/MS

Administrator

Text Box


Figure 5: Overlay of calibration curves of dibromochloroacetic acid and tribromoacetic acid in water by IC-MS/MS

Figure 6: Overlay of calibration curves of dibromoacetic acid and bromodichloroacetic acid in water by IC-MS/MS

nitrate, and 100 mg/L ammonium chloride preservative,for a total chloride concentration of 316 mg/L. The resultsare shown in Table 5. Excellent recoveries andreproducibility were achieved for most of the samples.However, difficulty was observed when quantitating lowlevels of DBCAA in matrix. DBCAA does not ionize asstrongly as the other analytes in the method and is verysusceptible to temperature changes in the column.

Method detection limits (Table 6) were calculated by

seven replicate injections of 1.0 ppb of each analyte andthe equation MDL=t99%xS(n-7), where: t is Student’s t at99% confidence intervals (t99%, n=7 = 3.143) and S is thestandard deviation. Table 6 compares these results to thecalculated MDL values of EPA Method 552.2, which usesliquid-liquid extraction and methylation of the carboxylicacids before determination by GC-ECD. The resultsobtained by the IC-MS/MS method were comparable tothose achieved in EPA Method 552.2.

Administrator

Text Box









www.thermo.com

AN62963_E 01/09S



Legal Notices©2009 Thermo Fisher Scientific Inc. All rights reserved. IonPac, ASRS, and Dionex are registered trademarks of Dionex Corporation. All other trademarksare the property of Thermo Fisher Scientific Inc. and its subsidiaries. This information is presented as an example of the capabilities of Thermo FisherScientific Inc. products. It is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others.Specifications, terms and pricing are subject to change. Not all products are available in all countries. Please consult your local sales representative fordetails.


Conclusion

IC-MS/MS is a powerful tool used in the quantitation ofhaloacetic acid samples. When compared to theconventional EPA methods using GC with electroncapture, using IC-MS/MS to analyze for haloacetic acidssaves analysts several hours of sample preparationbecause there is no requirement for sample pre-treatment. The resolution between the matrix peaks andhaloacetic acids is excellent, which allows for minimuminterference in detection.

Excellent recoveries and reproducibility wereachieved when samples were spiked into a simulatedmatrix containing 250 mg/L of each of chloride andsulfate, 150 mg/L bicarbonate, 30 mg/L of nitrate and100 mg/L ammonium chloride preservative for a totalchloride concentration of 316 mg/L. Results arecomparable to those achieved in EPA Method 552.2.

Analyte Average RT %RSD RT Average Area %RSD AreaMCAA 12.59 0.00 764439 2.34MBAA 14.06 0.27 1627886 2.91DCAA 24.44 0.02 11236488 3.98BCAA 26.88 0.18 2468467 4.85DBAA 30.09 0.16 731710 3.26TCAA 39.05 0.24 4855405 10.98

BDCAA 45.13 0.04 1212887 4.78DBCAA 43.55 0.07 1064 22.20TBAA 47.44 0.25 1333 17.60

Calculated EPA MethodAnalyte MDL (µL/L) 552.2 MDL (µL/L)MCAA 0.203 0.273MBAA 0.392 0.204DCAA 0.097 0.242BCAA 0.136 0.251DBAA 0.100 0.066TCAA 0.403 0.079

BDCAA 0.159 0.091DBCAA 0.459 0.468TBAA 0.407 0.820

References 1. U.S. Environmental Protection Agency, Microbial Health Effects Tables:

Potential Adverse Health Effects from High/Long-term Exposure toHazardous Chemicals in Drinking Water, 2002.

2. U.S. Environmental Protection Agency, Method 552.1, Determination ofHaloacetic Acids and Dalapon in Drinking Water by Ion ExchangeLiquid-Solid Extraction and Gas Chromatography with ElectronCapture Detection, Rev. 1.0, 1992.

3. U.S. Environmental Protection Agency, Method 552.2, Determination ofHaloacetic Acids and Dalapon in Drinking Water by Liquid-LiquidExtraction, Derivatization, and Gas Chromatography with ElectronCapture Detection, Rev 1.0, 1995.

4. U.S. Environmental Protection Agency, Method 552.3, Determination ofHaloacetic Acids and Dalapon in Drinking Water Liquid-LiquidMicroextraction, Derivatization, and Gas Chromatography withElectron Capture Detection, Rev 1.0, 2003.

5. Slignsby, R.; Saini, C.; Pohl, C.; Jack, R. The Measurement of HaloaceticAcids in Drinking Water Using IC-MS/MS–Method Performance,Presented at the Pittsburgh Conference, New Orleans, LA, March 2008.

Table 5. Reproducibility of area and retention time in the TSQ Quantum Access for seven injectionsof 2 ppb concentration in simulated matrix

Table 6. Calculated MDL response of HAA9 on the TSQ Quantum Access

Administrator

Text Box


Analysis of Regulated Pesticides inDrinking Water Using Accela and EQuanJonathan R. Beck and Charles Yang; Thermo Fisher Scientific, San Jose, CA

Key Words

• TSQ Quantum

• AccelaLC System

• EQuan

• LC-MS/MS

• PesticideAnalysis

• Water Analysis


Introduction

Pesticides are used throughout the world to control peststhat are harmful to crops, animals, or people. Becauseof the danger of pesticides to human health and theenviron ment, regulatory agencies control their use and setpesticide residue tolerance levels. The limits of detection(LODs) for many of these substances are at the parts-per-trillion (ppt) level. In order to achieve this levelof detection, offline sample pre-concentration is oftenperformed. However, these sample preparation procedurescan be time consuming, adding as much as one to twodays to the total analysis time. Therefore, a method foronline sample pre-concentration that bypasses the offlinesample pre-concentration provides a significant timesavings over conventional methods.

We describe a method for online sample cleanup andanalysis using the Thermo Scientific EQuan system. Thismethod couples a Fast-HPLC system with two ThermoScientific Hypersil GOLD LC columns–one for pre-concentration of the sample, the second for the analyticalseparation–and an LC-MS/MS instrument. Large volumesof drinking water samples (1 mL) can be directly injectedonto the loading column for LC-MS/MS analysis, thuseliminating the need for offline sample pre-concen trationand saving overall analysis time. Using this configuration,run times of six minutes are achieved for the analysis ofa mixture of pesticides. For separation prior to analysisusing an LC-MS/MS instrument, Fast-HPLC allows forsignificantly shorter run times than conventional HPLC.

Goal

To demonstrate the use of Fast-HPLC and a large volumeinjection to analyze sub-ppb concentrations of regulatedpesticides in drinking water samples.


Sample Preparation Bottled drinking water was spiked with a mixture ofthe following pesticides: carbofuran, carbaryl, diuron,daimuron, bensulfuron-methyl, tricyclazole, azoxystrobin,halosulfuron-methyl, flazasulfuron, thiodicarb, andsiduron. Concentrations were prepared at the followinglevels: 0.5, 1, 5, 10, 50, 100, 500, and 1000 pg/mL (ppt).

No other sample treatment was performed prior toinjection. The mass transitions and collision energiesfor each compound are listed in Table 1.

HPLCFast-HPLC analysis was performed using the ThermoScientific Accela High Speed LC System. A 1 mL watersample was injected directly onto a 20 mm× 2.1 mm ID,12 µm Hypersil GOLD™ loading column in a highaqueous mobile phase at a flow rate of 1 mL/min (seeFigure 1a). After approximately one minute, a 6-portvalve on the mass spectrometer was switched via theinstrument control software. This enabled the loadingcolumn to be back flushed onto the analytical column(Hypersil GOLD 50× 2.1 mm ID, 1.9 µm), where thecompounds were separated prior to introduction into themass spectrometer (Figure 1b). After all of the compoundswere eluted from the analytical column at a flow rate of

14 151 269.21 Daimuron

24 182 435.11 Halosulfuron-methyl

15 372 404.16 Azoxystrobin

20 137 233.19 Siduron

24 182 408.08 Flazasulfuron

22 149 411.13 Bensulfuron-methyl

20 72 233.05 Diuron

10 145 202.14 Carbaryl

14 165 222.10Carbofuran

14 88 355.06 Thiodicarb

10 106 190.09 Tricyclazole

Collision Energy (eV) Product Mass (m/z) Precursor Mass (m/z) Analyte

Table 1: List of mass transitions and collision energies for each compoundanalyzed.

Figure 1a: 6-port valve positionone (load position), for loading thesample onto the loading column.

Figure 1b: 6-port valve positiontwo (inject position), for eluting thecompounds trapped on the loadingcolumn onto the analytical column.

Administrator

Text Box


850 µL/min, the 6-port valve was switched back to thestarting position. The loading and analytical columns werecleaned with a high organic phase before being re-equilibrated to their starting conditions. The total runtime for each analysis was six minutes. The mobile phasesfor the analysis were water and acetonitrile, bothcontaining 0.1% formic acid. The gradient profile for eachpump is shown in Figure 2.

The pressure at the beginning of the gradient wasmonitored. At a flow rate of 850 µL/min (at the initialgradient conditions with the flow going through only theHypersil GOLD 50× 2.1 mm, 1.9 µm column), the back -pressure for the Fast-HPLC system was approximately450 bar. For comparison, an earlier method which useda Hypersil GOLD 50× 2.1 mm, 3 µm column had abackpressure of approximately 150 bar at a flow rateof 200 µL/min.

MSMS analysis was carried out on the Thermo Scientific TSQQuantum Access triple stage quadrupole mass pectrometerwith a heated electrospray ionization (H-ESI) probe. TheMS conditions were as follows:Ion source polarity: Positive ion mode Spray voltage: 4000 V Vaporizer temperature: 450°CSheath gas pressure (N2): 50 units Auxiliary gas pressure (N2): 50 unitsIon transfer tube temperature: 380°CCollision Gas (Ar): 1.0 mTorrQ1/Q3 Peak Resolution: 0.7 uScan Width: 0.002 u


Chromatograms for the calibration standard at a con -cen tration of 500 pg/mL are shown in Figure 3. In theFast-HPLC run, all 11 of the individual analytes wereeluted before three minutes. In contrast, none of theanalytes in the standard HPLC run were eluted untilnearly eight minutes into the run. Further optimizationof the chroma tography for the Fast-HPLC would produceeven shorter run times.

Calibration curves for all 11 compounds weregenerated using Thermo Scientific LCQUAN 2.5 software.Excellent linearity was achieved for all of the compoundsanalyzed in this experiment. Figure 4 shows arepresentative calibration curve for the compoundazoxystrobin over the concentration range 0.5 to 1000pg/mL (ppt). The calibration curve fit parameters and thelimits of detection for the analytes are summa rized inTable 2. The final column in the table lists the MinimumPerformance Reporting Limit (MPRL) for thesecompounds as set by the Japanese Ministry of Health,Labour, and Welfare1. All of the compounds were detectedand quantified at levels well below these regulatoryrequirements.

Figure 2: Gradient profiles for the two LC pumps used in this experiment.The Fast-HPLC pump gradient is shown on the left, and the loading pumpgradient is show on the right.

Figure 3: Chromatograms showing the SRMs for each of the components inthe mixture. Two different HPLC conditions are shown: the Fast-HPLC runand the standard HPLC run. All compounds in the Fast-HPLC run are elutedin less than three minutes (circled in green). Those in the standard HPLCrun are eluted much later (circled in red). These chromatograms representa calibration level of 500 pg/mL (ppt).

Figure 4: Calibration curve for the compound azoxystrobin. This calibrationcurve covers the range from 0.5 to 1000 pg/mL (ppt)

Administrator

Text Box


Conclusion

The implementation of Fast-HPLC, coupled with theonline pre-concentration and sample preparation tech -nique EQuan, yielded analysis of 11 pesticides in drinkingwater in less than one-third the time of conventionalHPLC analysis. All of the compounds eluted within threeminutes, which included a one-minute loading time forthe sample to be pre-concentrated on the loading column.The total run time for the analysis was six minutes. TheFast-HPLC method can be further shortened to producefaster chromatographic run times.

The use of large volume injections achieved resultsbelow the MPRL regulatory requirements for each of the11 pesticides. Because the limits of detection were muchlower than the MPRL values, the integrated peaks yieldedexcellent signal-to-noise ratios and allowed for confidencein reporting the results.

Reference1 http://www.mhlw.go.jp/index.html (Japanese language version),

http://www.mhlw.go.jp/english/index.html (English language version)

50000.50.9974Azoxystrobin

30000.50.9973Siduron

30010.9944Flazasulfuron

40000.50.9933Bensulfuron-methyl

2001000.9978Diuron

5001000.9345Carbaryl

5010.9928Carbofuran

80050.9930Thiodicarb

8000.50.9972Tricyclazole

MPRL(ppt)

Limit ofDetection (ppt)R2Analyte

Table 2: List of calibration curve fit parameters, limits of detection, andMinimum Performance Reporting Levels (MPRL) for each compound fromthe Japanese Ministry of Health, Labour and Welfare. All calibrations werecarried our using a linear curve fit and a weighting factor of 1/X.

Administrator

Text Box






















www.thermo.com

AN62449_E 10/08S



Legal Notices©2007 Thermo Fisher Scientific Inc. All rights reserved. All other trademarks are the property of Thermo Fisher Scientific Inc. and its subsidiaries. Thisinformation is presented as an example of the capabilities of Thermo Fisher Scientific Inc. products. It is not intended to encourage use of these products inany manners that might infringe the intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are availablein all countries. Please consult your local sales representative for details.


Administrator

Text Box


Analysis of Triazine Pesticides in DrinkingWater Using LC-MS/MS (EPA Method 536.0)Jonathan Beck, Thermo Fisher Scientific, San Jose, CA, USA

Introduction

The US EPA has recently issued a draft form of a proposedmethod for the analysis of triazine compounds in drinkingwater.1 This method uses a simple method to directly analyzetriazine compounds using LC-MS/MS without requiring anysolid phase extraction (SPE) or other lengthy samplepreparation steps. This application note demonstrates theanalysis of these compounds over the concentration range0.25 – 5.0 ng/mL (ppb) using the Thermo Scientific TSQQuantum Access triple stage quadrupole mass spectrometerand the Thermo Scientific Accela HPLC system.


The following triazine and triazine degradates were analyzed:Atrazine, Atrazine-desethyl, Atrazine-desisopropyl,Cyanazine, Propazine, and Simazine, purchased fromSigma-Aldrich, St. Louis, MO, and Ultra Scientific, NorthKingstown, RI. The following internal standards were used:Atrazine-d5, Atrazine-desethyl-d7, Atrazine-desisopropyl-d5,Cyanazine-d5, Propazine-d14, and Simazine-d10, purchasedfrom C/D/N Isotopes, Inc., Pointe-Claire, Quebec, Canada.Standards and internal standard stocks were prepared insolutions of methanol and diluted to their appropriateconcentrations prior to analysis.

Sample Preparation

While no SPE was required for this method, samples weretreated as per the EPA’s draft method. The method callsfor the addition of ammonium acetate at 20 mM for pHadjustment and dechlorination and sodium omadine at 64 mg/L to prevent microbial degradation, both purchasedfrom Sigma-Aldrich, St. Louis, MO. All samples wereprepared in reagent water. All samples were spiked withthe internal standard solution, resulting in a finalconcentration of 5 ng/mL (ppb) for each internal standard.Calibration standards were prepared at the followinglevels: 0.25, 0.5, 1, 2, 2.5 and 5 ng/mL.

HPLC Conditions

Column: Thermo Scientific Hypersil GOLD 100 x 2.1 mm, 3 µmSolvent A: 5 mM Ammonium AcetateSolvent B: MethanolFlow Rate: 400 µL/minInjection Volume: 100 µLHPLC Gradient: Time %A %B

0:00 98 210:00 98 220:00 10 9025:00 10 9025:06 98 230:00 98 2

Mass Spectrometer Conditions

Ionization Source: Positive Electrospray Sheath Gas: 30 arbitrary unitsAuxiliary Gas: 10 arbitrary unitsESI Voltage: 3.5 kVIon Transfer Tube Temperature: 350 °CCollision Gas: 1.5 mTorrQ1/Q3 Peak Resolution: 0.7 DaScan Width: 0.01 Da

MS Parameters

Precursor Product Collision TubeCompound Mass Mass Energy Lens

Atrazine-desisopropyl 174 132 17 90Atrazine-desethyl 188 146 16 95Simazine 202 124 17 80Atrazine 216 174 16 85Propazine 230 124 17 80Cyanazine 241 214 15 100Atrazine-desisopropyl-d5 179 137 17 85Atrazine-desethyl-d7 195 147 17 95Simazine-d10 212 137 19 95Atrazine-d5 221 179 17 95Propazine-d14 244 196 18 95Cyanazine-d5 246 219 16 100

Key Words

• Drinking WaterAnalysis

• Herbicides


• Triazines



Administrator

Text Box



The triazine compounds eluted from the LC column in 20 minutes. A chromatogram of each compound and theinternal standards is shown in Figure 1. All peaks arechromatographically resolved from one another. Calibrationcurves were generated for each compound over the range0.25-5 ppb. All calibration curves exhibited excellentlinearity, ranging from 0.9964 for Atrazine-desethyl to0.9982 for Atrazine. The calibration curve for Simazine isshown in Figure 2. The other compounds exhibit similarlinearity, and are not shown in this application note.

Conclusion

The TSQ Quantum Access LC-MS/MS is an excellentchoice for the analysis of triazine compounds and theirdegradates. Linearity over the entire calibration range of0.25 to 5 ppb is observed. Separation of all the analytes isachieved with the Hypersil GOLD™ column allowing forunambiguous identification and quantitation of all of thecompounds in this application note.

References1. Smith, G.A., Pepich, B.V., Munch, D.J. “Determination of Triazine

Pesticides and their Degradates in Drinking Water by LiquidChromatography Electrospray Ionization Mass Spectrometry(LC/ESI/MS)” Draft 5.0, April 2007.







Africa+43 1 333 5034 127Australia+61 2 8844 9500Austria+43 1 333 50340Belgium+32 2 482 30 30Canada+1 800 530 8447China+86 10 8419 3588Denmark+45 70 23 62 60 Europe-Other+43 1 333 5034 127France+33 1 60 92 48 00Germany+49 6103 408 1014India+91 22 6742 9434Italy+39 02 950 591Japan +81 45 453 9100Latin America+1 608 276 5659Middle East+43 1 333 5034 127Netherlands+31 76 579 55 55South Africa+27 11 570 1840Spain +34 914 845 965Sweden / Norway /Finland+46 8 556 468 00Switzerland+41 61 48784 00UK+44 1442 233555USA+1 800 532 4752

www.thermo.com

AN62880_E 09/08M





Figure 1: Chromatogram of the triazine compounds at 2 ppb, and theirinternal standards

Figure 2: Calibration curve for Simazine, 0.25-5 ppb

Administrator

Text Box


Determination of Different Classes of PesticideResidues in Processed Fruits and Vegetablesby LC-MS Using the TSQ Quantum UltraAccording to EU Directive 91/414 EECEleni Botitsi, Anastasios Economou, Spyros Antoniou, and Despina Tsipi;General Chemical State Laboratory, Pesticide Residues Laboratory, Athens, Greece

Key Words


• Surveyor HPLCSystem

• H-SRM

• Food Safety

• PesticideResidues

• Sensitivity


Introduction

A diet rich in fruits and vegetables is thought to reducethe risk of some types of cancer, atherosclerosis, and heartdisease. However, commercially grown produce oftencontains high levels of pesticide residues that can lead toserious health problems when consumed. Due in large partto growing public concern over the amount of pesticideresidues in foods, the European Union (EU) has enactedseveral directives to fix Maximum Residue Limits (MRLs)for different pesticide residues in food of plant origin.MRLs represent the maximum amount of pesticideresidues that might be expected in a commodity producedunder conditions of good agricultural practice andtypically range between 0.01 mg/kg and 10 mg/kg1.Although MRLs are not maximum toxicological limits,care is taken to ensure that these maximum levels do notgenerate toxicological concerns. Thus far, MRLs havebeen set for approximately 250 active substances.To cover the full variety of agricultural raw commodities(approximately 260 products of plant and animal origin),MRLs must be established for more than 260,000pesticide/commodity combinations1,2.

In the EU, pesticides are regulated principally byDirective 91/414/EEC concerning the placing of plant-protection products on the market3. According to thislegislation, chemical substances or micro-organisms inpesticides are approved for use only if they have under gonea peer-reviewed safety assessment. All foodstuffs intendedfor human consumption or animal feed in the EU are nowsubject to a maximum residue limit for pesticides to protecthuman and animal health. Regulation (EC) 396/20054

consolidates in a single act all the limits applicable tovarious types of food and feed. It establishes MRLs forproducts of plant and animal origin at the Communitylevel, taking into account good agricultural practices.It was based on several substantial amendments in theCouncil Directives:

• 76/895/EEC5, which relates to the fixing of maximumlevels for pesticide residues in and on specific fruitsand vegetables

• 86/362/EEC6 for cereals and cereal products

• 86/363/EEC7 for products of animal origin

• 90/642/EEC8 for plant products

Additionally, more stringent legislation has beenestablished concerning pesticides in baby food. Since1999, the EU has introduced the Commission Directives1999/39/EC9 and 1999/50/EC10, which limit all pesticideresidues to an MRL value of 0.010 mg/kg in processedcereal-based foods and in fruit and vegetables intended forthe production of baby foods. MRLs below 0.010 mg/kghave been established for a few pesticides of highertoxicity, while the use of certain very toxic pesticides hasbeen completely prohibited in the production of babyfoods, as underlined in Commission Directives2003/13/EC11 and 2006/125/EC12.

New“active”ingredients entering the market toreplace compounds banned by Directive 91/414/ EECpossess considerably different physicochemical properties,and thus demand the development of multi-residueanalytical methods. Analytical methodologies used todetermine pesticide residues in foods must be capableof quantifying very low levels of residues as well asconfirming their identity. This task becomes moredifficult as MRLs are decreased and the number of targetpesticides and metabolites increases. Therefore, thechallenge is to develop a sensitive, cost-effective, multi-residue analytical method that can quickly identify andconfirm pesticide residues belonging to various chemicalclasses in food products. At the same time, the methodmust accurately quantify these residues at low levels, thusfulfilling the performance criteria described in“MethodValidation and Quality Control Procedures for PesticideResidues Analysis in Food and Feed,” EuropeanCommission Document SANCO 2007/313113.

Goal

To develop a multi-residue liquid chromatography-electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) method for the detection and quantificationof 45 pesticides, including parent compounds and theirtransformation products from different chemical classes,in various food matrices.


LC-ESI-MS/MS is the analytical technique of choice toassay environmental and food matrices with high sensi tivityand selectivity. The technique is especially well-suited forthe identification and quantification of polar and thermallylabile pesticides and metabolites down to mg/kg levels.

The pesticides included in this study are listed inTable 1.

Administrator

Text Box


Table 1: Retention times and compound-specific ESI(+)-MS/MS parameters

t R Parent ion Quantifier ion Quantifier ionCompound (min)a (m/z) (m/z) (m/z)Acephate 2.5 184 143 (10V) 125 (10V)Aldicarbb (Na+) 19.7 213 89 (22 V) 116 (22 V)Aldicarb-sulfoxide (Na+) 2.7 229 166 (9V) 109 (15V)Aldicarb-sulfone (Na+) 4.1 245 109 (25V) 166 (25V)Acetamiprid 17.3 223.0 126 (21 V) 90 (35 V)Azoxystrobin 28.8 404.2 372 (16 V) 344 (26 V)Carbaryl 24.0 202.1 145 (25 V) 127 (25 V)Carbofuran 23.1 222.1 123 (24 V) 165 (24 V)3-hydroxy-carbofuran (-H2O) 15.8 220 135 (15V) 163 (15V)Chlorpropham 29.6 214.0 172 (12 V) 154 (19 V)Carbendazim + benomylb 2.5 192.0 160 (22 V) 132 (31 V)Cyprodinil 25.3 226.1 93 (40 V) 77 (46 V)Demeton-S 26.0 259.0 89 (22 V) 116 (22 V)Demeton-S-methyl 22.8 253.0 61 (40 V) 89 (20 V)Demeton-S-methyl-sulfone 7.0 263.0 109 (30 V) 169 (20 V)Demeton-S-methyl-sulfoxide 3.3 247.0 169 (17 V) 109 (29 V)Dimethomorph Ac / Bc 26.8 388.1 301 (23 V) 165 (35 V)

27.3Disulfoton 33.8 275.0 89 (15 V) 61 (30 V)Disulfoton-sulfone 27.1 307.0 125 (20 V) 153 (20 V)Disulfoton-sulfoxide 24.2 291.0 185 (15 V) 157 (25 V)Ethoprofos 29.2 243.0 131 (21 V) 173 (21 V)Fenhexamidd 29.0 302.0 97 (22 V)

304.0 97 (26 V)Flusilazole 29.8 316.1 247 (21 V) 165 (31 V)Imazalil 21.4 297.0 159 (24 V) 255 (25 V)Imidachloprid 15.3 256.1 209 (22 V) 175 (22 V)Kresoxim-methyl 31.8 314.0 222 (14 V) 116 (19 V)Metalaxyl 24.8 280.1 220 (15 V) 192 (25 V)Methiocarb 27.6 226.0 169 (11 V) 121 (19 V)Methiocarb sulfoxide (Na+) 6.6 185.0 122 (23 V) 170 (23 V)Methomyl 5.0 163.0 106 (12 V) 88 (12 V)Myclobutanil 28.7 289.0 125 (35 V) 70 (25 V)Oxamyl (Na+) 4.2 242 70 (20V) 121 (20V)Penconazole 30.0 284.0 159 (35 V) 70 (35 V)Pirimicarb 7.3 239.1 182 (15 V) 72 (30 V)Propiconazole 30.8 342.0 159 (31 V) 69 (31 V)Propoxur 22.7 210.1 111 (17 V) 168 (10 V)Pyrimethanil 21.2 200.0 182 (35 V) 168 (35 V)Tetraconazoled 29.4 372.0 159 (38 V)

374.0 161 (31 V)Thiabendazole 2.5 202.0 131 (36 V) 175 (36 V)Thiachloprid 21.0 253.0 99 (45 V) 126 (25 V)Thiodicarb 23.3 355.0 88 (20 v) 108 (20 V)Thiophanate-methyl 22.5 343.0 151 (23 V) 311 (15 V)Triadimefon 28.9 294.1 197 (19 V) 225 (19 V)Triadimenol Ac/ Bc 27.1/27.5 296.1 70 (16 V) 99 (16 V)Triazophos 30.4 314.1 162 (19 V) 119 (33 V)

a Retention time b Benomyl was measured as carbendazim14

c Dimethomorph and triadimenol exist as two isomers with different retention timesd For fenhexamid and tetraconazole, the isotopic parent ions were selected due to the lack of a second sound transition

Administrator

Text Box


Sample Preparation A stock mix solution of all the pesticides was preparedat a concentration of 1 mg/L. Calibration solutions in theconcentration range 0.5-100 µg/L were prepared by serialdilution of the stock solution.

Samples were prepared for analysis using extractionwith ethyl acetate. Individual samples of fruits andvegetables were first homogenized. After homogenization,a 10.0 g sample was extracted using ethyl acetate andanhydrous sodium sulfate. The mixture was ultra-sonicated for 20 minutes. The mixture was filtratedthrough a thin layer of anhydrous sodium sulfate andthe filtrate was evaporated. The extracts were thenreconstituted in 5 mL of methanol. The solution wasdiluted with water and then filtered through a 0.45 µmsyringe filter14.

HPLCHPLC analysis was performed using the Thermo ScientificSurveyor HPLC System. Each 20 µL sample was injectedonto a 150 × 2.1 mm, 3.5 µm, C18 HPLC columnequipped with a 10 × 2.1 mm, 3.5 µm, C18 HPLC guardcolumn. A gradient LC method used mobile phases A(0.1% formic acid) and B (0.1% formic acid inacetonitrile) at a flow rate of 0.2 mL/min. The gradientwas: 0–3 min A:B = 90:10 (v/v), 3 –31 min A:B = 90:10(v/v) to A:B = 10:90 (v/v), 31–36 min A:B = 10:90 (v/v),36 –36.5 min A:B = 10:90 (v/v) to A:B = 90:10 (v/v),36.5 – 45 min A:B = 90:10 (v/v).

MS MS analysis was carried out on a Thermo Scientific TSQQuantum Ultra triple stage quadrupole mass spectrometerwith an electrospray ionization source.

The MS conditions were as follows:Ion source polarity: Positive Spray voltage: 4000 VSheath gas pressure (N2): 40 unitsAuxiliary gas pressure (N2): 10 unitsIon transfer tube temperature: 350°CCollision gas pressure (Ar): 1.0 mTorrQ1 resolution: 0.2 FWHM (H-SRM)Q3 resolution: 0.7 FWHMScan Type: H-SRMDwell time: 20–50 ms

The LC-MS/MS method was developed according tothe scheme shown in Figure 1. The run was divided intofour time segments based on the retention times of thetarget compounds. Multiple scan events were includedin each time segment. For each target compound, theprotonated molecule [M+H]+ was usually investigated,except in the cases of compounds where the adduct[M+Na]+ was the base peak in the ESI(+) spectra. Twotransitions were selected per compound in order toperform quantification and identification simultaneously.

The SRM transitions that were monitored aresummarized in Table 1. Identification criteria for thetarget compounds were based on the LC retention time(tR) and on the ratio of the two monitored transitionsfor each compound.13,14

Figure 1: LC-ESI-MS/MS method

Administrator

Text Box



Although LC-MS/MS is a selective technique, interferencesdue to isobaric compounds can appear in chromatograms.These isobaric interferences increase the chemicalbackground and can make it difficult to integrate thedesired analyte peak reproducibly. Among the compoundsincluded in this study were three sets of isobariccompounds and one set of compounds that share the samefragment ions, which increases the likelihood of cross-talk.Therefore, to eliminate the noise and lower the detectionlimits, all of the assays in this study were run in theHighly Selective Reaction Monitoring (H-SRM) modewith the Q1 FWHM peak width set at 0.214.

The H-SRM chromatograms of a mix solution ofcertain pesticides at a concentration of 1 µg/L are shownin Figure 2. Linearity of the method was proven for allcases because the R2 values were usually greater than 0.99for the linear regression equations (1/x weighted) in the

concentration ranges tested. The instrumental detectionlimits (IDLs) were, in most cases, below 0.5 µg/L. Figure 3displays the linearity plots of selected compounds.Linearity data for certain compounds are summarizedin Table 2.

Using the H-SRM mode reduced the matrix effectsby minimizing the chemical noise caused by co-elutingisobaric compounds. Consequently, the signal-to-noiseratio was enhanced in the complicated food matrices.This effect can be observed in the chromatograms inFigure 4, which show the analysis of a peas sample in theSRM and H-SRM modes. The top two SRM chroma -tograms illustrate the background in a blank peas extractwhereas the bottom two SRM chromatograms show thepeaks for methomyl in a peas extract spiked with 1 ppbof methomyl. The narrower window of the Q1 set at0.2 FWHM in the H-SRM mode improves the selectivityof the analysis and increases the signal-to-noise ratio.

Figure 2: SRM chromatograms for certain pesticides of the standard mix solution

Administrator

Text Box


0

5000000

10000000

Imidachlopridy = 12773x - 5101R2 = 0.9957

Thiabendazoley = 53717x - 4547R2 = 0.9933

Demeton-Sy = 37613x - 9921R2 = 0.9972

Disulfotony = 1646x - 11944R2 = 0.9903

Carbaryly = 46572x + 36911R2 = 0.9944

Acephatey = 77684x + 11680R2 = 0.9967

0

700000

1400000

0 50 100µg/L

µg/L

0

3000000

6000000

0

400000

0 50 100

0

2000000

4000000

0

90000

180000

Cl

m/z 256

H+

N N

CH2-N

NO2

NH

HN

OO

m/z 202

H+

m/z 184

H+

CH3O - P - NHCOCH3

OCH3

O

m/z 275

H+

CH3CH2O - P - SCH2CH2SCH2CH3

OCH2CH3

Sm/z 259

CH3CH2O - P - SCH2CH2SCH2CH3

H+

OCH2CH3

O

m/z 202

H+

NNH

S

N

0 10050µg/L

µg/L0 50 100

µg/L0 50 100

µg/L0 50 100

Figure 3: Linearity plots for certain compounds

Figure 4: H-SRM and SRM chromatograms of methomyl in pea sample matrix

Administrator

Text Box


The matrix-matched calibration curves of methomylat different Q1 settings are shown in Figure 5 and dataof the calibration curves are listed in Table 3. The signalitself is reduced by a factor of two when the Q1 FWHMpeak width is changed from 0.7 to 0.2, yet the linearityand accuracy are improved (as demonstrated by thecorrelation coefficients and back-calculated values ofthe matrix-matched standards at the low concentrationlevels, in Table 3).

Some samples were found to contain pesticide residues.Figure 6 displays SRM chromatograms of a sample offrozen peas that contained residues of triazophos andmyclobutanil. The confirmation of identity was basedon the ion ratio of monitored transitions in the sampleand in the standard solution according to the EUGuidelines for pesticide residues monitoring.13 Theconcentrations of the residues found in the sample werebelow the Maximum Residue Limits (MRLs)1,2.

Table 2: Linearity data and instrumental detection limits (IDLs) for certain pesticides

Peas Matrix Peas Matrix Peas Matrix Peas Matrix0.1 g/mL 0.1 g/mL 0.2 g/mL 0.2 g/mL

Q1: 0.2 FWHM Q1: 0.7 FWHM Q1: 0.2 FWHM Q1: 0.7 FWHM

1/xY = -2469.2 + Y = 4631.3 + Y = -3845.1 + Y = 10244 +

10863 X 18381.3 X 8212 X 19142 XR2 0.9966 0.9851 0.9945 0.9861

Accuracy of Matrix-Matched Calibration Curves (1/x)

1 µg/L 0.91 µg/L (91%) 0.68 µg/L (68%) 0.87 µg/L (87%) 1.24 µg/L (124%)5 µg/L 4.89 µg/L (97%) 4.74 µg/L (94%) 5.26 µg/L (105%) 4.56 µg/L (91%)

10 µg/L 9.78 µg/L (97%) 11.5 µg/L (85%) 10.5 µg/L (95%) 9.05 µg/L (90%)

Table 3: Linearity and accuracy data for methomyl in pea matrix

Linear regression Concentration IDLsCompound equations range (µg/L) R2 (µg/L)

Acephate Y=116806 +77683.7 X (1-100) 0.9967 0.5Aldicarb Y=-590.6 +1115.4 X (1-100) 0.9900 0.7Azoxystrobin Y=-363884 + 213698 X (0.5-100) 0.9912 0.2Carbaryl Y=36911 + 46572 X (0.5-100) 0.9903 0.3Carbendazim Y=10192 + 211684 X (0.5-100) 0.9964 0.1Carbofuran Y=11107 + 161251 X (0.5-100) 0.9920 0.2Chlorpropham Y=-5289 + 6893.5 X (1-100) 0.9954 0.6Cyprodinil Y=-57425 + 30565.4 X (0.5-50) 0.9931 0.3Demeton-S Y=-9921 + 37615 X (0.5-100) 0.9972 0.3Disulfoton Y=-11944 + 1646.2 X (5-100) 0.9903 1.5Disulfoton Sulfoxide Y=40274 + 141033 X (0.5-100) 0.9961 0.4Disulfoton Sulfone Y=-1633.2 + 8994 X (0.5-100) 0.9904 0.4Ethoprofos Y=-10106 + 40922 X (0.5-100) 0.9940 0.3Imidachloprid Y=-5101.1 +12773.2 X (0.5-100) 0.9957 0.3Kresoxim-methyl Y=-7877.4 + 3056.8 X (2-100) 0.9900 1.0Metalaxyl Y=28427.5 +117245 X (0.5-100) 0.9964 0.3Methiocarb Y=4861 + 48380.4 X (0.5-100) 0.9921 0.3Methomyl Y=-2440.7 +13847.8 X (0.5-100) 0.9990 0.4Myclobutanil Y=-16905.7 +10101.5 X (0.5-100) 0.9953 0.4Pirimicarb Y=23403 +168260 X (0.5-100) 0.9953 0.2Propoxur Y=9181 + 151300 X (0.5-100) 0.9947 0.2Pyrimethanil Y=-4723.7 + 9197.2 X (0.5-100) 0.9900 0.4Thiabendazole Y=-4546.8 + 53716.7 X (0.5-100) 0.9933 0.3Triazophos Y=-18350 +134057 X (0.5-100) 0.9954 0.3

Administrator

Text Box


Conclusion

A multi-residue LC-ESI-MS/MS method was developed forthe reliable confirmation and quantification of pesticidesfrom different chemical classes at low ppb levels in foodmatrices. The method uses the Highly Selective ReactionMonitoring (H-SRM) mode of the TSQ Quantum Ultratriple quadrupole mass spectrometer to effectively reducethe background interference and improve the signal-to-

noise ratios. For the pesticides investigated, satisfactoryprecision and accuracy were achieved and Limit ofQuantitation (LOQ) values of 0.010 mg/kg wereestablished. The method can be expanded to include morepesticides and their metabolites to improve the range ofpesticide residues monitored in food commodities.

0.00E+00

5.00E+04

1.00E+05

1.50E+05

2.00E+05

2.50E+05

0 1 2 3 4 5 6 7 8 9 10

µg / L

0.2 g/mL pea matrix at Q1 0.2 (FWHM) 0.2 g/mL pea matrix at Q1 0.7 (FWHM) 0.1 g/mL pea matrix at Q1 0.2 (FWHM) 0.1 g/mL pea matrix at Q1 0.7 (FWHM)

Q1: 0.2 Da

Q1: 0.7 Da

A. standard solution of pesticides B. Pea sample extract

TRIAZOPHOSTIC: 314.1 � 162, 119

SRM 1: 314.1 � 119[119] / [162] = 0.16*

SRM 1: 314.1 � 119[119] / [162] = 0.17

SRM 2: 314.1 � 162

TIC: 289 � 125, 70

SRM 1: 289 � 70

SRM 2: 289 � 125

TIC: 314.1 � 162, 119

SRM 2: 314.1 � 162

MYCLOBUTANILTIC: 289 � 125, 70

SRM 1: 289 � 70

SRM 2: 289 � 125

NL: 2.8E4

NL: 4.2E3

NL: 2.6E4

NL: 1.9E4

NL: 9.5E3

NL: 1.1E4

NL: 3.3E3

NL: 1.8E3

NL: 1.8E3

NL: 1.3E4

NL: 1.8E3

NL: 1.8E4

[70] / [125] = 0.91* [70] / [125] = 0.92*

Figure 5: Matrix-matched calibration curves of methomyl in pea extract at Q1: 0.2 (FWHM) and 0.7 (FWHM)

Figure 6: LC-ESI-SRM chromatograms of frozen pea sample extract, with residues of triazophos and myclobutanil

Administrator

Text Box









www.thermo.com

AN62726_E 05/08S





References1 http://www.europa.eu.int/comm/food/plant/protection/pesticides/index_en.htm2 European Food Safety Authority (EFSA), www.efsa.europa.eu3 Commission Directive 91/414/EEC of 15 July 1991 concerning the placingof plant protection products on the market. Official Journal of theEuropean Communities L 230: 1-32.

4 European Parliament and Council, Regulation (EC) 396/2005 of 23 Feb -ruary 2005, Maximum Residue Levels of Pesticides in Products of Plant orAnimal Origin. Official Journal of the European Communities L 70: 1-16.

5 Commission Directive 76/895/EEC of 23 November 1976, relating to thefixing of maximum levels for pesticide residues in and on fruit andvegetables. Official Journal of the European Communities L 340: 26-31.

6 Commission Directive 86/362/EEC of 24 July 1986 on the fixing ofmaximum levels for pesticide residues in and on cereals. Official Journalof the European Communities L 221: 37-42.

7 Commission Directive 86/363/EEC of 24 July 1986 on the fixing ofmaximum levels for pesticide in and on foodstuffs of animal origins.Official Journal of the European Communities L 221: 43-47.

8 Commission Directive 90/342/EEC of 27 November 1990 on the fixing ofmaximum levels for pesticide residues in and on certain products of plantorigin, including fruit and vegetables. Official Journal of the EuropeanCommunities L 350: 71-79.

9 Commission Directive 1999/39/EC of 6 May 1999 Amending Directive96/5/EC on processed cereal-based foods and baby foods for infants andyoung children. Official Journal of the European Communities L 124: p.8-10.

10 Commission Directive 1999/50/EC of 25 May 1999 Amending Directive91/321/EEC on infant formulae and follow-on formulae. Official Journalof the European Communities L 139: 29-31.

11 Commission Directive 2003/13/EC of 10 February 2003 AmendingDirective 96/5/EC on processed cereal-based foods and baby foods forinfants and young children. Official Journal of the European CommunitiesL 41: 33-35.

12 Commission Directive 2006/125/EC of 5 December 2006 on processedcereal-based foods and baby foods for infants and young children. OfficialJournal of the European Communities L 33: 16-35.

13 “Method Validation and Quality Control Procedures for Pesticide ResiduesAnalysis in Food and Feed” European Commission Document No.SANCO/2007/3131.

14 Helen Botitsi, Anastasios Economou and Despina Tsipi. “Development andvalidation of a multi-residue method for the determination of pesticides inprocessed fruits and vegetables using liquid chromatography–electrosprayionization tandem mass spectrometry.” Anal BioAnal Chem 2007, 389,1685-1695.

Administrator

Text Box


LC-MS/MS Analysis of Herbicides in DrinkingWater at Femtogram Levels Using 20 mL EQuanDirect Injection TechniquesJonathan R. Beck, Charles Yang, Thermo Fisher Scientific, San Jose, CA, USA

Introduction

As concerns grow over the toxic effects of herbicides andother chemicals in our environment, the need to accuratelymonitor these substances in drinking water and foodsbecomes even more critical. LC-MS/MS is routinely usedby the environmental and food industries to identify andquantify pesticide and herbicide residues. However, thismethod typically requires extensive offline samplepreconcentration methods, which can be expensive andtime-consuming, to meet the stringent requirements andlow limits of detection set forth by federal and internationalregulatory authorities. An online preconcentration andcleanup method has been developed that improves bothsensitivity and precision and yields unmatched throughput.

The Thermo Scientific EQuan system for onlinesample cleanup and analysis consists of a triple quadrupolemass spectrometer with an electrospray ionization source(ESI), two LC quaternary pumps, an autosampler, and twoLC columns having C18 selectivity – one for preconcentrationof the sample, the second for analytical separation. A 6-port valve switches between the columns and is controlledby the instrument software. In addition to quantitativeinformation, qualitative full scan product ion spectra arecollected in the same analytical run and data file, using atechnique called Reverse Energy Ramp (RER). This fullscan spectrum provides additional confirmatory informationfor the compounds being analyzed. The resulting production spectra can be library searched for positive identification,or ion ratios can be used to confirm the presence of aparticular compound, helping to eliminate “false positive”samples. This method uses drinking water for direct injectiononto the loading column, with no sample preparation oroffline concentration. This application note provides acomparison of the online sample preconcentration of 1 mL,5 mL, and 20 mL injections of drinking water samplesspiked with herbicide compounds.

Goal

To compare different large volume injections using a loadingcolumn and an analytical column with two HPLC pumps.


Sample Preparation Drinking water containing 0.1% formic acid was spikedwith a mixture of the following herbicides: ametryn, atraton,atrazine, prometon, prometryn, propazine, secbumeton,simetryn, simazine, terbuthylazine, and terbutryn (UltraScientific, North Kingstown, RI). The concentrations ofthe herbicides in the spiked water ranged from 0.1 pg/mLto 10 pg/mL. Calibration standards were prepared at thefollowing concentrations: 0.1, 0.5, 1.0, 5.0, and 10.0 pg/mL.

HPLCSpiked water samples and blank water samples (1 mL, 5 mL,or 20 mL) were injected directly onto a loading column(Thermo Scientific Hypersil GOLD 20 mm x 2.1 mm ID,12 µm) using an HTC PAL autosampler (CTC Analytics,Zwingen, Switzerland). After the sample was completelytransferred from the sample loop to the loading column, a6-port valve was switched to enable the loading column tobe back flushed onto the analytical column (HypersilGOLD™ 50 mm x 2.1 mm ID, 3 µm), where thecompounds were separated prior to introduction into themass spectrometer. After all of the compounds wereeluted, the valve was switched back to the startingposition. The loading and analytical columns were cleanedwith a high organic phase before being re-equilibrated totheir starting conditions (Figure 1a and 1b). Control andtiming of the 6-port valve was through the computer datasystem, Thermo Scientific LCQUAN.

Key Words


• EQuan System

• Herbicides

• QED

• Water Analysis


Figure 1: a) 6-port valve in position 1 (load position), for loading the sample onto the loading column. b) 6-port valve in position 2 (inject position), for eluting thecompounds trapped on the loading column onto the analytical column.

a b

Administrator

Text Box


Figure 2: The method setup screen for the CTC Autosampler, showing thecapability to perform multiple injections from the same vial. The red boxhighlights the parameters used to control the number of syringe fills fromtwo consecutive vials. In this example, a total of 20 mL will be injected.

Slightly different LC programs were used in eachmethod, depending on the volume of the sample injected.The loading pump flow rates ranged from 1 mL/min for 1 mL samples to 5 mL/min for 20 mL samples. This allowedthe run times at the higher injection volumes to be shortenedbecause the time to transfer the sample from the sampleloop to the loading column depends on the flow rate. Thesame LC program was used for the analytical column.

Two HPLC pumps were used for the analysis: one fortransferring the sample from the injection loop to theloading column, and one for back flushing the compoundsoff of the loading column and separating them on theanalytical column. The loading pump was a ThermoScientific Surveyor Plus LC pump and the analytical pumpwas a Thermo Scientific Accela U-HPLC pump.

The HTC autosampler was equipped with a 5 mLsyringe. To accommodate larger injection volumes (> 5 mL),a CTC™ macro sequence was programmed to allow formultiple syringe fills and deliveries to the sample loopfrom a 10 mL vial. For 20 mL samples, two 10 mL vialswere used and the macro allowed sampling from adjacentvials filled with the same sample. The macro is shown inFigure 2. Because this multi-sampling scheme can be quitetime consuming, the ability to perform “look-ahead”injections allows for significant time savings. The loop can be switched to an offline position during a run, andsubsequent samples can be prepared and injected while asample is being run.

MSMS analysis was carried out on a Thermo Scientific TSQQuantum Access triple stage quadrupole massspectrometer with an electrospray ionization (ESI) source.The MS conditions were as follows:

Ion Source Polarity: Positive ion modeSpray Voltage: 4000 VIon Transfer Tube Temperature: 300 °CSheath Gas Pressure: 30 arbitrary unitsAuxiliary Gas Pressure: 5 arbitrary unitsCollision Gas (Ar): 1.5 mTorrQ1/Q3 Peak Resolution: 0.7 DaScan Width: 0.002 Da

Quantitative and qualitative data were collected in thesame run and data file.


Chromatograms of the herbicide simazine at three differentinjection volumes are shown in Figure 3. A very small peakcan be seen for the 1 mL injection volume; however, theintegration is not shown in the chromatogram. Injectionsat higher volumes show superior signal-to-noise ratios andintensity, which allow for analysis of very low concentrationsamples (pg/mL and sub pg/mL). To test the reproducibilityof the multiple syringe fill method with a 20 mL loop,eight replicate injections were performed using the 1 pg/mLcalibration standard. The results of this study are shownin Table 1. No internal standard was used in this analysis;however, if one were to be included, the % RelativeStandard Deviations (RSD) values would likely improve.Table 1 also shows the peak areas and calculated differencein peak areas between the 1, 5, and 20 mL injections.

FPO

Volume to be pulled per syringe “injection”

Number of syringefills from the first vial.

Number of syringe fillsfrom the second vial (optional).

Administrator

Text Box


Factor Factor %RSD Compound Area, 1 mL Area, 5 mL Area, 20 mL 1 mL to 5 mL 5 mL to 20 mL (n = 8)

Atraton ND 1.16E+07 5.42E+07 N/A 4.69 11.15Simetryn ND 4.27E+06 1.94E+07 N/A 4.56 8.93Prometon/Secbumeton 3.26E+06 1.07E+07 4.80E+07 3.30 4.47 9.89Ametryn 4.34E+06 1.42E+07 5.99E+07 3.27 4.22 11.59Simazine 3.18E+05 1.28E+06 5.70E+06 4.03 4.44 5.32Prometryn/Terbutryn 6.19E+06 1.89E+07 7.61E+07 3.05 4.02 3.99Atrazine 1.26E+06 4.45E+06 1.55E+07 3.53 3.49 4.97Table 1: Reproducibility for 20 mL injections (n = 8) at a 1 pg/mL concentration level, without an internal standard.

Figure 3: Chromatograms showing the injection of simazine with 1, 5, and 20 mL injection volumes. The concentration of simazine is 1 pg/mLfor all three injections.

In addition to quantitative data, qualitative data wascollected for each analyte using Quantitation-EnhancedData-Dependent MS/MS (QED-MS/MS) scanning with theReverse Energy Ramp (RER) scan function. The reverseenergy ramp allows the collision energy in Q2 to beramped from a high energy to a lower energy as Q3 isscanning the product ions from Q2 from low mass to highmass. This provides a rich product ion spectrum that canbe used for library searching or ion ratio calculations tohelp eliminate “false positive” results. The RER provides a much “richer” product ion when compared to a Q3product ion scan collected with a static collision energy.For this experiment, the collision energy for the RER wasset to 25 eV and the ramp value was set to 20 eV. Thisresults in a ramp from 45 eV at the low mass range ofQ3. As Q3 scans to higher masses, the collision energy in Q2 is ramped lower and ends at a collision energy of 25 eV. Figure 4 shows the full scan Q3 spectrum that was collected during the analytical run for the calibrationstandard at a level of 1 pg/mL. It also shows a rampillustrating the collision energy ramp applied to Q2.

Administrator

Text Box


Conclusion

Using a preconcentration column in tandem with ananalytical HPLC column allowed for the quantitation of a triazine herbicide mixture over the concentration range0.1 – 10.0 pg/mL. Direct 20 mL injections were performedwith the two HPLC columns. The large injection volumecapabilities of the EQuan system eliminated the need forlaborious and expensive offline preconcentration usingsolid phase extraction. Injection volumes ranging from 1 mL to 20 mL are possible using this configuration, thusoffering flexibility for laboratories based on theirsensitivity and reporting requirements.

Acknowledgements

We would like to thank Scott Harrison from LEAPTechnologies (Carrboro, NC) for assistance with the CTCmacro for multiple syringe fills and Mark Harrison fromThermo Fisher Scientific (Hemel-Hempsted, UK) for furtherassistance with the macro and for technical discussions.







Africa+43 1 333 5034 127Australia+61 2 8844 9500Austria+43 1 333 50340Belgium+32 2 482 30 30Canada+1 800 530 8447China+86 10 8419 3588Denmark+45 70 23 62 60 Europe-Other+43 1 333 5034 127France+33 1 60 92 48 00Germany+49 6103 408 1014India+91 22 6742 9434Italy+39 02 950 591Japan +81 45 453 9100Latin America+1 608 276 5659Middle East+43 1 333 5034 127Netherlands+31 76 579 55 55South Africa+27 11 570 1840Spain +34 914 845 965Sweden / Norway /Finland+46 8 556 468 00Switzerland+41 61 48784 00UK+44 1442 233555USA+1 800 532 4752

www.thermo.com

AN62858_E 10/08M



Legal Notices©2008 Thermo Fisher Scientific Inc. All rights reserved. CTC is a trademark of CTC Analytics AG. All other trademarks are the property of Thermo FisherScientific Inc. and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scientific Inc. products. It is not intended toencourage use of these products in any manners that might infringe the intellectual property rights of others. Specifications, terms and pricing are subject to change.Not all products are available in all countries. Please consult your local sales representative for details.


Figure 4. QED-MS/MS Q3 spectrum for a 1pg/mL injection of atrazine. The collision energy was 25 eV and the ramp was 20 eV.

Administrator

Text Box


LC-MS/MS Determination of Malachite Greenand Leucomalachite Green in Fish ProductsTao Ding1, Jingzhong Xu1, Bin Wu1, Huilan Chen1, Chongyu Shen1, Fei Liu2, and Kefei Wang3

1 Food Laboratory, APFIC, Jiangsu Entry-Exit Inspection and Quarantine Bureau (JSCIQ), Nanjing, China2 Thermo Fisher Scientific, Shanghai, China3 Thermo Fisher Scientific, San Jose, CA, USA


Key Words

• Hypersil GOLD

• Surveyor HPLC


• Food Safety

• H-SRM

Introduction

Malachite Green (MG, see Figure 1), a triphenylmethanedye, is an effective and inexpensive fungicide used inaquaculture, particularly in Asian countries. Duringmetabolism MG reduces to Leucomalachite Green (LMG),which has been shown to accumulate in fatty fish tissues.

Both MG and LMG have demonstrated putativecarcinogenic activity, and thus have been banned for usein aquaculture by both the U.S. FDA and European Union(EU). But trace levels of MG and LMG residues continueto be found in fish products. In a 2005 report,[1] malachitegreen was found in 18 out of 27 live eel or eel productsimported from China to Hong Kong local market andfood outlets, resulting in a government recall of allremaining products to be destroyed.

Based on European Commission decision2002/657/EC, an analytical test method to detect MG andLMG must have a Minimum Required Performance Limit(MRPL) of 2 µg/kg of total malachite greens (MG+LMG)in fish muscle. Detection of MG and LMG has beenreported by using UV-Vis, fluorescence spectrometry andmass spectrometry coupled to HPLC separation. Amongthese detection techniques, the sensitivity and selectivityare poor with UV-Vis, and the fluorescence detectionrequires a post-column oxidation (e.g. with lead oxide) toconvert LMG to MG. Only mass spectrometry allows fordetection of both LMG and MG without post-columnoxidation, and with superior sensitivity and selectivity.[2]

In this work, we report an LC-MS/MS method todetect MG and LMG in roasted eel meat using a triplequadrupole mass spectrometer operated in highly selectivereaction monitoring (H-SRM) mode. The method issensitive and selective, and has been validated for routinedetection of < 0.5 µg/kg of MG+LMG. Moreover, wedemonstrate the capability of using H-SRM to reduce thechemical noise in complex sample matrices to improvedetection of ultra-low level MG and LMG.

Experimental

Chemicals and Reagents

All chemicals were of reagent grade or better. MG oxalatesalt and LMG were from Sigma-Aldrich (St Louis, MO,USA), and d6-LMG from WITEGA (Berlin, Germany).

Sample Preparation

Extraction:1. To 5.00 g of homogenized roasted eel meat, add 50 µL

1 µg/mL of d6-LMG as internal standard (ISTD), 1 mL0.25 g/L hydroxylamine hydrochloride (NH2OH-HCl),1 mL 0.05 mol/L p-toluenesulfonic acid, 2 mL 0.1mol/L NH4Ac-HAc buffer (pH 4.5), and 40 mLacetonitrile.

2. Homogenize for 2 min.3. Centrifuge the mixture at 3000 rpm for 3 min.4. Collect the supernatants into a 250-mL separation

funnel.5. Extract the meat once more with 20 mL acetonitrile.

Liquid-Liquid Extraction:

1. To the acetonitrile crude extract in the separationfunnel, add 30 mL of dichloromethane (DCM) and35 mL DI water, shake for 2 min.

2. Collect the DCM.3. Extract the aqueous phase one more time with

20 mL DCM.4. Evaporate the combined DCM solvent to dryness, and

reconstitute in 3 mL of formic acid/acetonitrile (2:98).

Solid Phase Extraction (SPE):

1. Condition the Oasis 60 mg/3 cc MCX cartridge(Waters, Milford, MA, USA) with 3 mL acetonitrile,and 3 mL 2% v/v formic acid aqueous solution.

2. Load the sample (at ~0.2 mL/min).3. Wash with 2 mL formic acid:acetonitrile (2:98) and

6 mL of acetonitrile.4. Elute with 4 mL NH4Ac (5 mol/L and pH 7)/MeOH

(5:95).5. Evaporate the MeOH at 45°C under reduced pressure6. Dilute to 1.0 mL with initial mobile phase of water

(0.1% v/v formic acid)/MeOH (70:30)7. Filter with a 0.45 µm syringe filter before injection

to LC-MS.

NCH3

CH3

N

CH3

H3CN

CH3

CH3

N

CH3

H3C

Reduction

Oxidation

Malachite Green (MG) Leucolmalachite Green (LMG)

Figure 1: Structure and Conversion of Malachite Green andLeucomalachite Green

Administrator

Text Box


Note: For very fatty roasted eel tissues, prior to the SPEwith the MCX, the extracts after liquid-liquid extractionwere cleaned up with a Superclean 60 mg/3 cc LC-Alumina N cartridge (Waters, Milford, MA, USA):

1. Condition the cartridge with 3 mL acetonitrile2. Load the sample, collect the elute3. Wash with 3 mL acetonitrile, collect the elute to be

combined with elute in (2)4. Add 120 µL of formic acid to the combined elute.

Chromatography Conditions

HPLC: Thermo Scientific Surveyor HPLC Column: Thermo Scientific Hypersil GOLD

CN 50 × 2.1 mm 5 µmMobile Phase: A: Methanol

B: Water with 0.1% v/v Formic AcidGradients: Time (min) A%

0.0 30%2.0-6.0 90%

6.1 30%10.0 30%

Flow Rate: 220 µL/minInjection volume: 10 µL

Mass Spectrometry Conditions

MassSpectrometer: Thermo Scientific TSQ Quantum

Discovery MAX Source: ESI+, 4000 VSheath Gas: 40 unitAuxiliary Gas: 5 unitCapillaryTemperature: 350°CSource CID: -10 VQ1 Peak Width(FWHM): 0.7 Da (0.2 Da for H-SRM)Q3 Peak Width(FWHM): 0.7 DaCollision Gas: Ar (1.5 mTorr)SRM Transitions: See Table 1Scan Time: 0.1 s


Eel meat tissues are in general fatty, and the roasted eelmeat contains additional cooking oil and flavor chemicals,making the sample matrix complicated. The extractionof MG and LMG from roasted eel meat involves sampleextraction, liquid-liquid extraction, and one or two stepsof SPE clean up. A similar method using the liquid-liquidextraction (with DCM) and one step SPE clean up (withSCX) were also reported for analysis of MG and LMGin both the raw and the “processed eel products.”[3] Forextraction of MG and LMG in other raw fish meats, theprocedure could be simplified. For example, Roudaut etal. have recently reported the following method withoutusing the SPE for salmon, trout, tilapia and catfish:[4]

Use 2 g of homogenized sample Add 200 µL standard solution

(for spike experiment only)Add 200 µL ISTD solution 20 ng/mLAdd 600 µL Water (800 µL for unknown)Add 2 mL Hydroxylamine HCl 5 g/LStir the mixture for 10 minAdd 8 mL acetonitrileStir 10 min at 100 rpmCentrifuge for 5 minFilter on 0.45 umInject 20 µL to a TSQ Quantum™

Figure 2 shows the comparison of SRM andH-SRM chromatograms of a matrix matched standard(i.e., standard spiked into a blank roasted eel extractsample after sample preparation) containing 0.02 pg/µL(0.2 pg on-column) MG and 0.1 pg/µL (1 pg on-column)LMG and with 1 pg/µL ISTD. As shown, with H-SRM,signal-to-noise (S/N) ratios have improved significantlyfrom 2-5 to 20-25. Note that the S/N improved despitethe absolute signal (measured by the peak areas)decreasing by approximately half, indicating that the gainsin S/N are from eliminating noise (isobaric interferences)in the sample matrix. The instrument detection limit in thecurrent matrix is thus estimated to be 0.1 pg for MG and0.5 pg for LMG with H-SRM based on 10× S/N. Thesedetection limit values, corresponding to 0.004 µg/kg and0.02 µg/kg for MG and LMG in meat tissues, respectively,have far exceeded our current requirement to detect<0.5 µg/kg of MG+LMG in roasted eel meat.

The response linearity was evaluated over the rangeof 0.05-8.0 µg/kg using matrix matched standard solutions.The correlation coefficients obtained are >0.99 (weightfactor = 1/X). Figure 3 shows the representative calibra-tion curves.

The analytical method was validated by analyzingfortified roast eel samples at 1, 2 and 5 µg/kg levels forboth LG and LMG, corresponding to 0.5×, 1×, and 2.5×MRPL, respectively. Seven replicates were performed ateach level. The results are summarized in Table 2.Excellent recovery values of 90-106% were obtainedwith RSD% ranging from 3.7 to 11%.

Product IonPrecursor Ion (Collision Energy)

MG (M*) 329.1 313 (33)*208 (48)

LMG (MH+) 331.3 239 (31)*316 (18)

d6-LMG (MH+) 337.2 240 (30)

Table 1: SMR Transitions and Collision Energy Values for MG and LMG

Administrator

Text Box


MG

(329.1>313)

LMG

(331.3>239)

D6-MG

(337.2>240)

Figure 2: Comparison of SRM (left) and H-SRM (right) Chromatograms of a Matrix Matched Standard Containing 0.02 pg/µL MGand 0.05 pg/µL LMG (10 µL injection).

Figure 3: Representative calibration curves for MG and LMG with matrix matched standard solutions.

Spike (µg/kg) 1.0 2.0 5.0MG 95 (5.8) 101 (3.7) 90 (7.5)LMG 106 (7.0%) 94 (11) 92 (4.7)

Table 2: Recovery% (RSD%) of MG and LMG (ISTD Corrected)in Roasted Eel Meat (n=7)

Administrator

Text Box



Conclusions

A highly sensitive and selective LC-MS/MS method usingthe TSQ Quantum Discovery MAX has been developedfor determining Malachite Green and LeucomalachiteGreen in roasted eel meat. The method shows excellentlinearity (0.05 to 8.0 µg/kg), accuracy (90-106% recovery)and reproducibility (4-11% RSD), far exceeding theEU’s requirement of MRPL of 2 µg/kg of MG+LMG.The method has been implemented at the JSCIQ lab forroutine monitoring of <0.5 µg/kg (MG+LMG) in roastedeel and other fish products (with variations of samplepreparation procedures).

Highly selective reaction monitoring (H-SRM) hasbeen shown to reduce the chemical noise effectively inthe complicated sample matrix, which should be usefulto further improve the method sensitivity and specificity(i.e., to eliminate both false positive and false negative)in support of enforcement of a “zero tolerance” policytoward the use of MG and LMG for aquaculture.

References

1. http://news.gov.hk/en/category/healthandcommunity/050819/html/050819en05002.htm

2. DR Doerge, MI Churchwell, TA Gehring, YM Pu and SM Plakas,“Analysis of Malachite Green and Metabolites in Fish Using LiquidChromatography Atmospheric Pressure Chemical Ionization MassSpectrometry”, Rapid Communications in Mass Spectrometry, 12,1625-1634 (1994).

3. Hubert Tang PO, Twinnie Tso. S.C. and Clare Ho, “Determination ofMalachite Green and Leucomalachite Green in Fish Products”, 5thInternational Symposium on Hormone and Veterinary Drug ResidueAnalysis, Antwerp, Belgium, May 16-19, 2006.

4. B Roudaut, B Delepine, M Bessiral and P Sanders, “Malachite Green andLeucomalachite Green in Fish Flesh by Liquid Chromatography TandemMass Spectrometry (Validation of A Confirmatory Method)”, 2nd AOACInternational Symposium on Recent Advances in Food Analysis, Prague,Czech Republic, November 2-4, 2005.


©2007 Thermo Fisher ScientificInc. All rights reserved. Alltrademarks are the property ofThermo Fisher Scientific Inc.and its subsidiaries.

Specifications, terms and pric-ing are subject to change. Notall products are available in allcountries. Please consult yourlocal sales representative fordetails.








www.thermo.com

AN62296_E 03/07S


Administrator

Text Box


Multi-residue Analysis of Pesticides in Foodusing GC/MS/MS with the TSQ Quantum GCKuniyo Sugitate, Michiko Kanai, Thermo Fisher Scientific, Yokohama, JapanMasahiro Okihashi, Division of Food Chemistry, Osaka Prefectural Institute of Public Health, Osaka, Japan Dipankar Ghosh, Thermo Fisher Scientific, San Jose, CA, USA


Key Words

• TSQ Quantum GC

• H-SRM

• PesticideResidues in Food

• Positive ListSystem

• QED

• SRM

Introduction

Food safety concerns are on the rise amongst consumersworldwide. In 2006, sweeping changes were made to theFood Hygiene Law in Japan regarding residues of agricul-tural chemicals, including pesticides, in foods. As a result,standard residue values were established for approximately800 pesticides. All food items produced in or importedinto Japan are required to meet the standards establishedby this law. If pesticide residues in any food items exceedthese standards, then the distribution and sale of the foodis prohibited. This Positive List System has had a significanteffect not only on the Japanese domestic production, butalso on much of the food exported to Japan from variousforeign countries such as China, the United States, and Taiwan.

There are numerous types of pesticides regularly used inthe agricultural industry, including insecticides, fungicides,herbicides, and growth regulators. Because each type hasdifferent physicochemical properties, there are limitationson simultaneous analysis. Among the pesticides for whichstandard values are currently set, GC/MS/MS can analyzeapproximately 300 compounds. The superior selectivity ofthis technique allows interference-free quantification, evenwith peak coelution, and provides positive confirmation ofvarious pesticides in a single analytical run.

To accurately monitor pesticide residues, a highthroughput multi-residue screening method that can quantitate a large number of pesticide residues during asingle analytical run is needed.

GoalTo simultaneously analyze 103 pesticides using the Thermo Scientific TSQ Quantum GC system, using SRMand H-SRM. Additionally, to show the utility of QEDMS/MS for structural confirmation of the analytes under-going quantification.


Sample Preparation

Green pepper, carrot, grapefruit and banana samples wereprepared for analysis using a method based on the simpleand quick QuEChERS approach.1 A 10 g sample of foodwas homogenized in a food processor and placed in apolypropylene centrifuge tube. The sample was extractedwith 20 mL of acetonitrile in a homogenizer. Then, 4 g ofanhydrous magnesium sulfate and 1 g of sodium chloridewere added and the resulting mixture was centrifuged.After centrifugation, the supernatant was loaded onto a

graphite carbon/PSA dual layer solid phase extraction column and eluted with 50 mL of acetonitrile/toluene (3:1).After the eluate was concentrated under reduced pressure, it was dissolved (1 g/mL) in 10 mL of acetone/n-hexane togive the test solution.

GC

GC analysis was performed using the Thermo ScientificTRACE GC Ultra System. The GC conditions were as fol-lows:

Column: Rxi-5MS 30 m x 0.25 mm I.D.,0.25 m df (Restek Corp., Bellefonte, PA)

Injection mode: Splitless with surge injection(200 kPa, 1 min)

Injection temperature: 240 °C Oven temperature: 80 °C (1 min) – 20 °C/min – 180 °C –

5 °C/min – 280 °C (10 min)Flow rate: Constant flow 1.2 mL/minTransfer line temperature: 280 °C

AS

The samples were injected through the Thermo ScientificTriPlus autosampler. The autosampler conditions were asfollows:

Injection volume: 1 µLInjection mode: Hot needle Syringe: 80 mm

MS

MS analysis was carried out on a TSQ Quantum™ GCtriple stage quadrupole mass spectrometer. The MS condi-tions were as follows:

Ionization mode: EI positive ion Ion volume: Closed EI Emission current: 25 µAIon source temperature: 220 °CScan type: SRM and H-SRMScan width: 0.002 a.m.u. Scan time: 0.01 s Peak width: Q1, 0.7 Da; Q3, 0.7 Da FWHMPeak width for H-SRM: Q1, 0.4 Da; Q3, 0.7 Da FWHMCollision gas (Ar) pressure: 1.2 mTorr

A total of 103 pesticides were analyzed to determinethe product ion to be used for quantitation. Table 1 liststhe SRM transitions and the optimum collision energy foreach of the compounds and a summary of the calibrationrange, linearity, and the reproducibility of each individualcompound at 5 ppb (ng/mL).

Page 1of 6

Administrator

Text Box


Precursor Product Collision CV(%)R.T. Ion (m/z) Ion (m/z) Energy R2 Range n=5

Mevinphos 6.44 192 127 10 0.9999 0.1-100 4.03

XMC 7.52 122 107 10 0.9999 0.1-100 2.55

Tecnazene 8.03 261 203 15 0.9996 0.1-100 5.41

Ethoprpphos 8.22 200 114 10 0.9981 0.1-100 7.91

Ethalfluralin 8.42 316 276 10 0.9997 0.1-100 4.14

Benfluralin 8.62 292 264 10 0.9989 0.1-100 1.86

Monocrotophos 8.62 192 127 10 0.9754 5-100 19.47

α-BHC 9.03 219 183 15 0.9999 0.1-100 4.51

Dicloran 9.25 206 176 10 0.9994 0.1-100 2.30

Simazine 9.30 201 172 10 0.9999 0.1-100 4.33

Propazine 9.50 214 172 10 0.9998 0.1-100 1.99

β-BHC 9.57 219 183 15 1.0000 0.1-100 3.51

γ -BHC 9.73 219 183 15 0.9998 0.1-100 6.57

Cyanophos 9.78 243 109 10 0.9996 0.1-100 3.56

Pyroquilon 9.90 173 130 20 0.9996 0.1-100 2.95

Diazinon 4.94 304 179 15 0.9995 0.1-100 4.40

Phosphamidon-1 10.06 264 127 10 0.9989 0.1-100 10.31

Prohydrojasmon-1 10.12 184 83 20 0.9992 0.1-100 7.39

δ-BHC 10.26 219 183 15 0.9994 0.1-100 5.17

Prohydrojasmon-2 10.66 264 127 10 0.9972 0.1-100 17.11

Benoxacor 10.7 259 120 15 0.9999 0.1-100 3.30

Propanil 10.95 262 202 10 0.9993 0.1-100 3.65

Phosphamidon-2 10.97 264 127 10 0.9970 0.1-100 8.77

Dichlofenthion 10.99 279 223 15 0.9994 0.1-100 2.21

Dimethenamid 11.06 230 154 10 0.9996 0.1-100 2.51

Bromobutide 11.09 232 176 10 0.9990 0.1-100 5.91

Paration-methyl 11.24 263 109 10 0.9982 0.1-100 3.74

Tolclofos-methyl 11.38 265 250 15 0.9998 0.1-100 2.52

Ametryn 11.43 227 170 10 0.9999 0.1-100 0.90

Mefenoxam 11.57 249 190 10 0.9995 0.1-100 5.81

Bromacil 11.98 205 188 15 0.9988 0.1-100 3.87

Pirimiphos-methyl 12.00 305 276 10 0.9995 0.1-100 4.08

Quinoclamine 12.18 207 172 10 0.9989 0.1-100 4.24

Diethofencarb 12.34 225 125 15 0.9985 0.1-100 4.64

Cyanazine 12.52 225 189 10 0.9994 0.1-100 3.41

Chlorpyrifos 12.57 314 258 15 0.9991 0.1-100 3.37

Parathion 12.59 291 109 15 0.9962 0.1-100 9.76

Triadimefon 12.67 208 111 25 0.9986 0.1-100 6.10

Chlorthal-dimethyl 12.73 301 223 20 1.0000 0.1-100 1.23

Nitrothal-isopropyl 12.78 236 148 15 0.9974 0.1-100 5.53

Fthalide 13.04 272 243 10 0.9993 0.1-100 4.32

Fosthiazate 13.05 195 103 10 0.9956 5-100 6.29 13.12

Diphenamid 13.1 239 167 10 0.9997 0.1-100 4.67

Pyrifenox-Z 13.64 262 200 15 0.9979 0.2-100 4.54

Fipronil 13.79 367 213 25 0.9991 0.1-100 3.49

Allethrin 13.67 123 81 10 0.9991 5-100 3.79

Dimepiperate 13.87 145 112 10 0.9987 0.1-100 3.74

Phenthoate 13.87 274 121 10 0.9987 0.1-100 1.82

Quinalphos 13.88 146 118 10 0.9984 0.1-100 1.96

Paclobutrazol 14.45 236 125 15 0.9961 0.1-100 7.41

Endosulfan-α 14.67 241 206 15 0.9996 0.1-100 4.54

Butachlor 14.73 237 160 10 0.9998 0.1-100 5.26

Imazamethabenz- 14.81 256 144 20 0.9932 2-100 12.09methyl

Butamifos 15.00 286 202 15 0.9958 0.1-100 4.66

Napropamide 15.01 271 128 5 0.9989 0.1-100 8.96

Precursor Product Collision CV(%)R.T. Ion (m/z) Ion (m/z) Energy R2 Range n=5

Flutlanil 15.06 173 145 15 0.9986 0.1-100 1.93

Hexaconazole 15.06 214 172 15 0.9924 0.1-100 8.98

Profenofos 15.28 337 267 15 0.9968 0.1-100 6.61

Uniconazole-P 15.38 234 137 15 0.9966 0.1-100 11.37

Pretilachlor 15.37 162 132 15 0.9982 0.1-100 6.72

Flamprop-methyl 15.66 276 105 10 0.9986 0.1-100 3.93

Oxyfluorfen 15.69 361 300 10 0.9980 0.5-100 6.07

Azaconazole 15.79 217 173 15 0.9981 0.1-100 7.07

Bupirimate 15.82 316 208 10 0.9982 0.1-100 4.65

Thifluzamide 15.84 449 429 10 0.9972 0.1-100 2.75

Fenoxanil 16.25 293 155 20 0.9989 0.1-100 3.73

Chlorbenzilate 16.43 251 139 15 0.9976 0.1-100 0.81

Pyriminobac- 16.76 302 256 15 0.9986 0.1-100 2.70methyl-Z

Oxadixyl 16.86 163 132 10 0.9998 0.1-100 3.72

Triazophos 17.30 257 162 10 0.9941 0.2-100 6.72

Fluacrypyrim 17.38 189 129 10 0.9988 0.1-100 2.15

Edifenphos 17.72 310 173 10 0.9927 0.1-100 7.95

Quinoxyfen 17.74 272 237 10 0.9993 0.1-100 4.50

Lenacil 17.78 153 136 15 0.9979 0.1-100 5.19

Trifloxystrobin 18.01 222 162 10 0.9966 0.1-100 8.47

Pyriminobac- 18.19 302 256 15 0.9982 0.1-100 2.12methyl-E

Tebuconazole 18.39 250 125 20 0.9907 0.2-100 13.03

Diclofop-methyl 18.51 253 162 15 0.9991 0.1-100 2.14

Mefenpyr-diethyl 19.15 253 189 20 0.9992 0.1-100 3.35

Pyributicarb 19.24 165 108 10 0.9973 0.1-100 2.00

Pyridafenthion 19.46 340 199 10 0.9940 0.2-100 4.71

Acetamiprid 19.39 152 116 20 1.0000 50-100 –

Bromopropylate 19.64 341 185 15 0.9956 0.1-100 3.72

Piperophos 19.84 320 122 10 0.9939 0.2-100 7.51

Fenpropathrin 19.98 265 210 10 0.9973 0.1-100 6.87

Etoxazole 20.06 300 270 20 0.9969 0.1-100 8.84

Tebufenpyrad 20.10 333 171 20 0.9978 0.5-100 13.35

Anilofos 20.31 226 157 15 0.9948 0.2-100 5.56

Phenothrin-1 20.49 183 165 10 0.9967 5-100 16.13

Tetradifon 20.54 356 229 10 0.9998 0.2-100 4.17

Phenothrin-2 20.66 183 165 10 0.9968 0.1-100 3.79

Mefenacet 21.22 192 136 15 0.9955 0.1-100 4.90

Cyhalofop-buthyl 21.23 357 229 10 0.9967 0.1-100 5.52

Cyhalothrin-1 21.30 181 152 20 0.9975 0.2-100 3.21

Cyhalothrin-2 21.66 181 152 20 0.9984 0.2-100 6.67

Pyrazophos 22.06 373 232 10 0.9963 0.1-100 10.46

Bitertanol 22.80 170 141 20 0.9873 0.1-100 6.76 22.97

Pyridaben 23.18 147 117 20 0.9958 0.1-100 1.29

Cafenstrole 24.03 100 72 5 0.9958 0.1-100 9.77

Cypermethrin-1 24.72 181 152 20 0.9983 2-100 9.29

Halfenprox 24.79 263 235 15 0.9979 0.1-100 10.25

Cypermethrin-2 24.92 181 152 20 0.9982 2-100 6.91

Cypermethrin-3 25.06 181 152 20 0.9985 2-100 16.27

Cypermethrin-4 25.13 181 152 20 0.9948 2-100 13.79

Fenvalerate-1 26.47 167 125 10 0.9977 0.1-100 3.11

Flumioxazin 26.50 354 176 20 0.9937 0.1-100 9.66

Fenvarelate-2 26.91 167 125 10 0.9979 0.1-100 3.26

Deltamethrin+ 28.15 181 152 20 0.9967 0.2-100 8.20 Tralomethrin

Tolfenpyrad 29.11 383 171 20 0.9968 2-100 4.84

Imibenconazole 30.35 375 260 15 1.0000 50-100 –

Table 1: Retention times, SRM conditions, calibration range, linearity, and the reproducibility of each individual pesticide residue compound

Page 2 of 6

Administrator

Text Box



Figure 1 shows an example calibration curve for Propazine at 0.1-100 ppb with a corresponding chromatogram at 1 ppb, showingexcellent reproducibility (r2 = 0.9998).

Figure 2 shows examples of GC/MS/MS chromatograms of variouspesticides in which 1 ppb of each pesticide was added to green pepper.Even at this extremely low concentration (1/10 of the uniform standardvalue for pesticides), it was possible to make measurements withremarkably high sensitivity with the TSQ Quantum GC.

Figure 3 shows the chromatograms for cypermethrin, fenvalerateand deltamethrin (+ tralomethrin). Cypermethrin is a syntheticpyrethroid compound with a high detection ratio in agricultural produce. In addition to having a slow elution time in the GC, it has 4 peaks that are due to different isomers that must be resolved.As the chromatograms show, measurements with good sensitivitywere obtained even at the low concentration of 5 ppb.

Figure 1: Calibration curve (0.1-100 ppb) and SRM chromatogram (1 ppb) for Propazine

Figure 2: GC/MS/MS chromatograms of various pesticides at 1 ppb in green pepper samplesPage 3 of 6

Administrator

Text Box


Figure 3: Chromatograms for cypermethrin, fenvalerate and deltamethrin

Figure 4: Comparison of SRM Mode with H-SRM Mode. (a) Flamprop-methyl in grapefruit (1 ppb). (b) Parathion in green-pepper (1 ppb).Page 4 of 6

Administrator

Text Box


Figure 5: Chromatogram from a carrot test sample (upper row) and the MS/MS spectrum obtained with QED (lower row)

Advantages of H-SRMH-SRM is an acronym for Highly-Selective ReactionMonitoring, which is a more advanced form of SelectiveReaction Monitoring (SRM). H-SRM can eliminatechemical noise, lower detection limits, and reduce thelikelihood of generating false positives. For many pesti-cides that are subject to matrix-dependent interference,the measurements can be successfully carried out usingthe H-SRM mode. With H-SRM, the precursor ion isselected with a smaller peak width. The more stringenttolerance accounts for the higher selectivity, which canlower LOQs and increase precision and accuracy at thelimits of detection. The effects of H-SRM over SRM areillustrated for flamprop-methyl in grapefruit andparathion in green-pepper in Figure 4.

Structural Confirmation with QEDQED MS/MS stands for Quantification Enhanced by DataDependent MS/MS. A QED scan on a triple quadrupoleinstrument delivers an information rich mass spectrumthat can be used for structural confirmation of analyteswhile undergoing quantification by SRM (or H-SRM). Thespecificity provided by H-SRM followed by QED MS/MSprovides uncompromised quantitation performance atlow levels followed by a fast, highly-specific full MS/MSscan for confirmation. Figure 5 shows the QED scanresults obtained from a carrot test sample spiked with 10 ppb diazinon.

Page 5 of 6

Administrator

Text Box


Zero Cross-talkCross-talk can potentially occur when fragment ions fromone SRM transition remain in the collision cell while asecond SRM transition takes place. This can cause signalartifacts in the second SRM transition’s chromatogram. It can be especially problematic when different SRM eventshave the same product ions formed from different precursorions. However, the orthogonal design of the collision cellin the TSQ Quantum eliminates cross-talk. Figure 6 showsthe absence of cross-talk between two different SRM transitions of paclobutrazol and thifluzamide. Both yield aproduct ion of m/z 125, but no artifacts are seen in eitherchromatogram with a scan time of 10 ms. Similarly, theSRM transitions of triszophos and diclofop-methyl 5 alsoshow no evidence of cross talk, even though they bothyield product ions at m/z 162.

ConclusionSimultaneous analysis was carried out on multi-componentpesticide residues in food products using a quadrupoleGC/MS/MS system, the TSQ Quantum GC. Results obtainedindicated excellent sensitivity (0.1 ppb), reproducibility(10% at 5 ppb) and linearity (R2 > 0.995) in the range of0.1-100 ppb. No cross-talk was observed for the analysisof closely eluting multi-component mixtures. Using H-SRM,interferences from the sample matrix background weresubstantially reduced, leading to improved LOQs. In addition, QED provided MS/MS structural confirmationof the analytes undergoing quantification.

References1. Okihashi, M.; Kitagawa, Y.; Akutsu, K.; Obana, H.; Tanaka, Y. “Rapid

method for the determination of 180 pesticide residues in foods by gaschromatography/mass spectrometry and flame photometric detection”; J Pestic Sci 2005, 30(4), 368-77.

2. The Japanese Ministry of Health, Labour, and Welfare:www.mhlw.go.jp/english/topics/foodsafety/positivelist060228/index.html (English)www.mhlw.go.jp/topics/bukyoku/iyaku/syoku-anzen/zanryu2/index.html(Japanese)








www.thermo.com

AN387_E 02/08M


Figure 6: No cross-talk was observed in the SRM transitions of paclobutrazol and thifluzamide or in the SRM transitions of triszophos and diclofop-methyl.



Administrator

Text Box



Key Words

• TSQ Quantum

• EQuan

• LC-MS/MS

• Pesticides

• QED

• RER

• Water Analysis

Overview

In recent years, the use of LC-MS/MS for pesticide analysishas become increasingly popular due to the amenabilityof electrospray ionization (ESI) for polar compounds.This application note describes the multi-residue assayof a group of pesticides in drinking water by using theThermo Scientific EQuan system. The method utilizesSRM for quantification, followed by Data DependentQED-MS/MS and library searching for the structural con-firmation of these analytes.

Traditionally, water samples are extracted in a verytime consuming manner, using 1 L of sample that is con-centrated 100-1000 times before analysis. This extractioncan take several hours or even days of preparation time.Also, the added expense of extraction cartridges leads tomore expense per sample for labs. EQuan is a solution thatallows for automated online preconcentration of sampleswithout the need for extensive sample preparation ormanual intervention.

Introduction

There is an increasing emphasis for using multiple SRMs(selected reaction monitoring) and the use of Ion RatioConfirmation (IRC) to positively confirm the presence ofbanned or controlled substances in samples. For example,the 2002/657/EC European Commission Decision dictatesIdentification Points (IPs) that must be met for a sampleto be deemed “positive.” These criteria can include thenumber of MS/MS transitions, ion ratios, or the typeof mass spectrometer used (high resolution Vs. lowresolution).

The Quantitation Enhanced Data-Dependent (QED)scan on a regular triple quadrupole instrument delivers aninformation rich MS/MS which can be used to confirm theexistence of compounds while they are being quantified.When using QED, a “full scan MS/MS” mass spectrumis obtained by Data Dependent scanning for confirmatoryanalysis during the single reaction monitoring experiment(SRM), which is used for routine quantitation. Once aparticular SRM transition reaches a “user set” intensitythreshold, the instrument automatically triggers QED,using an innovative new technique called Reversed EnergyRamp (RER) which produces the high sensitivity production spectrum. The RER function linearly ramps the colli-sion energy from a high to low value, while scanning Q3.The RER scan generates a highly sensitive, fragment-richMS/MS spectrum that can be used to positively confirmthe existence of a compound.

Quantitation-Enhanced Data-Dependent (QED)Scanning of DrinkingWater Samples Using EQuanfor Pesticide Analysis on a Triple Stage QuadrupoleJonathan Beck, Thermo Fisher Scientific, San Jose, California, USAMihoko Yamaguchi and Kaori Saito, Thermo Electron K.K., Yokohama, Japan

Many pesticide samples are regulated at a very lowlevel (ppt, or ng/L levels), and in order to detect com-pounds at these low levels, time consuming extractionand concentration of samples is required before analysis.EQuan utilizes two LC pumps, a large volume auto -sampler, two HPLC columns, and a Thermo ScientificTSQ Quantum Mass Spectrometer to reduce samplepreparation time and to analyze the samples at the con-centration levels that are required.


Samples–Drinking water was spiked with tricyclazole(0.8), carbaryl (0.5), carbofuran (0.05), asulam (2.0),diruon (0.2), siduron (3.0), daimuron (8.0), carpropamid(0.4), thiodicarb (0.8), azoxystrobin (5.0), flazasulfuran(0.3), bensulfuron methyl (4.0), and halosulfon methyl(3.0) at concentration levels from 0.5 ppt (pg/mL) to1000 ppt. These compounds are all regulated by theJapanese Ministry of Health Labour and Welfare. Thereporting level for each compound, as set by the JapaneseMinistry of Health, Labour, and Welfare, in ppb (µg/L)is given in parenthesis after each compound.

HPLC Conditions–Two pump systems were used, aThermo Scientific Surveyor L-Pump for loading the 1 mLsample onto the loading column (Thermo ScientificHypersil GOLD 20× 2.1 mm 12µ), and a Surveyor™ MSPump Plus for eluting the compounds off of the loadingcolumn and separation on the analytical column (HypersilGOLD™ 50× 2.1 mm 3µ). The mobile phase for bothpumps was Water with 0.1% Formic Acid (A), andAcetonitrile with 0.1% Formic Acid (B). For the loading L Pump, the gradient used is shown in Table 1, and thegradient used for the analysis pump is shown in Table 2.A divert valve on the mass spectrometer is programmedby the data system to control the loading and elution ofthe two LC columns. In this experiment, the valve is in theload position from 0 to 1.5 minutes to allow for the entire1 mL sample to collect on the loading column beforeswitching to the analysis position until all of the analytesare eluted, 12.5 minutes in this case. After switching backto the loading position, the loading column can be rinsedand re-equilibrated by the loading L pump. A schematic ofthe EQuan system is shown in Figure 1.

Administrator

Text Box


of 8

Flow Rate Time %A %B (mL/min)

0 95 5 1.01.5 95 5 1.02 95 5 0.0

12.5 95 5 0.012.6 5 95 1.014.5 5 95 1.014.6 95 5 1.017 95 5 1.0

MS ConditionsThermo Scientific TSQ Quantum DiscoveryIon source and polarity: ESI, Positive ion modeSpray Voltage: 4500 VSheath Gas: 45 units (N2)Auxiliary Gas: Not UsedTransfer Tube Temperature: 330°CCollision Gap Pressure: 1.0 units (Ar)

MS Scan Functions

Two different scan functions, a SRM (selected reactionmonitoring) followed by a data dependent QED scanfunction were selected in the method. The SRM transi-tions can be seen in Figure 2a, and the QED scan functioncan be seen in Figure 2b.

Figure 1: Diagram showing the EQuan setup used in this experiment.

Table 1: Gradient program for the loading pump. The flow is turned off from2 to 12.5 minutes to conserve mobile phase, and the column is rinsed from12.6 to 14.5 minutes with a high organic phase, before re-equilibrating tostarting conditions.

Time %A %B0.00 95 51.50 95 510.0 0 10012.0 0 10012.1 95 517.0 95 5

Table 2: Gradient program for the analysis pump.The flow rate for this analysis is 0.2 mL/min.

Administrator

Text Box


of 8

Figure 2a: SRM transitions monitored in the experiment.

Figure 2b: QED scan function. A RER from 55 to 10 eV triggered by a SRM transition greater than 5.0×104 counts.Dynamic exclusion was used to allow only one QED spectrum to be collected for each SRM.

Administrator

Text Box


of 8


Figures 3 and 4 show the chromatogram of the 10 pptstandard from 5 to 11.5 minutes for all of the analyteslisted in the Experimental Conditions section. This level isfive times lower than the lowest MRL for the mixture ofcompounds (Carbofuran, 50 ppt). Asulam, the peak show -

ing the lowest S/N in at this concentration level is 200times lower than the MRL. Excellent linearity is obtainedfor all of the compounds over the concentration range1 ppt to 500 or 1000 ppt at the high end. Figure 5 showsthe calibration curve for Asulam. Table 3 summarizes thecalibration data for each compound, individually.

5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5

10.86

10.14

9.88

9.50

5.39

8.55

8.84

7.77

9.99

0

50

100

0

50

100

0

50

100

0

50

100

0

50

100

0

50

100

0

50

100

0

50

100

Asulam

Tricyclazole

Carbaryl

Carbofuran

Diuron

Siduron

Carpropamide

Daimuron

Time (min.)

Rela

tive

Abun

danc

e

Figure 3: A chromatogram at a concentration level of 10 ppt for the first eight compounds (in order of increasingprecursor ion mass) analyzed in the mixture, from 5 to 11.5 minutes.

5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0

10.50

9.63

9.69

9.74

10.86

0

50

100

0

50

100

0

50

100

0

50

100

0

50

100

Thidicarb

Azoxystrobin

Flazasulfuron

Bensulfuron-methyl

Halosulfuron-methyl

Time (min.)

Rela

tive

Abun

danc

e

Figure 4: A chromatogram at a concentration level of 10 ppt for the last five compounds (in order of increasingprecursor ion mass) analyzed in the mixture, from 5 to 11.5 minutes.

Administrator

Text Box


of 8

While excellent linearity and quantitative results weregathered using SRM transitions, additional QED datawere collected during each run for each compound. Anexample of a QED full scan MS/MS spectrum is shownin Figure 6 for the compound Carbofuran. This QEDscan function fragmented the precursor ion m/z 222 forCarbofuran over a reverse energy ramp of 10 to 55 eV,as specified in the second scan function (Figure 2b).

Using the built in environmental compound libraryavailable for the TSQ Quantum that includes over 1000compounds, Carbofuran was the #1 hit in the list of

possible compounds (Figure 7). This feature allows foradditional positive confirmation data for compounds thatare analyzed, providing the required number of IPs neededfor the positive presence of a compound. Library search -ing for all of the compounds in the experimental mixtureyielded matches for either the first or second compound inthe list of possible compounds. Furthermore, the librarysearches of the QED scans for the two compounds Diuronand Siduron, which both have precursor masses of m/z 233,correctly identified each unique compound, based ondifferences in their QED spectra.

0 20 0 400 60 0 800 1000ppt

0

1000000

5500000

5000000

4500000

4000000

3500000

3000000

2500000

2000000

1500000

500000

Area

Figure 5: Calibration curve for the analyte Asulam from 1 to 1000 ppt.

Retention HighCompound Time (min) Concentration EquationTricyclazole 7.7 500 ppt y = 21149 + 112678x R2 = 0.9993

Carbaryl 8.7 500 ppt y = 36518 + 78565x R2 = 0.9964Carbofuran 8.5 500 ppt y = 54509 + 290697x R2 = 0.9977

Asulam 5.3 1000 ppt y = -726 + 5356x R2 = 0.9992Diuron 9.4 500 ppt y = 758 + 32087x R2 = 0.9988Siduron 9.8 500 ppt y = 51461 + 88505x R2 = 0.9994

Daimuron 10.0 500 ppt y = 144173 + 285515x R2 = 0.9963Carpropamide 10.8 500 ppt y = 10377 + 37079x R2 = 0.9999

Thidicarb 8.8 500 ppt y = 16505 + 35334x R2 = 0.9959Azoxystrobin 9.6 1000 ppt y = 45456 + 198901x R2 = 0.9978Flazasulfuron 9.6 1000 ppt y = -3499 + 86802x R2 = 0.9989

Bensulfuron-methyl 9.5 1000 ppt y = -2657 + 65708x R2 = 0.9945Halosulfuron-methyl 10.4 500 ppt y = 1944 + 40565x R2 = 0.9981

Table 3: Results of pesticide calibration curves. All curves were a linear curve fit with a weighting factor of 1/x.

Administrator

Text Box


of 8

40 60 80 100 120 140 160 180 200 220m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rela

tive

Abun

danc

e

30.66 43.1656.74

67.1677.24 87.57

94.88105.45

116.58 133.62

136.96

122.81

164.96

141.18154.43 173.10

186.93193.34

205.27

218.03

221.97

80.46

57.58

Figure 6: QED spectrum of Carbofuran at the 5ppt calibration level. Searching against the standard library available on the TSQ Quantum instrument platformyields a positive confirmation.

Figure 7: Library search result for the QED spectrum generated at the 5 ppt calibration level. Carbofuran, highlighted in blue is the #1 hit in the listof possible compounds.

Sample Spectrum

Library Spectrum

Administrator

Text Box


of 8

Conclusions

The Quantitation Enhanced Data Dependent scan function,standard on all TSQ Quantum mass spectrometers, allowsthe user to obtain confirmatory data following quantitativeanalysis. This is of particular significance when analyzingenvironmental pollutants in water samples. EQuan, withits large injection volume, allows for significant time savingsover traditional SPE concentration methods, and allowsfor detection and quantitation of compounds at levels wellbelow the regulatory requirements.

The built-in library of over 1000 compounds in theindustry standard NIST format can help users to positivelyidentify compounds based on EU regulations.Additionally, users have the ability to add to or replacethe spectra in the library to increase their positive hitprobabilities when searching the library.

References

The 2002/657/EC European Commission Decision can be found on theWorld Wide Web at:

http://ec.europa.eu/food/food/chemicalsafety/residues/lab_analysis_en.htm

The Japanese Ministry of Health, Labour, and Welfare can be found on theWorld Wide Web at:

http://www.mhlw.go.jp/index.html (Japanese)

http://www.mhlw.go.jp/english/index.html (English)

Administrator

Text Box




©2007 Thermo FisherScientific Inc. All rightsreserved. All trademarksare the property of ThermoFisher Scientific Inc. and itssubsidiaries.

Specifications, terms andpricing are subject to change.Not all products are availablein all countries. Please consultyour local sales representativefor details.

AN62265_E 01/07S






















www.thermo.com

Administrator

Text Box


Simultaneous Detection of 88 Pesticideson the TSQ Quantum DiscoveryUsing a Novel LC-MS/MS Method Dipankar Ghosh1, Lutz Alder2, Mark Churchill1, Wilhelm Gebhardt1, Eric Genin1, and Jeanette Klein2

1Thermo Fisher Scientific, San Jose, CA, USA;2Federal Institute for Health Protection of Consumers & Veterinary Medicine, POB 330013, D-14191 Berlin, Germany


Key Words


• Cross-talk

• EPA

• LC-MS/MS

• Pesticides

Overview

Pesticide residues in food are strictly regulated accordingto the provisions of US Environmental Protection Agency(EPA) CFR Title 40. Several hundred sections in Part 180detail the maximum pesticide residue (tolerance) for awide variety of foods. A pesticide’s allowable tolerance(measured in ppm) can span several orders of magnitude,depending upon the food source. For example, the toler-ance for captan in cattle fat is 0.05 ppm, while 100 ppmof captan is acceptable in lettuce and spinach.

To analyze the large numbers of samples whose pesticidetreatment history is usually unknown, the US Food andDrug Administration (FDA) uses analytical methodscapable of simultaneously determining a number of pesti-cide residues. These cost-effective multi-residue methods(MRMs) can determine about half of the approximately400 pesticides and their metabolites with EPA tolerances.Most commonly, residues in extracts are separated by GCor HPLC, and then detected using UV absorbance, nitro -gen phosphorus detection, or electron capture detection.

Due to its specificity in identifying compounds, LC-MS/MS is emerging as the technique of choice foridentifying and quantifying pesticides. The most commonlyused MRMs can also detect many metabolites, impurities,and alteration products of pesticides.

Conventional MS/MS methods generally require exten -sive optimization of operating parameters for each targetanalyte or even for compounds belonging to the samechemical class, significantly impacting analytical through -put. The objective of this work was to demonstrate theuse of the Thermo Scientific TSQ Quantum Discoveryin developing an automated, generic, high-throughputLC-MS/MS screening method to simultaneously detectand quantitate nearly 100 pesticides following minimalseparation using an HPLC.

Goals

• Develop a multi-residue LC-MS/MS screening methodto detect 88 analytes using a single, automated experiment with a short chromatographic time scale

• Demonstrate the utility of using different time segmentsand scan events

• Illustrate the large linear dynamic range for pesticideanalysis in a multi-residue context

• Exhibit the absence of “cross-talk” between co-elutingcomponents


Chemicals and Reagents Water, methanol, and acetic acid were HPLC grade andpurchased from J. T. Baker Chemicals, France.

SamplesPesticides listed in Table 1 were purchased from Sigmaunless otherwise noted. Standards solutions of 0.1, 0.5, 1, 5, 10 and 50 pg/µL were prepared in methanol.

Sample Analysis HPLC analysis was performed on the Thermo ScientificSurveyor HPLC System, using a Thermo ScientificAQUASIL C18 50× 2.1 mm column. Mobile phase A waswater/methanol 80/20 (v/v) and mobile phase B wasmethanol/water 90/10 (v/v) – both contained 0.05% aceticacid. Solvent was pumped at 200 µL/min and analyteseluted using a linear gradient of 100% A to 100% B over11 minutes, holding at 100% B for 12 minutes, thenreturning to 100% A in 2 minutes.

Mass SpectrometryInstrument: TSQ Quantum DiscoverySource: ESIIon polarity: PositiveSpray voltage: 3.5 kVSheath/Auxiliary gas: NitrogenSheath gas pressure: 50 (arbitrary units)Auxiliary gas pressure: 15 (arbitrary units)Ion transfer capillary temperature: 350 °C Scan type: SRMCID conditions: Ar at 1.5 mTorr

Administrator

Text Box


of 6

Multi-residue OptimizationOne of the most time consuming parts in the developmentof a large multicomponent assay is the optimization ofMS/MS parameters for each analyte. The TSQ QuantumDiscovery allows multicomponent optimization of MS/MSparameters to be carried out automatically, thus allowingfor faster method development. Up to eight SRM transi-tions can be optimized simultaneously, either from a singleparent component or from multiple components. In effect,this means the ability to carry out the optimization proce-dure 11 times for 88 pesticides (instead of 88 times if theywere carried out singly), thus saving a significant amountof time in method development. An example of this isgiven in Figure 1a, displaying the simultaneous optimiza-tion of eight SRM transitions from four pesticides. Thestructures of these compounds are shown in Figure 1b.

MS Instrument MethodTo accommodate such a large number of components overa short time range, the acquisition time was divided intotwo segments, each containing three scan events. Allowingfor analyte overlap between the time segments, a total of59 SRM transitions were performed in segment one and56 SRM transitions in segment two, with dwell times of20 ms for each transition. A graphical representation ofthe actual instrument method is shown in Figure 2.

10 20 30 40 50 60

100

80

60

40

20

0

Rela

tive

Inte

nsity

Collision Energy (V)

Figure 1a: Simultaneous multicomponent optimization of MS/MS parameters

COMPOUND PARENT PRODUCT COLLISION M/Z M/Z ENERGY V

Bendiocarb 224.1 167.1 10Bendiocarb 224.1 109.0 22

Pyrimethanil 200.1 107.1 28Pyrimethanil 200.1 182.1 28

Thiabendazol 202.0 175.0 30Thiabendazol 202.0 131.1 36

Cyprodinil 226.1 93.1 40Cyprodinil 226.1 77.1 46

Figure 2: Splitting the acquisition time into two time segments and three scan events improves instrument performance for complex screening analyses

Mwt Pesticide Structure (da) Use

Bendiocarb 223 Insect control against public health, industrial & storage pests

Pyrimethanil 199 Fungal controlon vine/fruits, ornamentals

Thiabendazol 201 Fungal control on cotton, barley, bananas

Cyprodinil 225 Fungal control in cereals, grapes,strawberries

O

O

O O

HN

NH N

N

N

NHS

N

NH N

N

Figure 1b: Structures of the four pesticides used to generate the optimization graph of Figure 1a

Administrator

Text Box


of 6

Results and DiscussionFigure 3a shows the LC-MS/MS chromatogram generatedfrom the pesticide mix eluting over a chromatographictime scale of 16 minutes. The complexity of the chromato -

gram can be seen by expanding the area from 8 to 11minutes (Figure 3b), where several different pesticidescan typically be observed to co-elute.

Figure 3a: LC-MS/MS chromatogram of 88 pesticides at 50 pg/µL

Figure 3b: Detection of minor components under the larger peaks

Administrator

Text Box


of 6

All compounds were mixedat the same concentration andthe SRM method allows eventhose with low responses to bedetected under other analytes.A summary of the results forthese pesticides at 50 pg/μLis tabulated in Table 1. As isclearly evident, excellent lin-earity was observed with thecoefficient of correlations ofmost components varyingfrom 0.9900 to 0.9998.

Daminozoid 1.04 4674516 50.000 48.743 Y = 56652 + 94739.2 × R2 = 0.9988Methamidophos 1.27 3829646 50.000 49.984 Y = −27105.9 + 77160.3 × R2 = 0.9992

Acephate 1.37 2535282 50.000 47.557 Y = 51553.4 + 52226.8 × R2 = 0.9955Omethoat 1.87 1516561 50.000 51.212 Y = −20184.5 + 30007.4 × R2 = 0.9985

Propamocarb 2.13 4988264 50.000 49.355 Y = −31459.9 + 101707 × R2 = 0.9997Aldicarb-sulfoxid 2.45 15089 50.000 90.324 Y = 1383.55 + 151.735 × R2 = 0.4642

Butocarboxim-sulfoxid 2.45 92079 50.000 59.539 Y = −5384.2 + 1636.97 × R2 = 0.9372Butoxycarboxim 2.62 30232 50.000 63.008 Y = −761.85 + 491.892 × R2 = 0.8296

Aldoxycarb 2.55 210393 50.000 49.163 Y = −5547.18 + 4392.29 × R2 = 0.9977Pymetrozin 2.48 6401121 50.000 50.687 Y = −40420.7 + 127084 × R2 = 0.9995

Carbendazim 3.63 67750507 50.000 49.534 Y = −61584.2 + 1.369e + 006 × R2 = 0.9999Methomyl 3.56 298259 50.000 57.081 Y = 9692.33 + 5055.42 × R2 = 0.9934

Demeton-S-methyl-sulfon 3.71 3006988 50.000 49.419 Y = −13010.2 + 61109.5 × R2 = 0.9995Oxidemeton-methyl 50.000 N/F Y = −38265.8 + 97760.8 × R2 = 0.9989

Monocrotophos 4.17 1366861 50.000 52.403 Y = −16543.9 + 26399.4 × R2 = 0.9964Ethiofencarb-sulfon 4.30 277016 50.000 48.786 Y = −1760.21 + 5714.23 × R2 = 0.9983

3-hydroxy-carbofuran 4.97 292068 50.000 51.737 Y = −3310.44 + 5709.23 × R2 = 0.9871Ethiofencarb-sulfoxid 4.87 1110901 50.000 48.253 Y = −24680.6 + 23533.9 × R2 = 0.9964

Thiabenzadol 5.17 24648309 50.000 49.962 Y = −58379.8 + 494513 × R2 = 0.9999Dimethoat 5.28 3611384 50.000 50.512 Y = −18998 + 71871.1 × R2 = 0.9997

Vamidothion 5.40 472217 50.000 51.537 Y = −5505.94 + 9269.53 × R2 = 0.9982Imidacloparid 5.59 1506719 50.000 52.085 Y = −9529.36 + 29111.3 × R2 = 0.9969

Metamitron 5.52 2252454 50.000 49.143 Y = 1941.62 + 45795.4 × R2 = 0.9993Quinmerac 5.58 7922156 50.000 50.826 Y = −21139.4 + 156284 × R2 = 0.9995

Clethodim-imin-sulfon 5.67 2327412 50.000 48.892 Y = −1206.9 + 47627.9 × R2 = 0.9992Pirimicarb 5.73 20969231 50.000 50.280 Y = −50637 + 418055 × R2 = 0.9999

Clethodim-imin-sulfoxid 6.12 5154573 50.000 49.645 Y = 4532.14 + 103738 × R2 = 0.9998Butocarboxim 6.41 3903019 50.000 48.852 Y = −2919.87 + 79954.2 × R2 = 0.9992

Aldicarb 6.50 2151719 50.000 49.942 Y = 11575.3 + 42852.7 × R2 = 0.9998PyridateXX 7.02 8244302 50.000 49.736 Y = −6792.58 + 165899 × R2 = 0.9999Thiacloprid 7.20 5497747 50.000 50.557 Y = −37790.7 + 109491 × R2 = 0.9997

Propoxur 7.21 2762590 50.000 49.906 Y = −7780.97 + 55512 × R2 = 0.9996Thiophanat-methyl 7.53 1420464 50.000 50.642 Y = 731.145 + 28034.9 × R2 = 0.9989

Bendiocarb 7.44 917258 50.000 49.829 Y = −4657.68 + 18501.5 × R2 = 0.9991Carbofuran 7.44 19195703 50.000 49.441 Y = −84755.6 + 389971 × R2 = 0.9996

Cinosulfuron 7.58 794473 50.000 47.468 Y = −5092.19 + 16844.4 × R2 = 0.9956Triasulfuron 7.62 440622 50.000 48.954 Y = −7849.1 + 9161.13 × R2 = 0.9973

5-hydroxy-clethodim-sulfon 8.13 356038 50.000 49.817 Y = −1197.11 + 7170.92 × R2 = 0.9995Ethiofencarb 8.09 1655866 50.000 51.049 Y = −2896.19 + 32493.3 × R2 = 0.9989

Metsulfuron-methyl 8.10 534961 50.000 52.066 Y = −8787.37 + 10443.5 × R2 = 0.9964Nicosulfuron 8.18 55407 50.000 43.245 Y = −7332.14 + 1450.79 × R2 = 0.9559

Carbaryl 8.21 665994 50.000 53.046 Y = −8256.25 + 12710.6 × R2 = 0.9937Chlorosulfuron 8.45 490049 50.000 49.618 Y = −7636.28 + 10030.3 × R2 = 0.9977

Isoxaflutole 8.63 2483727 50.000 50.356 Y = −30166.7 + 49922.5 × R2 = 0.9996Amidosulfuron 8.58 44104 50.000 51.395 Y = 1474.67 + 829.433 × R2 = 0.9605

Metalaxyl 8.72 2766611 50.000 50.460 Y = −19113 + 55207.1 × R2 = 0.9998Imazalil 8.64 7172377 50.000 51.281 Y = −146813 + 142726 × R2 = 0.9951Atrazin 8.84 22304226 50.000 49.756 Y = −79337.7 + 449862 × R2 = 0.9998

3,4,5-Trimehacarb 9.08 5876561 50.000 49.047 Y = −31771.1 + 120463 × R2 = 0.9994Clethodim-sulfon 9.14 606123 50.000 47.974 Y = −6241.14 + 12764.5 × R2 = 0.9963

Desmedipham 9.42 1513363 50.000 48.104 Y = −38755.8 + 32266.2 × R2 = 0.9944Phenmedipham 9.43 1349750 50.000 49.328 Y = −31936.5 + 28010.3 × R2 = 0.9982

Pyrimethanil 9.13 11739201 50.000 49.761 Y = −72893 + 237379X × R2 = 0.9996Isoproturon 9.22 7708543 50.000 50.569 Y = −36587.8 + 153161 × R2 = 0.9996

Fenpropimorph 9.38 41217540 50.000 50.340 Y = −1.58954e + 006+850358 × R2 = 0.9877Thiodicarb 9.39 131141 50.000 46.936 Y = −1721.31 + 2830.73 × R2 = 0.9925

Flazasulfuron 9.65 275363 50.000 51.509 Y = −8454.51 + 5510.05 × R2 = 0.9964Bensulfuron-methyl 9.57 337368 50.000 44.644 Y = −60.8872 + 7558.17 × R2 = 0.9836Clethodim-sulfoxid 9.60 673868 50.000 47.765 Y = −9675.27 + 14310.6 × R2 = 0.9951

Diuron 9.75 2269026 50.000 50.361 Y = −9647.22 + 45246.5 × R2 = 0.9996Prosulfuron 9.93 166599 50.000 42.400 Y = −2469.03 + 3987.42 × R2 = 0.9581

Azoxystrobin 9.85 5924744 50.000 50.076 Y = −109126 + 120494 × R2 = 0.9973Methiocarb 9.95 1164890 50.000 49.681 Y = −8196.01 + 23612.5 × R2 = 0.9991Promecarb 9.95 1530917 50.000 51.120 Y = −20134.2 + 30341.5 × R2 = 0.9990

Iprovalicarb 10.09 2190910 50.000 50.988 Y = −17552.3 + 43313.4 × R2 = 0.9994 Fenhaxamid 10.25 2163320 50.000 51.113 Y = −21459.1 + 42744.1 × R2 = 0.9990

Linuron 10.26 1668513 50.000 48.797 Y = −26949.6 + 34745.1 × R2 = 0.9980Triflusulfuron-methyl 10.19 5971 50.000 54.571 Y = 2451.06 + 64.4984 × R2 = 0.6233

Cyprodinil 10.45 14661158 50.000 50.662 Y = −222152 + 293778 × R2 = 0.9982Spiroxamine 10.39 75041736 50.000 50.560 Y = −2.9315e + 006 + 1.54218e + 006 × R2 = 0.9880Metolachlor 10.64 15042434 50.000 49.975 Y = −148631 + 303973 × R2 = 0.9992Tebufenzoid 10.83 1226295 50.000 55.610 Y = −14352.5 + 22309.7 × R2 = 0.9985

Thiofanox 10.93 118596 50.000 -129.926 Y = 110983 − 58.591 × R2 = 0.0058Fenoxycarb 11.08 409691 50.000 49.579 Y = −27661.2 + 8821.35 × R2 = 0.9956

Fentin-hydroxide 11.02 4603640 50.000 50.016 Y = −89525.3 + 93833.1 × R2 = 0.9963Diflubenzuron 11.30 700297 50.000 49.001 Y = −7945.49 + 14453.7 × R2 = 0.9879

Tebuconazol 11.38 7987902 50.000 49.908 Y = −170977 + 163478 × R2 = 0.9964Rimsulfuron 50.000 N/F N/A

Haloxyfop-methyl 11.83 8823982 50.000 48.905 Y = −344151 + 187469 × R2 = 0.9885Indoxacarb 11.88 370334 50.000 43.918 Y = −28514.8 + 9081.59 × R2 = 0.9678Triflumuron 11.81 1448162 50.000 50.860 Y = −72372.4 + 29896.2 × R2 = 0.9902

Clethodim 12.15 856888 50.000 59.816 Y = −5040.92 + 14409.7 × R2 = 0.9991Fluzifop-P-butyl 12.26 8028262 50.000 51.498 Y = −365417 + 162992 × R2 = 0.9866

Haloxyfop-ethoxyethyl 12.27 3551906 50.000 52.226 Y = −131575 + 70529.2 × R2 = 0.9887Flurathiocarb 12.44 1424774 50.000 48.972 Y = −65299.8 + 30427.4 × R2 = 0.9828

Quizalofop-ethyl 12.52 9664875 50.000 51.959 Y = −380173 + 193325 × R2 = 0.9886Flufenoxuron 13.73 40186 50.000 44.371 Y = −10159.7 + 1134.66 × R2 = 0.9607

Pyridate 15.28 66084 50.000 40.512 Y = −18878.2 + 2097.21 × R2 = 0.8556

Component Specified Calculated

Name RT Area Amount Amount Equation

Table 1: Results of pesticideanalysis at 50 pg/µL

Administrator

Text Box


of 6

Linearity

Peak areas were used for quantitation and the resultantlinearity of responses are plotted in Figure 4 for five different compounds. Although no internal standard wasused during the assay, excellent correlation coefficientvalues were observed between 0.9990 and 0.9998 for the five components metamitron, clethodimin-sulfoxide,isoxaflutole, iprovalicarb, and methiocarb.

Absence of Cross-talk

High-throughput characterization of very complex mixturesrequires rapid analysis of coeluting analytes. An effectiveway to accomplish this is by reducing the dwell andinterscan times. However, cross-talk can occur in triplequadrupole instruments when short scan times areemployed because the fragment ions from one SRMtransition are often scanned out during another transition.This is due to some fragment ions from one transitionstill residing in the collision cell when the next transitionstarts, resulting in signal artifacts. However, the patenteddesign of the orthogonal collision cell of the TSQQuantum Discovery virtually eliminates cross-talk.

This was demonstrated during the pesticide assay bymonitoring the SRM transitions of three components: triasulfuron, metasulfuron-methyl and chlorosulfuron.These compounds have different precursor ions but all generate a product ion at m/z 167 (see Table 2). The chromatograms in Figure 5 show the transitions for these compounds, with no evidence of any cross-talk.

250,000

200,000

150,0000

100,0000

500,000

0

0 10 20 30 40 50

Area

Isoxaflutole

2,500,000

2,000,000

1,500,000

1,000,000

500,000

0

Area

0 10 20 30 40 50

Metamitron

1,200,000

1,000,000

800,000

600,000

400,000

200,000

0

0 10 20 30 40 50

Area

Methiocarb

2,200,000

2,000,000

1,800,000

1,600,000

1,400,000

1,200,000

1,000,000

800,000

600,000

400,000

200,000

0

Area

Iprovalicarb

0 10 20 30 40 50

5,500,000

5,000,000

4,500,000

1,500,000

1,000,000

0

0 10 20 30 40 50

4,000,000

3,500,000

3,000,000

2,500,000

2,000,000

500,000

Area

Clethodim-sulfoxide

Figure 4. Linearity of response for the five components: metamitron, clethodim-sulfoxide, isoxaflutole, iprovalicarb, methiocarb

Pesticide Retention time (min) SRM transition (m/z)

Triasulfuron 7.62 402 > 167Metasulfuron-methyl 8.10 382 > 167Chlorosulfuron 8.45 358 > 167

Table 2: Characteristics of the three pesticides – triasulfuron, metasulfuron-methyl, and chlorosulfuron – used to demonstrate zero cross-talk on the TSQ Quantum Discovery

Administrator

Text Box


Conclusions

An LC-MS/MS screening assay to monitor 88 pesticidesusing minimal LC separation was developed using theTSQ Quantum Discovery. It was possible to detect allcomponents within a chromatographic time scale of 16minutes by performing SRM transitions during two user-determined time segments. Even with dwell times of only 20 ms, no cross-talk interference was observedduring the analysis.

Typical food monitoring applications require screeningfor tens to hundreds of pesticides. Although conventionaldetection is accomplished using UV absorbance, nitrogenphosphorus, or electron capture detection, LC-MS/MSprovides superior sensitivity, and more importantly, specificity of identification as compared with these othercommonly used techniques. The LC-MS/MS-based methoddescribed here, with its speed, sensitivity, and specificity,is highly applicable to both the environmental monitoringand agrochemical industries operating within EPA andFDA criteria.

CD-ROMThe data generated for this application note, alongwith the instrument and processing methods, areavailable on a CD-ROM from Thermo Fisher Scientificat www.thermo.com/quantum.

References

U.S. Environmental Protection Agency website at www.epa.gov: USEnvironmental Protection Agency Code of Federal Regulations 40. Chapter I,Subchapter E, Part 180 details the tolerances and exemptions from tolerancesfor pesticide chemicals in food.

Pesticide Analytical Manual Volume 1, Sections 605-606 (describes MS appli-cations and benefits). Transmittal No. 94-1 (1/94), Form FDA 2905a (6/92).Available on the FDA website at www.cfsan.fda.gov. Chapter 3 describesmulti-class multi-residue methods, while Chapter 4 provides selective multi-residue methods.

7.62

8.10

8.45

100

90

80

70

60

50

40

30

20

10

0100

90

80

70

60

50

40

30

20

10

0

100

90

80

70

60

50

40

30

20

10

0

6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5

Time (min)

Rela

tive

Abun

danc

e

Triasulfuron

Metasulfuron-methyl

Chlorosulfuron

Figure 5: No cross-talk was observed when the pesticides triasulfuron, metsulfuron-methyl, and chlorosulfuron were detected

AN62488_E 11/07S










www.thermo.com



Administrator

Text Box


Testing LC-MS System Robustnesswith Automated Sample CleanupUsing Red Wine as a MatrixJonathan Beck, Thermo Fisher Scientific, San Jose, CA, USA


Key Words


• EQuan System


• Pesticides

Introduction

Achieving low limits of detection (LODs) of pesticides,antibiotics and veterinary residues in food residues anddrinking water is of paramount importance in order tomonitor the regulatory levels as stated by the US, Japaneseand EU directives. These substances pose a significanthealth threat and therefore, need to be accurately detectedat the lowest levels, typically at part per trillion (ppt).Traditionally, LC-MS/MS has been used by the environ-mental and food industries for the identification andquantitation of these residues. However, this methodologytypically requires extensive offline sample preparation,which can be time consuming and expensive.

The Thermo Scientific EQuan environmental quan ti ta -tion system consists of a Thermo Scientific TSQ Quantumseries mass spectrometer, two Thermo Scientific SurveyorHPLC pumps with a preconcentration column, an analyticalcolumn, and a CTC autosampler. The unique capabilities ofEQuan for online preconcentration and cleanup of samplesresult in improved sensitivity and precision, as well asunmatched throughput.

In previous experiments, using the EQuan system foronline sample preconcentration and detection of pesticidesin ground water yielded lower limits of detectioncompared to standard injection techniques. See Table 1.

Typically, when red wine is analyzed using LC-MS/MS,some form of sample preparation and/or extraction isnecessary prior to injection. In this application note, theEQuan system was tested for robustness using a matrixof neat red wine spiked with a mixture of pesticides usinglarge volume (1000 µL) injections.

Goal

To test the robustness of an LC-MS system for anautomated online preconcentration system using adirty matrix.


Sample PreparationRed Burgundy wine was spiked with a mixture of nineherbicides and six fungicides at a level of 500 pg/mL(500 ppt). The following herbicides were analyzed:atrazine, cyanazine, simazine, propazine, trietazine,metazachlor, propachlor, pendimethalin, and propyzamide.The following fungicides were analyzed: flutriafol,triadimefon, epoxiconazole, flusilazole, tebuconazole,and propiconazole. No other sample treatment wasperformed prior to injection.

HPLCHPLC analysis was performed using an HTC PAL™

Autosampler with two LC quaternary pumps and two LCcolumns, the first for preconcentration of the sample andthe second for the analytical analysis. A sample of 1000µL of the spiked neat wine was injected directly onto theThermo Scientific Hypersil GOLD 20 × 2.1 mm, 12 µmloading column in a high aqueous mobile phase (seeFigure 1a). After 1 minute, a six-port valve on the massspectrometer was switched by Thermo Scientific LCQUAN

2.5 instrument control software. This enabled the loadcolumn to be back flushed onto the analytical column(Hypersil GOLD™ 50× 2.1mm, 3 µm), where the com-

1 mL 100 µL 1 mL 100 µL 1 mL 100 µL 1 mL 100 µLInjection Injection Gain Injection Injection Gain Injection Injection Gain Injection Injection Gain

Area Area Factor Area Area Factor Area Area Factor Area Area FactorPropham Isoproturon Diuron Linuron

1 ppt 5.53E+04 NA 1.97E+04 1.73E+03 115 ppt 2.17E+04 3.35E+05 3.17E+04 11 4.15RE+04 5.65E+03 7 6.96E+03

10 ppt 2.71E+04 6.68E+05 4.90E+04 14 8.25E+04 1.18E+04 7 1.99E+0450 ppt 5.09E+04 3.33E+06 2.82E+05 12 4.47E+05 3.72E+04 12 5.91E+04 7.98E+03 7

100 ppt 6.51E+04 6.54E+06 5.24E+05 12 8.83E+05 7.60E+04 12 1.34E+05 2.50E+04 5500 ppt 2.47E+05 3.00E+04 8 3.11E+07 2.60E+06 12 4.65E+06 3.80E+05 12 7.36E+05 1.28E+05 6

1000 ppt 5.29E+05 5.69E+04 9 5.81E+07 5.23E+06 11 9.39E+06 7.63E+05 12 1.43E+06 2.47E+05 65000 ppt 2.59E+06 2.82E+05 9 2.58E+08 2.44E+07 11 4.95E+07 3.68E+06 13 9.49E+06 1.25E+06 8

Table 1: Calculations demonstrating the gain in peak areas due to larger injection volumes in ground water samples

Administrator

Text Box


PEEK™ TubingInternal Flow Routing

Autosampler

3 µm Hypersil GOLDAnalytical Column

12 µm Hypersil GOLDPreconcentration

Column

TSQ QuantumSeries MS

6-PortValve

SurveyorLC Pump Plus

SurveyorMS Pump Plus

Preconcentrate

Waste LC Pump

MSLoad Pumpand CTC

1 mL/min

PreconcentrationColumn

Analytical Column

400 µL/min

Analyze

Waste LC Pump

MSLoad Pump/CTC

PreconcentrationColumn

Analytical Column

pounds were separated prior to introduction into the massspectrometer (see Figure 1b). After all of the compoundswere eluted from the analytical column, the 6-port valvewas switched back to the starting position, and theloading and analytical columns were cleaned with a high

organic phase before being re-equilibrated to their startingconditions. The total run time for each analysis was 22minutes. The mobile phases for the analysis were waterand methanol, both with 0.1% formic acid.

Figure 1: The schematic of the EQuan system used for this assay

Figure 2: Ion sweep cap after several hundred injections,showing contamination from red wine

Figure 3: Electrospray ionization source with the electrospray probe removed,showing the main spray pattern directed towards the drain

a b

Administrator

Text Box


MSMS analysis was carried out on a Thermo Scientific TSQQuantum Ultra triple quadrupole mass spectrometer withan electrospray ionization source. The MS conditions wereas follows:

Electrospray ionization: PositiveSpray voltage: 3.0 kVIon transfer tube temperature: 350 °CSheath gas pressure: 45 arbitrary unitsAuxiliary gas pressure: 5 arbitrary unitsIon sweep gas pressure: 3 arbitrary unitsCollision gas (Ar): 1.0 mTorrQ1/Q3 peak resolution: 0.7 DaScan width: 0.002 DaThe source of the mass spectrometer was adjusted so

that the ESI probe was off axis to prevent contaminationof the ion transfer tube. The position of the probe was setso that the main spray pattern of the electrospray hit theThermo Scientific Ion Sweep cone below the center lineand off to the left by about 0.5 cm. The probe depth wasset to position “C” on the electrospray probe. An ionsweep gas of three arbitrary units was set to prevent anylarge droplets from entering the ion transfer tube of themass spectrometer.


The back pressure of the loading column and the analyticalcolumn were monitored over the course of the wine injec-tions to determine if the columns were becoming cloggedwith any particulates from the wine. Over 600 injections,the back pressure on the 12 µm loading column remainedat approximately 20 bar under the starting conditions ofthe analytical run, while the back pressure on the 3 µm

analytical column remained at approximately 72 bar.The resulting spray pattern of the electrospray can

be seen in Figure 2. A thick deposit of red wine residueis clearly visible from just below the center of the sweepcone to the outside radius. The red wine spray can also beseen on the inside of the electrospray housing in Figure 3.In the picture, the drain is dark purple in color, illustratingthat the main excess spray of the red wine was directed tothe bottom of the ion source and away from the mainorifice of the mass spectrometer. Additionally, the ESIprobe can be adjusted to be closer to the ion transfer tube,which increases robustness by allowing less side scatterfrom the electrospray beam, thus focusing the main spraypattern lower on the ion sweep cap.

The reproducibility of the method is shown in Figure4. The graph plots the peak area for metazachlor for 164injections of red wine. The first four injections wereexcluded from the %RSD calculation. Because the loadingcolumn was new at the beginning of the runs, severalinjections were required to condition the column beforea stable peak area was achieved. A representative chro-matogram is shown in Figure 5.

As shown in Figure 6, after several hundred injectionsof the spiked red wine matrix, no degradation in columnperformance or source robustness was observed. In total,over 600 injections were made on the system with no lossin column performance.

Conclusion

This application note demonstrates the robustness of theTSQ Quantum Ultra triple quadrupole mass spectrometerand an online extraction and preconcentration method.The described sample cleanup technique improves signal-

to-noise ratios by a factor of 10 to100 (based on injection volume)for low concentration samples inred wine matrices. Preliminaryresults using onionand tobacco matrices have yieldedsimilar results in terms of columnperformance and mass spectrometerrobustness. Further studies will beconducted in other matrices, as wellas with other pesticides, herbicides,and insecticides.

References1 Wang, Y.; Catana, F.; Yang, Y.; Roderick, R.;

van Breemen, R.B. “An LC-MS Method forAnalyzing Total Resveratrol in Grape Juice,Cranberry Juice, and in Wine”; J Agric.Food Chem. 2002, 50(3), 431-435.0 50 100 150 200

Injection Number

Metazachlor Peak Area

12000000

10000000

8000000

6000000

4000000

2000000

0

Are

a

Figure 4: Scatter plot of the peak area for 164 injections (1000 µL) of metazachlor spiked in red wine.The %RSD is 9% when the first four points are excluded.

Administrator

Text Box


Simazine202 > 124

Propachlor212 > 170

Propazine/Trietazine230 > 188

Cyanazine241 > 214

Simazine202 > 124

Propyzamide256 > 190

Metazachlor282 > 212

Triadimefon294 > 197

Pendimethalin282 > 212

Flutriafol302 > 123

Tebuconazole308 > 125

Flusilazole316 > 247

Epoxyconazole330 > 121

Propiconazole342 > 159

Figure 5: Example chromatograms for a 1000 µL injection of spiked red wine

RT: 8.13-13.32

9 10 11 12 13Time (min)

100

90

80

70

60

50

40

30

20

10

0

9TH Injection 321ST Injection

Figure 6: Different injections of metazachlor (retention times have been offset for greater visibility)

AN62495_E 11/07S


Legal Notices©2007 Thermo Fisher Scientific Inc. All rights reserved. HTC PAL is a trademarks of CTC Analytics AG, Zwingen, Switzerland. All other trademarks are theproperty of Thermo Fisher Scientific Inc. and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scientific Inc.products. It is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Specifications,terms and pricing are subject to change. Not all products are available in all countries. Please consult your local sales representative for details.

View addional Thermo Scientific LC/MS application notes at: www.thermo.com/appnotes








www.thermo.com


Administrator

Text Box


UHPLC Separation of Triazine Herbicides at Elevated TemperatureDave Thomas, Thermo Fisher Scientific, San Jose, CA USA

Goal:

Increase throughput of the HPLC method for triazineherbicides by employing ultra high-speed liquidchromatography at elevated temperature on a heat stableHypercarb column.

IntroductionTemperature is a key variable in high performance liquidchromatography (HPLC), influencing solute diffusionrates, mobile phase viscosity, and solubility. For example,as column temperature increases, analyte diffusionincreases. Increased analyte diffusion generally leads to anincrease in the optimum linear velocity of the separation,so that equivalent chromatographic efficiency andresolution can be achieved at a higher flow rate.Furthermore, elevating the temperature reduces theoperating backpressure. The net result is that separationscan be performed faster without exceeding the pressurelimitations of the instrument.

This application uses a porous graphitic carbonstationary phase thermostatted in a high temperaturecolumn oven to separate triazine herbicides 5 to 10 timesfaster than is typical with conventional HPLC. Thetriazines and degradation products are separated on theThermo Scientific Accela High Speed LiquidChromatograph in 2 minutes on a Thermo ScientificHypercarb 3 µm, 1 x 100 mm column operated at 160 °C.This application note also documents the performance ofthe high temperature liquid chromatographic method,including precision of retention time and peak area,resolution, and spike recovery from several environmentalwater matrices.

Experimental

Instrumentation

Thermo Scientific Accela HPLC system with PDA DetectorThermo Scientific ChromQuest 5.0 Chromatography Data System (CDS)Polaratherm Series 9000 Total Temperature Controller (Selerity Technologies)

Chromatographic conditions

Column: Thermo Scientific Hypercarb 3.0 µm, 1 x 100 mm(35003-101046)

Mobile phase: A: water B: acetonitrileGradient: Time %A %B

0.00 75 251.00 70 302.20 10 902.30 75 254.00 75 25

Flow rate: 500 µL/minDetector: PDA, 238 nm, 10-mm flow cell, 11nm bw, 20 Hz,

0s rise timeColumn temp.: 160 °C (housed in Selerity temperature controller)Injection: 5 µL sample loop, 2 µL partial loop injection

Syringe Speed: 4 µL/secFlush Speed: 100 µL/secFlush Volume: 400 µLWash Volume: 200 µLFlush/Wash source: Bottle with 90:10 methanol:water

Chemicals

Water, LC/MS-grade Fisher Scientific W6Acetonitrile, LC/MS-grade Fisher Scientific A998Methanol, LC/MS-grade Fisher Scientific A456Atrazine Supelco 49085Ametryn ULTRA PST-024Cyanazine ULTRA PST-1360Deisopropylatrazine, 1000 mg/L SPEX CertiPrep S-1135Desethylatrazine, 1000 mg/L SPEX CertiPrep S-1145Propanil, 1000 mg/L SPEX CertiPrep S-3155Propazine ULTRA PST-850Prometryn ULTRA PST-840Simazine ULTRA PST-1130Simetryn Chem Service PS-381

Consumables

Autosampler vials, 1.8 mL glass, Thermo Scientific yellow septa A4954-010Backpressure assembly Upchurch P-788Ferrules, high temperature Selerity Technologies BM0054Mixer, 50 µL in-line static Thermo Scientific 109-99-032Mobile Phase Preheater, Selerity Technologies AD1040.005” x 70 cmSyringe filters, 0.45 µm Nylon Thermo Scientific A5307-010Sample Loop, 5 µL Thermo Scientific 109-99-007

Key Words

• Accela UHPLC

• EnvironmentalAnalysis

• Herbicides

• High Temperature

• Hypercarb LCColumn

• Triazines


Administrator

Text Box


Mobile Phase

Proportioned mobile phase: Filled Solvent Reservoir BottleA of the Accela pump with fresh HPLC-grade water andpurged the solvent line with at least 30 mL of the water.Connected a fresh bottle of HPLC-grade acetonitrile toReservoir B and purged as above.

Calibration Standards

Individual Stock Solutions, 1000 mg/L: Accuratelyweighed 10 mg (0.010 g) of each neat compound into a10-mL volumetric flask, added 5 mL acetonitrile, andsonicated to dissolve. Brought to volume with acetonitrileand mix. Used desethyl atrazine, deisopropyl atrazine andpropanil, purchased as solutions of 1000 mg/L inmethanol, as received.

Combined Intermediate Standard 100 mg/L: Used acalibrated pipette to deliver 1000 µL of each individualstock solution to a 10-mL volumetric flask. Brought tovolume with acetonitrile and mix.

Calibration standards: Used a calibrated pipette to dilutethe intermediate standard with mobile phase in volumetricglassware to 30, 10, 3, 1, 0.3, 0.1, and 0.03 mg/L.

Samples

Samples of surface water (Salinas River, Monterey County,CA), ground water (domestic well, Santa Cruz county,CA), and drinking water (San Jose, CA tap water) werecollected in accordance with established procedures,stored at 4 - 8 °C, and were filtered through a 0.45 µmnylon syringe filter into a glass autosampler vial beforeanalysis.

System Preparation

To ensure good performance of this application, preparethe system as directed in Appendix A.

Results

Separation of seven triazine herbicides, two triazinedegradation products often found in environmentalsamples, and propanil is shown in Fig 1. To optimize thisseparation, we adjusted the mobile phase composition toelute the first analyte with a capacity factor k’ > 2,thereby improving resolution of the target analytes fromsample matrix junk. Analytes spanning a wide range ofpolarity are well resolved by the combination of hightemperature, solvent gradient, and the selectivity of theHypercarb stationary phase. Note that because of thereduced viscosity of the mobile phase at 160 °C, thisseparation occurs at a linear flow rate of 15 mm/s —equivalent to a flow rate of over 10 mL/min on a 4.6 mmi.d. column. The system backpressure under theseconditions is less than 4000 psi (272 bar).

Method performance is characterized by peakresolution, linear calibration range, limits of detection,and precision of retention time and peak area, assummarized in Table 2. MDLs for each analyte weredetermined by performing seven replicate injections ofLC/MS-grade water fortified at a concentration of three tofive times the estimated instrument detection limits,calculating the standard deviation of the measuredconcentration, and using the equation given in the figurecaption. Note the good precision of retention time andpeak area, as this reflects the temperature stabilitymaintained by the high temperature oven.

We analyzed several environmental water samples todemonstrate the efficacy of this method with real matrices.Samples of surface, ground and municipal drinking waterwere analyzed before and after fortification with a knownamount of each target analyte. Spike recovery wascalculated as the amount of each analyte found in thespiked sample divided by the amount expected (i.e., theamount determined in the blank plus the amount added inthe spike). Even with the dirtiest matrix, Salinas Riverwater, the target analytes are well separated from the earlyeluting matrix peaks (Figure 2) and recovery of the spikedanalytes exceeds 80% (Table 3).

Administrator

Text Box


1 2 30

50

100

Minutes

mAU

6 78 9

10

12

3 11

5

2

4

1

A

B

C

D

Figure 2: Chromatograms of four environmental water samples spiked with 200 µg/L each of triazinesand propanil. Chromatograms obtained on the Accela High Speed LC by reversed phase chromatographywith UV absorbance detection at 215 nm. Peaks: see Figure 1. Samples: top trace A, surface water(Salinas River); trace B, ground water (Simoes’ well); trace C, drinking water (San Jose tap); bottomtrace D, HPLC-grade water. Conditions: see text for details.

Minutes

0 1 2 3

mAU

100

50

0

67

8

910

12

3

115

2

4

1

Figure 1: Separation of triazine herbicides, degradation products, and propanil on the Accela HighSpeed LC by reversed-phase chromatography with UV absorbance detection at 215 nm. Peaks: seeFigure. Sample: overlay of 30 injections of triazines in HPLC-grade water with 20% acetonitrile.Conditions: see text for details.

Column: Hypercarb 3 µm, 100 x 1 mmTemperature: 160 ºCFlow rate: 500 µL/minDetector: Accela PDA at 215nm, 20Hz, 0s rise timeInjection: 2 µL partial loop from 5 µL loopSolvents: A: Water

B: AcetonitrileGradient: Time (min) A% B%

0.00 75 251.00 70 302.20 15 852.10 75 254.00 75 25

Samples: triazines and propanil in 20% acetonitrilePeaks: 1. Melamine

2. Unknown3. Deisopropylatrazine4. Desethylatrazine5. Cyanazine6. Propazine7. Prometryn8. Atrazine9. Ametryn10. Simazine11. Simetryn12. Propanil

Administrator

Text Box


Table 1. Useful properties of some triazine herbicides and degradation productsa

4-deethyl atrazine 6-deisopropyl atrazine Cyanazine Propazine PrometrynCAS# 6190-65-4 1007-28-9 21725-46-2 139-40-2 7287-19-6

Formula C6H10ClN5 C5H8ClN5 C9H13ClN6 C9H16ClN5 C10H19N5SMW (g/mol) 187.633 173.606 240.697 229.713 241.361pKa 0.87 1.7 4.05Log Po-w 1.51 1.15 2.22 2.93 3.51Water solubility, mg/L 3200 670 170 8.6 33MeOH solubility, g/L 45 (ethanol) 6.2 (toluene) 160

Atrazine Ametryn Simazine Simetryn PropanilCAS# 1912-24-9 834-12-8 122-34-9 1014-70-6 709-98-8

Formula C8H14ClN5 C9H17N5S C7H12ClN5 C8H15N5S C9H9Cl2NOMW (g/mol) 215.687 227.334 201.66 213.307 218.082pKa 1.7 4.1 1.62 4 2.29Log Po-w 2.61 2.98 2.18 2.8 3.07Water solubility, mg/L 34.7 209 6.2 450 152MeOH solubility, g/L 18 510 400 540

ahttp://toxnet.nlm.nih.gov

NH

Cl

OCl

CH3

NN

HN

NHNS

CH3

H3CCH3

N

N

NH

N

NH

Cl

H3C CH3

NN

NH

NHNS

H3C

CH3

CH3

CH3

N

N

NH

N

Cl

H3C

CH3

NH

CH3

N

HN

N

HNN

CH3

CH3

SH3C

CH 3

CH3

N

N

NH

N

NH

Cl

CH3

H3C CH3

CH3

HN

N N

CH3H3C

NHN Cl

N

H3C

N

NHN

N

NH2

Cl CH2N

N

HN

N

H2N

Cl

CH3

CH3

Table 2. Performance of high temperature method for triazines performed on Hypercarb 3 µm, 1 x 100 mm column at 160 °C.

Precision,Retention Precision,

Linear range, MDLb Time Peak AreaAnalyte k’a Ra mg/L r2 µg/L % RSDc % RSDc

deisopropylatrazine 2.4 1.1 0.03 – 10 0.9999 6 0.25 0.39desethylatrazine 2.8 1.2 0.03 – 30 0.9999 16 0.22 0.89cyanazine 5.6 7.2 0.03 – 30 0.9995 40 0.16 0.47propazine 6.1 1.5 0.03 – 30 0.9996 23 0.13 0.82atrazine 7.7 3.9 0.03 – 30 0.9995 14 0.10 0.87simazine 9.2 3.9 0.03 – 30 0.9994 30 0.09 0.92prometryn 10.4 3.0 0.03 – 30 0.9999 32 0.08 0.97ametryn 11.9 4.3 0.03 – 100 0.9995 8 0.04 0.64simetryn 13.1 3.7 0.03 – 100 0.9999 16 0.04 0.57 propanil 15.1 6.3 0.03 – 30 1.0000 25 0.02 0.42

a Capacity factor (k’) and Resolution (R) calculated according to Reference 1.b Detection limit MDL = σts,99 where tσ,99 = 3.14 for n = 7 replicates of the standard.c for n = 30 replicates.

Administrator

Text Box


Table 3. Percent recovery of analytes spiked into selected environmental water matrices. n = 3 replicates.

Analyte HPLC water Drinking water Ground water Surface water deisopropylatrazine 94.0 104 104 93.0desethylatrazine 93.0 110 98.5 85.0cyanazine 92.5 102 94.0 97.0propazine 97.0 106 104 101atrazine 94.5 104 101 98.5simazine 98.0 102 100 103prometryn 104 117 116 110ametryn 97.0 104 101 98.5simetryn 99.0 98.5 104 100propanil 101 103 104 80.0

Conclusion

A separation performed at 160 °C on the Accela highspeed chromatography system equipped with a heat stableHypercarb column and high temperature column ovenresolves 11 triazine herbicides in about two minutes withretention time and peak area precision better than 1%RSD for thirty replicates.

References

1. United States Pharmacopeia 30-National Formulary 25, United StatesPharmacopeia, Rockville, Maryland 20852-1790, USA.

Suppliers

Chem Service, West Chester, PA, USA (http://www.chemservice.com)Selerity Technologies, Inc., Salt Lake City, UT, USASigma-Aldrich, St. Lois, MO, USA (http://www.sigmaaldrich.com)Supelco, Bellefonte, PA, USA (http://www.sigmaaldrich.com)Thermo Fisher Scientific, Waltham, MA, USA (http://www.thermofisher.com) ULTRA Scientific, No. Kingstown, RI, USA (http://www.ultrasci.com)

Administrator

Text Box









www.thermo.com

AN62960_E 1/09S



Legal Notices©2009 Thermo Fisher Scientific Inc. All rights reserved. PolaraTherm is a trademarket of Selerity Technologies. Vespel is a registered trademark of E. I. duPont de Nemours and Company. All other trademarks are the property of Thermo Fisher Scientific Inc. and its subsidiaries. This information is presented asan example of the capabilities of Thermo Fisher Scientific Inc. products. It is not intended to encourage use of these products in any manners that mightinfringe the intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are available in all countries. Pleaseconsult your local sales representative for details.


Appendix A.

System Preparation

Pump: Always plumb the Accela system with precut andpolished 0.005” i.d. high-pressure tubing and highpressure fittings as shown in Figure 15 of the Accela PumpHardware Manual (Document 60157-97000 Revision B).For all tubing connections that you make, ensure that thetubing end is square-cut and burr-free. Firmly push thetubing into the injection valve port as you tighten thehigh-pressure fitting in order to maximize peak efficiency.Prime the pulse dampener and purge the solvent lines asinstructed in Chapter 4 of the Accela Pump manual.Verify that the pump is performing well by monitoring thepressure pulsation and by testing the pump proportioningaccuracy as described in Chapter 5 of the pump manual. Ifyour Accela pump does not include an inline 35 µLdynamic mixer, then install a 50 µL static mixer betweenthe inline high pressure filter and the Accela AS mobilephase preheater.

AS: Open the Instrument Configuration and verify thatthe Accela AS Configuration entry for “Dead volume” iscorrect (the calibrated volume in µL written on thetransfer tubing between the injection port and injectionvalve). Verify that the entry for “Loop size” is correct forthe currently installed sample loop. Fill the Flush reservoirwith 90:10 (v/v) methanol:water and flush the syringewith solvent to purge any air bubbles from the syringe and

tubing. Use the Wash/Flush conditions specified under“Conditions” to ensure low carryover between injections.Consult the Accela Getting Connected manual (Document60057-97001 Revision A) for details.

Polaratherm Column Oven: Install the Total TemperatureController according to the Polaratherm Series 9000Installation and Operation Manual. Install the Hypercarb,3 µm 1 x 100 mm column, by using a 70-cm length ofprecut and polished 0.005” i.d. high-pressure tubing withmobile phase preheater. It is important to use a heat stableHypercarb column as this does not contain any PEEKcomponents that will degrade at the temperatures used inthis method. Use the high temperature graphite/Vespelferrules and fittings described in the Series 9000 manual.Ensure that the tubing is fully pushed into the columninlet when you tighten the high-pressure fitting.

Detector: Use a 10 mm light-pipe flow cell. Install a 250psi backpressure regulator after the flow cell outlet tosuppress bubble formation in the flow cell. Verify that thedeuterium lamp has been used for less than 2000 hours.

Use Direct Control or a downloaded method toequilibrate the Accela system under the conditions shownabove. Create a method based on these operatingconditions and then create a sequence to make severalinjections of HPLC grade water. The system is ready torun standards and samples when the peak-to-peakbaseline oscillation is between 50 – 200 µAU/min (averageof 10 1-min segments) and no significant peaks elute inthe retention time window of the analytes.

Administrator

Text Box


Utility of H-SRM to Reduce Matrix Interferencein Food Residue Analysis of Pesticides byLC-MS/MS Using the TSQ Quantum DiscoveryYoko Yamagishi, Thermo Fisher Scientific, C-2F 3-9 Moriya-cho Kanagawa-ku, Yokohama 221-0022, Japan


Key Words


• Surveyor HPLC

• H-SRM

• Pesticides

• Zero Cross-talk


Sample Analysis

HPLC analysis was performed on the Thermo ScientificSurveyor HPLC System, using a Thermo ScientificHyPURITY C18 150 × 2.1 mm 5µm column. Mobilephase A was water, mobile phase B was methanol, andmobile phase C was water containing 10 mM ammoniumacetate. Solvent was pumped at 200 µL/min and analyteseluted using a linear gradient of 20% B to 99% B over15 minutes, holding at 99% B for 3 minutes, and thenreturning to 20% B for 5 minutes. Mobile phase C washeld at 1% throughout the run.

Mass SpectrometryInstrument: TSQ Quantum Discovery

Positive ESISpray Voltage: 5kVSheath/Auxiliary gas: NitrogenSheath gas pressure: 40 (arbitrary units)Auxiliary gas pressure: 40 (arbitrary units)Ion transfer capillary temperature: 380°C Scan type: SRM or H-SRMCID conditions: Ar at 1.0 mTorr

Negative ESI

Spray Voltage: 4.25 kVSheath/Auxiliary gas: NitrogenSheath gas pressure: 50 (arbitrary units)Auxiliary gas pressure: 5 (arbitrary units)Ion transfer capillary temperature: 350°CScan type: SRM or H-SRMCID conditions: Ar at 1.0 mTorr

MS Instrument Method

Thirty-five pesticide residue compounds were analyzed tofind the product ion to be used for quantitation. Three ofthe compounds were ionized using negative electrospray,while the remaining 32 were ionized using positive electro-spray in two different runs. A table of the compoundslisting SRM transitions and the optimum collision energyare shown in Table 1.

Introduction

With the recent trend of increased concern about foodsafety, the number of regulated pesticide residues in foodhas increased rapidly. In Japan, a new positive list systemfor monitoring pesticide residues will take effect in 2006.Consequently, an accurate high throughput multi-pesticidescreening method which can quantitate high number ofpesticide residues during a single analysis is required.

LC-MS/MS is fast becoming the technique of choicefor the identification and quantitation of pesticideresidues. This is due, in part, to the ease of sample prepa-ration and chromatographic conditions that LC-MS/MSallows, when compared to other techniques such as GCor HPLC with UV absorbance, nitrogen phosphorusdetection, or electron capture detection. However, it canbe extremely challenging to quantitate multi-pesticideresidues in food because of interference from complexsample matrices. Although matrix-related interferencescan be decreased by various sample clean up procedures,the analytical instrument used for the quantitation alsohas to be highly selective and sensitive. The uniqueHighly-Selective Reaction Monitoring (H-SRM) detectionmethod available with the Thermo Scientific TSQQuantum has proven to be very useful for this purpose.The analytical results of 35 pesticide residues in foodwith the H-SRM detection method are reported in thisapplication note.

Goals• Illustrate the effectiveness of H-SRM for reducing back-

ground interference and improving s/n

• Develop a multi-residue LC-MS/MS screening methodto detect 35 pesticides, and

• Exhibit the absence of “cross-talk” between co-elutingcomponents.

Administrator

Text Box


of 6


Figure 1a shows the chromatogram of the 32 pesticidesin positive ESI, and Figure 1b shows the three pesticidesunder negative ESI, all eluting over a chromatographictime scale of 18 minutes. While some compounds co-elute,the specificity of the H-SRM method allows for the indi-vidual quantitation and detection of each component,even at very low levels. A summary of the calibrationrange, linearity, and the reproducibility of each individualcompound at 5 ppb (ng/mL) is tabulated in Table 2.

Effect of H-SRM on Detection Limits

H-SRM is an acronym for Highly-Selective ReactionMonitoring (H-SRM), which is a more advanced form ofSelective Reaction Monitoring (SRM). Although tradi-tional SRM is a selective technique by itself, it still can notcompletely eliminate the interference from some foodmatrix components. Sometimes, it is possible to get incor-rect qualitative results or the quantitative analysis can notreach the required detection limits of targeted compoundsdue to matrix-related interferences. The traditional SRMexperiment, using a triple quadrupole instrument, isusually conducted with unit resolution (0.7 FWHM) forthe precursor ion. With the more advanced H-SRM, theprecursor ion is selected with a peak width of 0.1-0.2FWHM. The more stringent tolerance accounts for thehigher selectivity, which can lower LOQs and increaseprecision and accuracy at the limits of detection. This canalso, in effect help reduce the overall bench time requiredfor sample preparation.

Precursor Product Collision RetentionCompound Name Ion (m/z) Ion (m/z) Energy (V) Time (min)Oxamyl 237.17 72.0 15 3.9Imidacloprid 256.12 209.1 16 6.3Acetamiprid 223.12 126.0 23 7.3Aldicarb 208.17 116.0 8 9.0Propoxur 210.16 111.0 14 10.3Carbofuran 222.16 165.1 14 10.4Bendiocarb 224.14 167.0 10 10.4Carbaryl 202.15 145.0 10 11.0Ethiofencarb 226.13 107.0 14 11.3Pirimicarb 239.22 182.1 16 11.5Methabenzthiazuron 222.12 165.0 17 11.9MIPC 194.17 95.0 20 11.9Diuron 233.06 72.1 19 12.4Azoxystrobin 404.17 372.1 15 12.8BPMC 208.19 152.0 10 13.1Siduron 233.20 137.0 17 13.2Linuron 249.09 182.0 18 13.2Methiocarb 226.14 169.1 10 13.4Daimuron 269.21 151.1 14 13.7Cumyluron 303.14 185.0 14 13.9Tebufenozide 353.24 133.0 19 14.7Iprodione 330.07 245.1 15 14.7Diflubenzuron 311.04 158.0 14 14.8Etobenzanid 340.08 121.0 36 15.2Cyprodinil 226.18 93.0 38 15.2Phoxim 299.08 129.0 12 15.4Bitertanol 338.21 269.2 10 15.6Hexythiazox 353.13 228.0 16 16.8Piperonyl butoxide 356.26 177.1 13 17.2Flufenoxuron 489.09 158.0 20 17.4Fenpyroximate 422.26 366.1 15 17.6Chlorfluazuron 540.03 382.9 20 17.8Teflubenzuron 379.00 339.0 12 17.08Hexaflumuron 459.02 439.0 12 16.04Lufenuron 509.00 326.0 18 16.77(Positive in Black, Negative in Red)

Table 1: Summary of SRM transitions used for the analysis

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18Time

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

5.5

6

6.5

7

7.5

8

8.5

9

9.5

10

3. 9

NL : 5. 29E5 m s2 237->72

6. 3

NL : 8. 68E5 m s2 256->209

7. 3

NL : 2. 12E6 m s2 223->126

9. 0

NL : 1. 25E5 m s2 208->116

10. 3

NL : 2. 30E6 m s2 210->111

10. 411. 910. 4

NL : 3. 23E6 m s2

11. 0

NL : 2. 90E6 m s2 202->145

11. 3

NL : 2. 47E6 m s2 226->107

11. 5

NL : 2. 75E6 m s2 239->182

11. 96

NL : 2. 10E6 m s2 194->95

12. 4

NL : 8. 19E5 m s2 233->72

12. 8

NL : 6. 49E6 m s2 404->372

13. 1

NL : 1. 75E6 208->152

13. 2

NL : 1. 99E6 233->137

13. 2

NL : 9. 56E5 m s2 249->182

13. 4

NL : 3. 75E6 m s2 226->169

13. 7

NL : 7. 47E6 m s2 269->151

13. 9

NL : 5. 10E6 m s2 303->185

14. 7

NL : 1. 59E6 m s2 353->133

14. 7

NL : 9. 47E4 m s2 330->245

14. 8

NL : 1. 18E6 m s2 311->158

15. 2

NL : 4. 49E5 m s2 340->121

15. 2

NL : 7. 01E5 m s2 226->93

15. 4

NL : 1. 47E6 m s2 299->129

15. 6

NL : 5. 71E5 m s2 338->269

16. 8

NL : 7. 23E6 m s2 356->177

17. 2

NL : 2. 02E6 m s2 353->228

17. 4

NL : 1. 02E6 m s2 489->158

17. 6

NL : 5. 31E6 m s2 422->366

17. 8

NL : 5. 20E5 m s2 540->382

Relat

ive A

bund

ance

NL : 4. 57E6 m s2 222->165

Figure 1a: LC-MS/MS chromatogram of 32 pesticides at 10 ng/mL, positive ESI

Administrator

Text Box


of 6

Compound R2 Range (ppb) CV(%), n=5

Oxamyl 1.000 0.01-100 1.79Imidacloprid 0.9994 0.05-100 2.84Acetamiprid 0.9987 0.05-100 1.17Aldicarb 0.9993 0.05-100 6.89Propoxur 0.9997 0.01-100 1.70Carbofuran 0.9996 0.05-100 0.95Bendiocarb 0.9992 0.01-100 2.30Carbaryl 0.9999 0.01-100 1.44Ethiofencarb 0.9996 0.01-100 2.64Pirimicarb 0.9995 0.01-100 3.55Methabenzthiazuron 0.9989 0.01-100 1.73MIPC 0.9987 0.01-100 1.26Diuron 0.9987 0.05-100 2.23Azoxystrobin 0.9989 0.01-100 2.60BPMC 0.9999 0.05-100 1.57Siduron 0.9989 0.05-100 1.59Linuron 0.9989 0.05-100 4.04Methiocarb 0.9997 0.01-100 1.88Daimuron 0.9992 0.01-100 3.03Cumyluron 0.9993 0.01-100 3.17Tebufenozide 0.9995 0.05-100 1.83Iprodione 0.9979 0.5-100 6.17Diflubenzuron 0.9997 0.01-100 2.98Etobenzanid 0.9997 0.05-100 1.82Cyprodinil 0.9998 0.1-100 4.49Phoxim 0.9997 0.05-100 3.14Bitertanol 0.9996 0.05-100 3.54Piperonyl butoxide 0.9996 0.01-100 1.65Hexythiazox 0.9999 0.01-100 2.43Flufenoxuron 0.9997 0.01-100 3.63Fenpyroximate 0.9999 0.01-100 2.22Chlorfluazuron 0.9987 0.01-100 2.77Teflubenzuron 0.9986 0.01-100 2.35Hexaflumuron 0.9973 0.01-50 1.58Lufenuron 0.9998 0.01-10 2.56

Table 2: Calibration range and linearity of each compound, as well as thereproducibility of each compound at 5 ppb

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18T im e ( m in )

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

16.44

16.77

17.08

NL : 2. 99E5 m s2 459->439

NL : 1. 01E5 m s2 509->326

NL : 2. 01E5 m s2 379->339

Relat

ive A

bund

ance

Figure 1b: LC-MS/MS chromatogram of 3 pesticides at 10 ng/mL, negative ESI

The effects of H-SRM over SRM are clearly illus-trated for the three pesticides Iprodione, Biteranol andEtobenzanid in Figures 2 a, b, and c.


In order to quantitate mixtures of many compoundsaccurately, it is necessary to use short scan speed toensure sufficient data points for integration. It is impor-tant that the system can maintain its sensitivity withoutcross-talk even at short scan speeds. Cross-talk occurswhen ions from one scan event are still present in thecollision cell when a second SRM transition is takingplace. This leads to signal artifacts in the next transi-tion’s chromatogram. This can be especially problem-atic when different SRM events have the same productions formed from different precursor ions. ThermoFisher Scientific’s patented design of the orthogonal col-lision cell used on the TSQ Quantum product line elim-inates cross-talk. Figure 3a shows the absence ofcross-talk between two different SRM transitions, pir-imicarb and linuron. Both yield a product ion of 182,and no artifacts are seen in either chromatogram, evenwhen magnified 100-1000 times. The same effect isshown in Figure 3b for diflubenzuron and flufenoxuronfor a common product ion of 158 for dwell times of 20msec.

Administrator

Text Box


of 6

1110 12 13 14 15 16 17 18 19 20Time (min) Time (min)

RT : 14. 7

17. 0

10 11 12 13 14 15 16 17 18 19 20

RT : 14. 7

?

100

95

90

85

80

75

70

65

60

55

50

45

40

35

30

25

20

15

10

5

0

100

95

90

85

80

75

70

65

60

55

50

45

40

35

30

25

20

15

10

5

0

Rela

tive

Abun

danc

e

Rela

tive

Abun

danc

e

SRM Mode H-SRM Mode

Figure 2a: Comparison of SRM mode and H-SRM mode for the analysis of the fungicide Iprodione

1 3 1 4 1 5 1 6 1 70

1 0

2 0

3 0

4 0

5 0

6 0

7 0

8 0

9 0

1 0 0R T : 1 5 .6 4

1 3 1 4 1 5 1 6 1 70

1 0

2 0

3 0

4 0

5 0

6 0

7 0

8 0

9 0

1 0 0R T : 1 5 .6 1

Rela

tive

Abun

danc

e

Rela

tive

Abun

danc

e

SRM Mode H-SRM Mode


Figure 2b: Comparison of SRM mode and H-SRM mode for the analysis of the fungicide Bitertanol

Administrator

Text Box


of 6

13 14 15 16 17 18

RT : 15. 218. 0

13 14 15 16 17 18 1919

RT : 15. 3

SRM Mode H-SRM Mode

100

95

90

85

80

75

70

65

60

55

50

45

40

35

30

25

20

15

10

5

0

Rela

tive

Abun

danc

e

100

95

90

85

80

75

70

65

60

55

50

45

40

35

30

25

20

15

10

5

0

Rela

tive

Abun

danc

e


Figure 2c: Comparison of SRM mode and H-SRM mode for the analysis of the herbicide Etobenzanid

Pirim icarb

SRM 239 182

Linuron

SRM 249 182

8 10 12 14 16

11. 49

13. 4 13. 6 15. 37. 69 9. 5

x 10

T im e ( m in )

13. 26

13. 4 15. 0

10. 9

10. 38. 93 11. 8

x 50

No cross-t alk

10

8

6

4

2

0

Relat

ive A

bund

ance

10

8

6

4

2

0

Relat

ive A

bund

ance

8 10 12 14 16

Figure 3a: No cross-talk is observed for the SRM transitions of primicarb andlinuron

Dif lubenzuron

SRM 311 158

Flufenoxuron

SRM 489 158

10 12 14 16 18 200

20

40

60

80

100

Rela

tive

Abun

danc

e

14. 84

17. 24 17. 52

13. 9412. 7910. 56

x 100

T im e10 12 14 16 18 20

0

20

40

60

80

100

Rela

tive

Abun

danc

e

17. 43

17. 9015. 7415. 5312. 9111. 08

x 1000

No cross t alk

Figure 3b: No cross-talk is observed for the MRM transitions of diflubenzuronand flufenoxuron

Administrator

Text Box


Conclusions

An H-SRM LC-MS/MS method to monitor 35 pesticideresidues was developed using the TSQ Quantum Discovery.All 35 pesticide residues were quantitated in 18 minutes.Using H-SRM, interferences from the sample matrix back-ground were substantially reduced, leading to improvedLOQs. Similarly, no cross-talk issues were detected forany of the tested analytes.

Compared with traditional single pesticide analysismethods, the sample preparation procedures are usuallysimplified in multi-pesticide analysis methods. This meansmore interference from the sample matrix may be presentmaking H-SRM the technique of choice for improvingdetection limits.

AN62487_E 11/07S










www.thermo.com



Administrator

Text Box


Zero Cross-talk on the TSQ Quantum Dipankar Ghosh, Mark Churchill, Eric Genin, and Alan Schoen; Thermo Fisher Scientific, San Jose, CA


Key Words

• TSQ Quantum

• Surveyor HPLC

• AQUASIL

• Environmentalsamples

• Quantitation

• SRM

Introduction

In assays involving closely eluting multi-target analytes,Selective Reaction Monitoring (SRM) is the mostcommonly used mass spectrometry technique for per-forming quantitative assays on triple quadrupole systems.Examples include DMPK assays, monitoring for drugs ofabuse in urine extracts, and pesticide residues monitoringin environmental samples.

The SRM experiment consists of three distinct events.First, precursor ions of a specific mass to charge (m/z)ratio are transmitted through the first quadrupole, whileall ions of different m/z values are filtered out. Secondly,the selected ions collide with a neutral gas present in thesecond quadrupole (collision cell) where they undergocollision induced decomposition (CID) reactions. Finally,product ions of specific m/z values are transmittedthrough the third quadrupole, after which they aredetected.

When performing LC-MS/MS analyses of multi com-ponent mixtures by SRM, it is often necessary to carryout such assays using very short dwell times of 10–20 mil-liseconds. In certain instances, especially in environmentalmonitoring of pesticides, it is possible to get a largenumber of parent compounds of the same chemical classeluting very close to each other in a short chromato-graphic timescale. Oftentimes, these compounds also giverise to exactly the same product ions which are used tomonitor the SRM transitions. For example, the com-pounds Triadimenol (mwt 295 amu), Tebuconazole(mwt 307 amu), Cyproconazole (mwt 291 amu), andHexaconazol (mwt 314 amu) have different precursorions, but all give rise to the same product ion at m/z 70.

One potential scenario in such assays is that whenSRM dwell times and inter scan times are very small, andsample concentrations are high, storage of product ionscan take place in the collision cell. In other words,fragment ions from the one transition can still be in thecollision cell when the next SRM transition is monitored.This can result in ‘Cross talk,’ which is used to describethe phenomenon when the fragment ions from one SRMtransition are scanned out during another transition. Evena 0.01% cross-talk effect can result in false positives beingobserved in the resulting quantitative data.

It is a well known fact that Thermo Scientific triplequadrupole systems, ranging from the earlier TSQ70 seriesto the current TSQ Quantum range, have never sufferedfrom any cross-talk issues due to the advanced design ofthe collision cell. The purpose of this application note isto demonstrate the absence of cross-talk on the TSQQuantum using examples from the analyses of pesticides.

Goal

To illustrate the absence of cross-talk in a screeningassay for 88 different pesticides with closely elutingmulti-component compounds belonging to the samechemical classes.


Chemicals and Reagents

Water, methanol and acetic acid were HPLC grade andpurchased from J T Baker Chemicals, France.

Samples

Pesticides (3,4,5-Trimehacarb, 3-hydroxy-carbofuran,5-hydroxy-clethodim-sulfon, Acephate, Aldicarb,Aldicarb-sulfoxid, Aldoxycarb, Amidosulfuron, Atrazin,Azoxystrobin, Bendiocarb, Bensulfuron-methyl,Butocarboxim, Butocarboxim-sulfoxid, Butoxycarboxim,Carbaryl, Carbendazim, Carbofuran, Chlorosulfuron,Cinosulfuron, Clethodim, Clethodim-imin-sulfon,Clethodim-imin-sulfoxid, Clethodim-sulfon, Clethodim-sulfoxid, Cyprodinil, Daminozoid, Demeton-S-methyl-sulfon, Desmedipham, Diflubenzuron, Dimethoat, Diuron,Ethiofencarb, Ethiofencarb-sulfon, Ethiofencarb-sulfoxid,Fenhaxamid, Fenoxycarb, Fenpropimorph, Fentin-hydroxide, Flazasulfuron, Flufenoxuron, Flurathiocarb,Fluzifop-P-butyl, Haloxyfop-ethoxyethyl, Haloxyfop-methyl, Imazalil, Imidacloparid, Indoxacarb, Iprovalicarb,Isoproturon, Isoxaflutole, Linuron, Metalaxyl,Metamitron, Methamidophos, Methiocarb, Methomyl,Metolachlor, Metsulfuron-methyl, Monocrotophos,Nicosulfuron, Omethoat, Oxidemeton-methyl,Phenmedipham, Pirimicarb, Promecarb, Propamocarb,Propoxur, Prosulfuron, Pymetrozin, Pyridate, PyridateXX,

Administrator

Text Box


of 6

Pyrimethanil, Quinmerac, Quizalofop-ethyl, Rimsulfuron,Spiroxamine, Tebuconazol, Tebufenzoid, Thiabenzadol,Thiacloprid, Thiodicarb, Thiofanox, Thiophanat-methyl,Triasulfuron, Triflumuron, Triflusulfuron-methyl,Vamidothion) were purchased from Sigma unless other-wise noted. A standard solution containing the above 88pesticides was prepared at 50 pg/mL in methanol.

Sample Analysis

HPLC analysis was performed on the Thermo ScientificSurveyor HPLC system, using a 50×2.1 mm ThermoScientific AQUASIL C18 column. Mobile phase A waswater/methanol 80/20 (v/v) and mobile phase B wasmethanol/water 90/10 (v/v)–both contained 0.05% aceticacid. Solvent was pumped at 200 mL/min and analyteseluted using a linear gradient of 100% A to 100% B over11 minutes, holding at 100% B for 12 minutes, thenreturning to 100% A in 2 minutes. MS analyses wascarried out using a Thermo Scientific TSQ QuantumDiscovery mass spectrometer.

Mass SpectrometryInstrument: TSQ Quantum DiscoverySource: ESIIon polarity: PositiveSpray voltage: 3.5 kVSheath/Auxiliary gas: NitrogenSheath gas pressure: 50Auxiliary gas pressure: 15Ion transfer capillary temperature: 350°C Scan type: SRMCID conditions: Ar at 1.5 mTorr

MS Instrument Method

To accommodate the analysis of a large number of com-ponents over a short time range, the acquisition time wasdivided into two segments, each containing three scanevents. Allowing for analyte overlap between the timesegments, a total of 59 SRM transitions were performedin segment one and 56 SRM transitions in segment two,with dwell times of 20 msec for each transition.


Figure 1 shows the LC-MS/MS chromatogram generatedfrom the pesticide mix eluting over a chromatographictime scale of 16 minutes.


A number of compounds from the list of pesticidesbelonging to the same class are highlighted in Tables 1, 2and 3. Although they have different parent molecularweights, they belong to the same, or similar, chemical classand have a common basic carbon skeleton. This results inthem giving the same structural product ions (i.e., iden-tical mass product ions), which are being monitored forthe SRM transition.

This was demonstrated during the pesticide screeningassay by extracting three examples of compound classescontaining different precursor ions but all generating thesame product ion mass.

Figure 1 : LC-MS/MS chromatogram of 88 pesticides at 50 pg/µL

Administrator

Text Box


of 6

Example 1: Triasulfuron, Metsulfuron-methyl andChlorosulfuron. These compounds have different pre-cursor ions but all generate a product ion at m/z 167(see Table 1 and Figures 1a and 1b). The chromatogramsshow the transitions for these compounds, with noevidence of cross-talk.

Retention Predicted productCompound Time (min) Parent ion ion fragment

Triasulfuron 7.62

Metsulfuron-methyl 8.07

Chlorosulfuron 8.45

SO

O NH

O

N

NN

OO

NH

Cl

m/z 402

O

N

N N

O

NH

m/z 167

S

O

O

O OO N N

N

O

NHNH

m/z 382

O N

N

N

O

NH

m/z 167

Cl

S

O

O

N

N

N O

O

NH

NH

m/z 358

O N

N

N

O

NH

m/z 167

Table 1: Triasulfuron, Metsulfuron-methyl and Chlorosulfuron

Time (min)6.5 7.0 7.5 8.0 8.5 9.0 9.5

100

80

60

40

20

0

100

80

60

40

20

0

100

80

60

40

20

0

7.62

8.07

8.45

Triasulfuronm/z 402 >167

Metsulfuron-methylm/z 382 >167

Chlorosulfuronm/z 358 >167

Time (min)6.5

0.0

1.0

0.8

0.6

0.4

0.2

0.0

1.0

0.8

0.6

0.4

0.2

0.0

1.0

0.8

0.6

0.4

0.2

7.0 7.5 8.0 8.5 9.0 9.5

7.62

8.07

8.45

Triasulfuronm/z 402 >167

Metsulfuron-methylm/z 382 >167

Chlorosulfuronm/z 358 >167

No cross-talk

No cross-talk

Figure 1a: Zero cross-talk was observed when the pesticides triasulfuron,metsulfuron-methyl and chlorosulfuron were detected. Peak height shownfull scale. Arrows indicate position of potential cross-talk

Figure 1b: Zero cross-talk (indicated by red arrows) was observed whenthe pesticides triasulfuron, metsulfuron-methyl and chlorosulfuron weredetected. Baseline magnified x100

Administrator

Text Box


of 6

Example 2: Isoproturon and Diuron. These com-pounds have different precursor ions but both generatea product ion at m/z 72 (see Table 2 and Figures 2a and2b). The chromatograms show the transitions for thesecompounds, with no evidence of cross-talk.


Isoproturon 9.22

Diuron 9.75

NH

N

O

m/z 207

N

O

m/z 72

Cl

NH

N

O

Cl

m/z 233

N

O

m/z 72

Table 2 : Isoproturon and Diuron

9.0 9.5 10.0 10.5 11.0 11.5Time (min)

0

20

40

60

80

1000

20

40

60

80

1009.22

9.75

Isoproturonm/z 207 >72

Diuronm/z 233 >72

9.0 9.5 10.0 10.5 11.0 11.5Time (min)

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

9.22

9.75

Isoproturonm/z 207 >72

Diuronm/z 233 >72

No cross-talk

Figure 2a : Zero cross-talk was observed when the pesticides Isoproturonand Diuron were detected. Peak height shown full scale. Arrow indicatesposition of potential cross-talk

Figure 2b : Zero cross-talk (indicated by red arrow) was observed when thepesticides Isoproturon and Diuron were detected. Baseline magnified x100

Administrator

Text Box


of 6

Example 3: Nicosulfuron and Linuron. These com-pounds have different precursor ions but both generatea product ion at m/z 182 (see Table 3 and Figures 3a and3b). The chromatograms show the transitions for thesecompounds, with no evidence of cross-talk.


Nicosulfuron 9.59

Linuron 10.26

N

ON

S

O

O N

N

O

O

O

NH NH

m/z 411

O

N

N

O

O

NH

m/z 182

NH

O

NO

Cl

Cl

m/z 249

NH •

ClOH

N

m/z 182

Table 3 : Nicosulfuron and Linuron

Time (min)9.0 9.5 10.0 10.5 11.0

0

20

40

60

80

1000

20

40

60

80

1009.59

10.26

Nicosulfuronm/z 411 >182

Linuronm/z 249 >182

Time (min)9.0 9.5 10.0 10.5 11.0

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.09.59

10.26

Nicosulfuronm/z 411 >182

Linuronm/z 249 >182

No cross-talk

Figure 3a: Zero cross-talk was observed when the pesticides Nicosulfuronand Linuron were detected. Peak height shown full scale. Arrow indicatesposition of potential cross-talk

Figure 3b: Zero cross-talk (indicated by red arrow) was observed when thepesticides Nicosulfuron and Linuron were detected. Baseline magnified x100

Administrator

Text Box


Why is there Zero Cross-talk on the TSQ Quantum?

The reason for not observing any cross-talk on the TSQQuantum can be ascribed to the way the collision cell(Q2) is operated. The ions are actively ejected radially byselecting Mathieu a and q operational parameters suchthat all ions of all masses become unstable. This instanta-neously dumps all ions from the collision cell. The offsetvoltages are also operated to keep the cell empty while Q1slews to the next precursor mass. Once Q1 is set, the cell(Q2) parameters are selected to accept and fragment theprecursor ions for the next transition. As this procedureis not dependent upon ion residence times, each SRMtransition becomes a discreet experiment, with no memoryof the one before, thus totally eliminating any possibilityof cross-talk.

It should be noted that for SRM and MRM assays,this technique is superior to clearing a collision cell usingaxial ejection with a supplementary field which is slowerand less efficient. Indeed, axial techniques applied on thistime scale always have some residual cross-talk.

Conclusions

An LC-MS/MS screening assay to monitor 88 pesticidesusing minimal LC separation was developed using theTSQ Quantum Discovery. It was possible to detect allcomponents within a chromatographic timescale of 16minutes by performing 88 SRM transitions. Even withdwell times of only 20 milliseconds no cross-talk inter-fered with the analysis. The absence of cross-talk wasdemonstrated by extracting and comparing the ion chro-matogams of similar classes of compounds with differentprecursor ion masses but with the same product ion mass.

References

U.S. Environmental Protection Agency website at www.epa.gov:US Environmental Protection Agency Code of Federal Regulations 40.Chapter I, Subchapter E, Part 180 details the tolerances and exemptionsfrom tolerances for pesticide chemicals in food.

Pesticide Analytical Manual Volume 1, Sections 605-606 (describes MSapplications and benefits). Transmittal No. 94-1 (1/94), Form FDA 2905a(6/92). Available on the FDA website at www.cfsan.fda.gov. Chapter 3describes multi-class multi-residue methods, while Chapter 4 providesselective multi-residue methods.

AN62646_E 01/08S


Legal Notices©2008 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific Inc. and its subsidiaries. This informationis presented as an example of the capabilities of Thermo Fisher Scientific Inc. products. It is not intended to encourage use of these products in any mannersthat might infringe the intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are available in all coun-tries. Please consult your local sales representative for details.

View additional Thermo Scientific LC/MS application notes at: www.thermo.com/appnotesThermo Fisher Scientific,San Jose, CA USA is ISO Certified.








www.thermo.com

Administrator

Text Box


t

PesticidesPesticides, Mycotoxins, Plant toxins

30163: High Resolution and Precise Mass Accuracy: A Perfect Combination for Food and Feed Analysis in Complex Matrices

High Resolution and Precise Mass Accuracy: A Perfect Combination for Food and FeedAnalysis in Complex MatricesMarkus Kellmann, Andreas Wieghaus, Helmut Muenster, Thermo Fisher Scientific, Bremen, GermanyLester Taylor, Dipankar Ghosh, Thermo Fisher Scientific, San Jose, CA, USA

Overview

Purpose: To demonstrate the analytical advantages of using highresolution (> 40,000) for the accurate screening of pesticides in complex matrices using a new benchtopThermo Scientific Orbitrap detector.

Methods:Use of ultra-high pressure liquid chromatography (U-HPLC) coupled with the Thermo Scientific Exactivemass spectrometer operating in high resolution mode. TheExactive™ detector is based on Thermo Scientific Orbitraptechnology.

Results: The combination of high resolution (15,000 – 50,000)accurate mass is required for the detections of pesticidesand mycotoxins.

IntroductionScreening of pesticides, mycotoxins and veterinary drugs isof great importance in regulated environments such asfood and animal feed analysis. Traditionally these type of experimentshave been carried out using triple quadru-pole instruments. This approach has certain limitations:

• no post acquisition re-interrogation of data

• limited number of compounds per analysis

• cannot screen unidentified unknowns

Because of these limitations, there is currently a trendtowards full scan MS experiments in residue analysis.Current screening approaches are performed using highperformance ToF instruments, with mass accuracies of < 5ppm and resolutions of about 15,000, coupled to

Ultra High Performance Liquid Chromatography (U-HPLC).

In complex sample matrices (e.g. food, feed, hair,honey) this limited resolution leads to inaccurate massmeasurements caused by unresolved background matrixinterferences. In this work we show a full scan screeningapproach using a novel single stage Orbitrap massspectrometer coupled to U-HPLC, capable of providinghigh mass accuracy at resolutions of up to 100,000.

Additionally we will discuss two aspects of theanalysis which also greatly benefit from very highresolution:

• resolving co-eluting, isobaric target compounds

• elemental composition determination

MethodsThe Exactive, a new non-hybrid single-stage Orbitrap™

mass spectrometer coupled to the Thermo ScientificAccela U-HPLC chromatograph was used to evaluate ahighly complex mixture of 116 pesticides, mycotoxins andplant toxins in different concentrations. A 12 min gradientwas applied to a Thermo Scientific Hypersil GOLD RP 50x 2 mm 1.9 µm particle packed column withwater/acetonitrile eluents. The method developed wasevaluated with respect to sensitivity, selectivity andlinearity in standard solutions and extracts from animalfeed. Mass measurements were performed at differentresolution settings (R = 15,000 and R = 50,000) to enablecomparisons to data acquired by ToF instruments and todemonstrate the advantage of ultra high resolution.Orbitrap detection was carried out using automaticcontrol of the number of ions entering the detector (AGC, target value = 106).

Key Words

• Exactive

• Food Analysis

• Pesticide Screening

• High Resolution MS


Figure 1: Schematic of the new Orbitrap benchtop mass spectrometer, including HCD collision cell for ”All Ion Fragmentation”.

Administrator

Text Box


Results

Resolution of Isobaric Pesticides In cases where isobaric compounds co-elute, erroneousmass accuracy and elemental composition assignment willoccur if the resolving power of the mass spectrometer isinsufficient to separate these compounds. Figure 2 showstwo pesticides Thiamethoxam (C8H10ClN5O3S) andParathion (C10H14NO5PS), which have protonatedmolecular ions (MH+) at 292.02656 and 292.04031,respectively. A resolution higher than 40,000 is needed toresolve the protontated molecular ion of these twocompounds completely. This is a pre-requisite for analysisof low concentration compounds in the presence of higherabundant ones. The example in Figure 2 shows anapproximate 1:3 mixture of both pesticides measured and simulated.

Influence on Elemental Composition DeterminationA limited resolution of 15,000 results in two majorlimitations. Firstly, the detection of unresolved doubletsmay result in significant mass errors which are outside thecharacteristic accuracy specification of the Exactiveinstrument. As a consequence, at lower resolution settingsthe mass windows for elemental compositiondetermination have to be increased, resulting in muchlarger number of elemental composition proposals for theunknown or targeted compounds. This can be seen for theexample (Figure 3) of Pirimicarb at m/z 239.1503. Due tothe presense of an isobaric interference the peak at 239shows a mass error of 6.5 ppm. At a resolution of 15,000the underlying interference causes an apparent shift tohigher mass, whereas at higher resolution (here 80,000)the doublet is clearly resolved and the mass accuracy iswell within instrument specifications.

Figure 2: Mass chromatogram of two isobaric pesticides measured at a resolution of 15,000 (left) and 50,000 (right). Superimposed is thesimulated mass trace at each resolution setting.

239.00 239.05 239.10 239.15 239.20 239.25m/z

m/z

0

20

40

60

80

100

Rela

tive

Abun

danc

e

239.15181C11H19O2N4

C11H19O2N4

Pirimicarb

Pirimicarb

6.50 ppm

239.00 239.05 239.10 239.15 239.20 239.250

20

40

60

80

100

Rela

tive

Abun

danc

e

239.15033

0.32 ppm

R=15,000

R=80,000 Resolved interference

Figure 3: Improved mass accuracy simply by increasing the resolution in order to resolve the doublet. The peak at 15,000 resolution splits up into 2 by increasing the resolution to 80,000.

Administrator

Text Box


In order to limit the number of candidate elementalcompositions to a single confident assignment, a sophisti-cated software algorithm is used. It takes into account thepeak height and mass accuracy of the monoisotopic peakand its isotopes. However, in order to function correctly,all of the accurate mass values for the isotopic peaks mustbe within specified limits. The absence of interferencepeaks can only be assured by use of high resolution, (ascan be seen in Figure 4). Here the fungicide Azoxystrobinis shown at resolutions of 15,000 and 80,000. Themedium resolution spectrum shows very good massaccuracy for the monoisotopic peak, but gives unusuallyhigh mass errors for the A+1 and A+2 ions. This is due toan interference at m/z 405.1452, which cannot be resolvedat medium resolution. Whereas the high resolutionspectrum shows exellent mass accuracy for all threemeasured isotopes. Determining elemental compositionsusing data acquired at ~15,000 resolution will result inmisleading or incorrect data. Only sufficient high resolution allows the determination of the accurate massof the complete molecular ion cluster and therefore allowsautomated assignment of an elemental formula with a high degree of confidence.

Analyzing highly complex samples such as extractsfrom food or animal feed, and the screening of regulatedsubstances including pesticides, mycotoxins and veterinarydrugs is a major analytical challenge for mass spectro-metry. On one hand the methodology must have a highintra scan dynamic range in order to detect lowconcentrated compounds in presence of high abundantmatrix ions, on the other hand high selectivity and highsensitivity is needed to avoid false positive, or even worse,false negative results. In our procedure we analyzed anextract from horse feed as an example of extremelycomplex matrix, spiked with a mixture of 116 pesticidesand mycotoxins. A dilution series ranging from 2 to 250ppb (for each compound) was measured in duplicates attwo different resolution settings. In addition a 100 ppbsample of the same mixture was analyzed at a resolutionof 50,000 in order to determine the maximum number of detectable substances for this method.

Figure 4: The importance of resolution for the molecular ion AND its isotopes. At high resolution the complete molecular ion cluster is correctly detected.

Administrator

Text Box


LC-MS analysis of the extracted spiked samplesshowed the presence of 95 out of 116 compounds at 100 ppb in matrix. Figure 5 shows the overlaid ionchromatograms for all 116 compounds (3 ppm window)at 50 ppb (in matrix). The number of recovered pesticidesin different concentrations is shown in Figure 5. The dataillustrates that a greater number of detected compounds(higher sensitivity) with an extraction window of 3 ppm athigher resolution setting. This is exemplified in Figure 6,where extracted ion chromatograms of Sulcotrion at 50 ppb for R = 15,000 and R = 50,000 are shown. Thehigher resolved spectrum displays two peaks, of which thesmaller one is Sulcotrion.

The lower resolved peak masks the pesticide signalcompletely. The only indication for the presence ofSulcotrion is a slightly broader peak or shoulder and a mass shift of the interfering ion towards higher masses.This would lead to a false negative result, if the analysiswas only performed at a resolution of 15,000, and is themajor reason, why the number of identified componentsin the case of R = 15,000 decreases disproportionately tothe measurements at higher resolving power (diagram,Figure 5).

Figure 5: Overlaid extracted ion chromatograms from a mixture of 116 pesticides and mycotoxins ata 100 ppb level. Extraction was done with 3 ppm mass window. The inset chart shows the numberof detected compounds at different concentrations (in matrix) at two different resolution settings.

Figure 6: Expanded view of the pesticide mixture at different resolution settings (top: 15,000 and bottom: 50,000). PesticideSulcotrion (m/z 328.02475) is masked under background ions at a resolution of 15,000 but is easily detected at 50,000 resolution(see also mass chromatogram inset).

Administrator

Text Box


For this reason, a dilution series was measured at higher resolution settings. One example (Meta-benthiazuron) is shown in Figure 7. It demonstratesexcellent linearity and sensitivity down to 2 ppb level (2 ng/mL).

Figure 7: Quantitation curve for Metabenthiazuron ranging from 2 to 250 ppb. The quantified peak for each concentration level demonstrates the high qualitydata even at the lowest level.

Conclusions • New benchtop Orbitrap mass spectrometer

demonstrates superior mass resolving power comparedto that obtained using TOF instruments.

• High resolving power (up to 100,000) provides precisemass accuracy for complex sample analysis

• High resolving power provides excellent sensitivity,linearity and selectivity in multi-residue screening of complex matrices.

• Fast scan speed (10 Hz) are fully compatible with theuse of U-HPLC fast chromatography methods

For the analysis of very complex samples it isadvantageous to select the appropriate scan speed andresolution in order to avoid unresolved isobariccompounds (matrix ions from analyte ions) and still allowunambiguous detection of low abundant species.

AcknowledgementsWe would like to thank Paul Zomer and Hans Mol fromthe RIKILT Institute for Food Safety in Wageningen, The Netherlands, for providing the samples.

Administrator

Text Box


AN30163_E 07/08C



View additional Thermo Scientific LC/MS application notes at: www.thermo.com/appnotesThermo Fisher Scientific(Bremen) GmbH is certifiedDIN EN ISO 9001:2000













Laboratory Solutions Backed by Worldwide Service and Support In addition to these







www.thermo.com

Administrator

Text Box


Index – BY COMPOUND CLASS

LC-MS Applications for Food, Beverage, and Water Testing Index-1


Index

BY COMPOUND CLASS

Acrylamide Quantitation of Acrylamide in Food Samples on the TSQ Quantum Discovery

by LC/APCI-MS/MS

Antibiotics Analysis of (Fluoro)quinolones in Honey with Online Sample Extraction and

LC-MS/MS

Analysis of Sulfonamides in River Water using EQuan, an Online Concentration Analysis System

Determination of Sulfonamide Antibiotics in Wastewater by Liquid Chromatography–Tandem Mass Spectrometry

Determination of Trace Level Nitrofuran Metabolites in Crawfish Meat by Electrospray LC-MS/MS on the TSQ Quantum Discovery MAX

Highly Selective Detection and Identification of Nitrofurans Metabolites in Honey using LC-MS/MS

LC-MS/MS Analysis of Malachite Green, Leucomalachite Green, Ciprofloxacin, and Tetracycline in Food Samples using a TurboFlow Method

Multi-class Antibiotic Screening of Honey Using Online Extraction with LC-MS/MS

On-line Enrichment HTLC/MS/MS Assay for Multiple Classes of Antibiotics in Environmental Water Sources

Simple and Rapid Analysis of Chloramphenicol in Milk by LC-MS/MS

Antimicrobials; Antibiotics LC/MS/MS Analysis of Anti-Infectives In Raw and Treated Sewage

Flavanoids Metabolomic Analysis of Green and Black Tea Extracts Using an LTQ

Orbitrap XL Hybrid Linear Ion Trap Mass Spectrometer

Flavonoids Direct Analysis of Red Wine Using Ultra-Fast Chromatography and High

Resolution Mass Spectrometry

Index – BY COMPOUND CLASS



Haloacetic Acids Analysis of Haloacetic Acids in Drinking Water by IC-MS/MS

Melamine; cyanuric acid Analysis of Melamine and Cyanuric Acid in Food Matrices by LC-MS/MS

Mycotoxins Analysis of Mycotoxins in Various Cattle Forages and Food Matrices with the

TSQ Quantum Discovery MAX

Pesticides Analysis of Haloacetic Acids in Drinking Water by IC-MS/MS

Analysis of Regulated Pesticides in Drinking Water Using Accela and EQuan

Analysis of Triazine Pesticides in Drinking Water Using LC-MS/MS (EPA Method 536.0)

Determination of Different Classes of Pesticide Residues in Processed Fruits and Vegetables by LC-MS Using the TSQ Quantum Ultra According to EU Directive 91/414 EEC

LC-MS/MS Analysis of Herbicides in Drinking Water at Femtogram Levels Using 20 mL EQuan Direct Injection Techniques

LC-MS/MS Determination of Malachite Green and Leucomalachite Green in Fish Products

Multi-residue Analysis of Pesticides in Food using GC/MS/MS with the TSQ Quantum GC

Quantitation-Enhanced Data-Dependent (QED) Scanning of Drinking Water Samples Using EQuan for Pesticide Analysis on a Triple Stage Quadrupole

Simultaneous Detection of 88 Pesticides on the TSQ Quantum Discovery using a Novel LC/MS/MS Method

Testing LC-MS System Robustness with Automated Sample Cleanup Using Red Wine as a Matrix

UHPLC Separation of Triazine Herbicides at Elevated Temperature

Utility of H-SRM to Reduce Matrix Interference in Food Residue Analysis of Pesticides by LC/MS/MS using the TSQ Quantum Discovery

Zero Cross-talk on the TSQ Quantum

Pesticides, Mycotoxins, Plant toxins High Resolution and Precise Mass Accuracy: A Perfect Combination for Food

and Feed Analysis in Complex Matrices

Index – BY MATRIX



Pharmaceuticals The Thermo Scientific Exactive Benchtop LC/MS Orbitrap Mass

Spectrometer

Pharmaceuticals, Personal Care Products, and Pesticides Detection of Pharmaceuticals, Personal Care Products, and Pesticides in

Water Resources by APCI-LC-MS/MS

Phenolic Pollutants Analyzing Phenolic Pollutants in Water Using U-HPLC

Toxins Analysis of Microcystins from Blue-green Algae Using the TSQ Quantum

Ultra LC-MS/MS System

BY MATRIX

Catfish Analysis of Melamine and Cyanuric Acid in Food Matrices by LC-MS/MS

Cattle forages Analysis of Mycotoxins in Various Cattle Forages and Food Matrices with the


High Resolution and Precise Mass Accuracy: A Perfect Combination for Food and Feed Analysis in Complex Matrices

Crawfish Determination of Trace Level Nitrofuran Metabolites in Crawfish Meat by

Electrospray LC-MS/MS on the TSQ Quantum Discovery MAX

Crop extracts Multi-residue Analysis of Pesticides in Food using GC/MS/MS with the TSQ

Quantum GC



Eel LC-MS/MS Determination of Malachite Green and Leucomalachite Green in

Fish Products

Index – BY MATRIX



Honey Analysis of (Fluoro)quinolones in Honey with Online Sample Extraction and

LC-MS/MS


Highly Selective Detection and Identification of Nitrofurans Metabolites in Honey using LC-MS/MS

Milk Simple and Rapid Analysis of Chloramphenicol in Milk by LC-MS/MS

Potato chips; breakfast cereals Quantitation of Acrylamide in Food Samples on the TSQ Quantum Discovery

by LC/APCI-MS/MS

Red wine Direct Analysis of Red Wine Using Ultra-Fast Chromatography and High



Shrimp, fish, pork LC-MS/MS Analysis of Malachite Green, Leucomalachite Green,

Ciprofloxacin, and Tetracycline in Food Samples using a TurboFlow Method

Tea Metabolomic Analysis of Green and Black Tea Extracts Using an LTQ


Water (drinking) Analysis of Haloacetic Acids in Drinking Water by IC-MS/MS

Analysis of Regulated Pesticides in Drinking Water Using Accela and EQuan




Index – BY PRODUCT



Water (environmental) Analysis of Microcystins from Blue-green Algae Using the TSQ Quantum

Ultra LC-MS/MS System


Analyzing Phenolic Pollutants in Water Using U-HPLC

Detection of Pharmaceuticals, Personal Care Products, and Pesticides in Water Resources by APCI-LC-MS/MS



Water (lab) Simultaneous Detection of 88 Pesticides on the TSQ Quantum Discovery

using a Novel LC/MS/MS Method

The Thermo Scientific Exactive Benchtop LC/MS Orbitrap Mass Spectrometer


Water (sewage) Determination of Sulfonamide Antibiotics in Wastewater by Liquid

Chromatography– Tandem Mass Spectrometry

LC/MS/MS Analysis of Anti-Infectives In Raw and Treated Sewage

BY PRODUCT

Thermo Scientific Accela U-HPLC system Analysis of Regulated Pesticides in Drinking Water Using Accela and EQuan


Analyzing Phenolic Pollutants in Water Using U-HPLC

Direct Analysis of Red Wine Using Ultra-Fast Chromatography and High Resolution Mass Spectrometry

Metabolomic Analysis of Green and Black Tea Extracts Using an LTQ Orbitrap XL Hybrid Linear Ion Trap Mass Spectrometer






Thermo Scientific Aria TLX-1 Analysis of (Fluoro)quinolones in Honey with Online Sample Extraction and

LC-MS/MS



Thermo Scientific Exactive Direct Analysis of Red Wine Using Ultra-Fast Chromatography and High


High Resolution and Precise Mass Accuracy: A Perfect Combination for Food and Feed Analysis in Complex Matrices

The Thermo Scientific Exactive Benchtop LC/MS Orbitrap Mass Spectrometer

Thermo Scientific LTQ Orbitrap XL Metabolomic Analysis of Green and Black Tea Extracts Using an LTQ


Thermo Scientific SEIVE Direct Analysis of Red Wine Using Ultra-Fast Chromatography and High


Thermo Scientific Surveyor HPLC System Analysis of Mycotoxins in Various Cattle Forages and Food Matrices with the




Determination of Sulfonamide Antibiotics in Wastewater by Liquid Chromatography– Tandem Mass Spectrometry









Thermo Scientific Surveyor Plus HPLC system Analyzing Phenolic Pollutants in Water Using U-HPLC

Thermo Scientific TSQ Quantum Analysis of Regulated Pesticides in Drinking Water Using Accela and EQuan




Thermo Scientific TSQ Quantum Access Analysis of Haloacetic Acids in Drinking Water by IC-MS/MS





Thermo Scientific TSQ Quantum Discovery Highly Selective Detection and Identification of Nitrofurans Metabolites in

Honey using LC-MS/MS

Quantitation of Acrylamide in Food Samples on the TSQ Quantum Discovery by LC/APCI-MS/MS

Simultaneous Detection of 88 Pesticides on the TSQ Quantum Discovery using a Novel LC/MS/MS Method


Thermo Scientific TSQ Quantum Discovery MAX Analysis of Mycotoxins in Various Cattle Forages and Food Matrices with the







Thermo Scientific TSQ Quantum GC Multi-residue Analysis of Pesticides in Food using GC/MS/MS with the TSQ

Quantum GC

Thermo Scientific TSQ Quantum Ultra Analysis of (Fluoro)quinolones in Honey with Online Sample Extraction and

LC-MS/MS

Analysis of Melamine and Cyanuric Acid in Food Matrices by LC-MS/MS

Analysis of Microcystins from Blue-green Algae Using the TSQ Quantum Ultra LC-MS/MS System



Determination of Sulfonamide Antibiotics in Wastewater by Liquid Chromatography–Tandem Mass Spectrometry





EFS Applications

Documents

Transcript of EFS Applications