Michelangelo Anastassiades

Community Reference Laboratoryfor Pesticide Residues

using Single Residue Methods

CRL-SRMSAUDIS TRAININGSTUTTGART, DEC. 2009

1 EU Reference Laboratoryfor Pesticide Residues

requiring Single Residue Methods

Training on Pesticide Residues Analysis Bangkok, 17-21 september 2012

Michelangelo Anastassiades

Matrix Effects in GC and LC







GC-Injection - Split/Splitless







By Hans Mol

GC-Injection – A Critical Step







GC-Injection – A Critical Step

By Hans Mol

Overload







The gas volume of the sample injected should be smaller than the internal volume of the liner.

If not, the liner is overloaded and backflash of the sample into various parts of the injector may occur. • Poor injection profiles• Bad peak-shapes• Adsorption onto various parts of inlet and possible carry over.

Typical liners for split/splitless injection are ~80 x 4mm = ~1ml volume

NOTE: Effective vol. available is lower as part of the liner will be filled with carrier gas.

Liner volume







boiling point densityC g/ml μl vapor/μl injection

water 100 1 1491acetonitrile 82 0.78 511acetone 56 0.79 366ethyl acetate 77 0.9 275toluene 111 0.87 254hexane 69 0.66 206isooctane 99 0.69 162GC: 30 m x 0.25 mm, 70C, 60kPa

Expansion-Volumes of Solvents

GC-Injection

MW

184158889286114







Large Volume Injections

By Hans Mol







PTV for Large Volume Injections

By Hans Mol







Requirements for robust GC analysis

• Avoid water- activates liner, retention gap, column- damages filament- has large expansion volume (upon evaporation)

Note: Water can be conveniently evaporated using PTV Azeotrop: MeCN:H2O = ~84:16- if H2O content is lower it will concentrate to 16% and then evaporate together w. MeCN; - if content is higher water will remain in the system

• Avoid non-volatile co-extractants (incl. Fat=triglycerides)- contaminate liner/retain less volatile analytes

• Avoid too high concentrations of volatile co-extractants- interfere in the chromatograms







High Selectivity in Instrumental Analysis gives more flexibility in sample preparathion

GC analysis

MIPs/Immunoaffinity

Column chromatography Solid Phase Extraction Dispersive SPESelective extraction solventGeneric extraction solvent

GC-FID

GC-ECD GC-NPD GC-FPD GC-MS (quad/IT/TOF) GCxGC-MS, GC-MS/MS, GC-hrMSGCxGC-hrMS, GC-QTOF

Combination should be fit-for-purpose

Sample preparation

Interdependence of analytical steps

By Hans Mol

Incr

easi

ng s

elec

tivity

In

crea

sing

sel

ectiv

ity







Questions:

What are the target analytes?

What is the matrix?

What is the desired LOQ?

What instrumentation is available in your laboratory?

LC-MS/MS available?

Analytes amenable to LC-MS/MS? yes no

use LC-MS/MS(+ GC for confirmatory purposes if possible)

yes no

GC analysis

Decisions Regarding Instrumental Analysis

By Hans Mol







Dealing with Matrix-Effects in GC

with the help of Analyte Protectants (APs)







WHAT ARE MATRIX-

EFFECTS ???













GC-Injector schematically

Syringe

GC-capillary

GC-linerHeater z.B.

250°C

Pneumatics

Sample injection

GC-Injection







Ratios:• Peak-Areas: ~ 1,5:1• Peak-Heights: ~ 4:1• Peak-Width (at half height): ~ 1:3

8.70 8.75 8.80 8.85 8.90 8.95 9.00 9.05 9.10 9.15 9.20 9.25 9.30 9.35 9.40 9.450

5000

10000

15000

20000

25000

30000

RT= 8,80 min

RT= 8,92 min

WITHOUT Matrix co-extractives(in pure solvent)

e.g. Calibration standard

WITH Matrix co-extractives(Strawberry-Extract)

Analyte: Atrazine ; Matrix: Strawberry

Stronger Tailing

Apex-Shift towards longer RTs!

Matrix-Induced Peak Enhancement OVERESTIMATION OF RESULTS!!







Analytes (interact with Active Sites)• Unwanted Retention/Tailing• Quasi-catalysed degradation (susceptible compounds)

„Matrix-Induced Peak Enhancement Effect“

Active Sites (on Surface of GC-Liner & Column )(Siloxanes & deposited non-volatile matrix-co-extractives)

Matrix-Components (in Excess) • Bloc active sites and protect analytes

GC-Liner

GC-Capillary







Number and type of active sitesNumber and type of active sites in the inlet and GC column in the inlet and GC column

Chemical structure of the analytesChemical structure of the analytes: : H-bondingH-bonding abilityability, , thermolability, volatilitythermolability, volatility

Analyte concentrationAnalyte concentration most pronounced at trace level most pronounced at trace level

Injection temperatureInjection temperature

Interaction timeInteraction time function of: analyte volatility (function of: analyte volatility (), inlet temperature (), inlet temperature (), flow ), flow raterate, , gas gas pressurepressure, , injection volumeinjection volume, , solvent expansion solvent expansion volumevolume, , column diamensionscolumn diamensions

Matrix type and concentrationMatrix type and concentration

Factors influencing in matrix-effects







START

FINISH

100 % 100 %

95 % 10 %

Influence of Analyte

Concentration







An additional effect… „Response Diminishment”Response Diminishment”

Accumulation of non-volatile matrix compounds on the surface of liner and a first part of GC column

Formation of new active sitesnew active sites

Increasing number of injections of matrix-containing extracts

Gradual decrease in analyte responsesand/or

increase in peak-tailing







Matrix EffectsMatrix Effects

Response Response enhancementenhancement

Response Response diminishmentdiminishment







Elimination Elimination oror CCompensation for ompensation for MMatrix atrix EEffects?ffects?

Analyte protectantsAnalyte protectants

Completely remove active sites in GC system

Completely remove matrix compounds

Standard additionMatrix-matched standardsIsotopically labeled internal standards

PracticalPractical

ELIMINATIONELIMINATION

COMPENSACOMPENSATIONTION

ImpossiblImpossible in e in

practicepractice

Impractical Impractical for routinefor routine







Compensation of matrix effects (1)

Method of standard additionsMethod of standard additionsextra effortextra effort possible inaccuracies due to :possible inaccuracies due to :

concentration dependence oconcentration dependence off matrix effects matrix effects non-linear response of detector non-linear response of detector deterioration of the system as the samples deterioration of the system as the samples

are injected are injected







Peak Ratio Analyte/ISTD

Added amount of analyte to the aliquotx

x: absolute amount of analyte in aliquot before spiking (y=0)

y-intercept= ׀x׀ Slope

Method of standard additions

Withdraw from extract multiple aliquots of same volume and spike them with increasing amounts of standard

No spiking here

Good linearity is paramount because of

extrapolation

In case of violations: Standardaddition








Use of matrix-matched standardisationUse of matrix-matched standardisation need for enough blank matrix (ideally exactly the same need for enough blank matrix (ideally exactly the same

as the samples) and its long-term storageas the samples) and its long-term storage

extra time, laboextra time, labouur, and expense for preparation of the r, and expense for preparation of the blank extracts for calibration standardsblank extracts for calibration standards

greater amount of matrix material injected onto the greater amount of matrix material injected onto the column in a sequence column in a sequence greater GC maintenance greater GC maintenance

potentially greater potential for analyte degradation in potentially greater potential for analyte degradation in the matrix solutionthe matrix solution








Isotopically labelled internal standardsIsotopically labelled internal standards

Generally not availableGenerally not available for all pesticides for all pesticides (or expensive if available) (or expensive if available)

Restriction in the use of MS techniquesRestriction in the use of MS techniques

Occupy Measurement time Occupy Measurement time (theoretical consideration) (theoretical consideration)

Obviate the need for Matrix-MatchingObviate the need for Matrix-Matching







Use of isotopically labelled ISTD in MS

MW 122.6

Formula: C5H13ClN

Chlormequat Chlormequat D4

MW 126.6

Formula C5H9D4ClN

Different mass, but exactly same behaviour during extraction, cleanup and chromatography!

CH3 N+

CH3

CH3

Cl

HH

H H

CH3 N+

CH3

CH3

Cl

DD

D D









Co-elutionCo-elution

Co-elutionCo-elution







N

O

O

SC

Cl

Cl

Cl

H

H

H H

N

O

O

SC

Cl

Cl

Cl

D

D

D D


MW 296.6

Formula C9H4Cl3NO2S

Folpet Folpet D4

MW 300.6

Formula C9D4Cl3NO2S








Degradation at• High pH• High temperatures

Phthalimide

Dicofol, Captafol, Tolylfluanid, Dichlofluanid, Pyridate

CAPTAN

e.g.:

• During sample prep. • In final extract• In GC-inlet

Similar behaviour:

FOLPET

Tetrahydrophthalimide

Degradation of Folpet and Captan







Alternative: Isotopically Labelled ISTDs

FLV Folpet D4

y = 12,78x + 0,228

R 2 = 0,9979

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0 0,005 0,01 0,015 0,02 0,025 0,03 0,035

FLV BS 138

y = 0,1576x + 0,003

R2 = 0,7960

0,002

0,004

0,006

0,008

0,01

0 0,005 0,01 0,015 0,02 0,025 0,03 0,035

Fläche pur

y = 3E+06x + 45408

R2 = 0,8318

0

20000

40000

60000

80000

100000

120000

140000

0 0,005 0,01 0,015 0,02 0,025 0,03 0,035

FLV C aptan D6

y = 71,164x + 2,312

R2 = 0,9979

0

1

2

3

4

5

6

7

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07

FLV BS 138

y = 1,0722x + 0,0252

R2 = 0,9585

0

0,02

0,04

0,06

0,08

0,1

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07

Fläche pur

y = 6E+06x + 152616

R2 = 0,9436

0

100000

200000

300000

400000

500000

600000

0 0,01 0,02 0,03 0,04 0,05 0,06 0,07

Only via Area

via ISTD (PCB 138)

via ISTD (Folpet D4)

R2 =0.9979

R2 =0.7960

R2 =0.8318

R2 =0.9979

R2 =0.9436

R2 =0.9585

Only via Area

via ISTD (PCB 138)

via ISTD (Captan D6)

Folpet in Papaya (MRL=0.01 mg/kg)

0.019 mg/kg

0.022 mg/kg

0.018 mg/kg

No

violation

0.030 mg/kg

0.033 mg/kg

0.030 mg/kg

Unacceptable R2

Should be at least 0.995

Captan in Blueberries (MRL=0.01 mg/kg)

Dicofol is the next to check using isotopically labelled ISTD

ISTDs added

to extract

Violation







THE USE OF ANALYTE PROTECTANTS







WHAT ARE ANALYTE PROTECTANTS?

?ANALYTE

PROTECTANT







PROTECTANT LORD ANALYTE

?







ANALYTE

PROTECTANT

DANGER

?







ANALYTE

PROTECTANT

?







PROTECTANT

ANALYTE

?







Dispersive SPE –Removal of Co-extractives

2,9

1,31,0

pH 5.4

pH 8.4pH 8.9

0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

No PSA PSA 25 mg/mL PSA 50 mg/mL

mg

/mL

0

1

2

3

4

5

6

7

8

9

10

pH

va

lue

Amount of Co-extractives in the extract (mg/kg)

pH of Acetonitril Extract

PSA Cleanup and effect on pH

Drawbacks: pH rises (degradation risk) Matrix-Induced Analyte

Protection reduced

Solutions:

Addition of Acids (see later)

Addition of Analyte Protectants







Analyte Protectants Principle

PSA cleanup

Addition of „Analyte

Protectants“(AP)

„ Protection“

Raw Extract

Cleaned up Extract

Cleaned-up Extract + AP

Standard + AP

Standard in Pure Solvent

Analyte Protectants help to reduce analyte Interactions with Active Sitesand thus Errors related to Matrix-Induced Peak Enhancement in GC







Extract acidification improves GC-behaviour of Folpet, Captan and Dicofol

But better results by addition of ANALYTE PROTECTANTS-MIX to extracts

Use of Analyte Protectants to improve GC-Analysis

Errors due to Matrix Effects with and without Analyte Protectants

010

2030

4050

6070

8090

100

Dic

hlo

rvos

Dic

hlo

benil

Acephate

o-P

henylp

henol

Phenm

edip

ham

Carb

ofu

ran

HC

H,

gam

ma-

Chlo

roth

alo

nil

Vin

clo

zolin

Para

thio

n-m

Carb

ary

l

Meth

iocarb

Dic

lofluanid

Para

thio

n-e

Chlo

zolin

ate

Toly

fluanid

Pro

cym

idone

Capta

n

Folp

et

Flu

benzim

ine

Ditalim

fos

Bupirim

at

Bin

apacry

l

Trifloxystr

obin

Tetr

am

eth

rin1

Dic

ofo

l

Acrinath

rin

Bitert

anol

Ipro

dio

ne

Chin

om

eth

ionat

TP

P

ME

AN

ALL

Ab

so

lute

Err

or

[%]

Against Solvent w/o APs

Against solvent with AP

(no peak in

solvent)

Overestimations if quantified against

solvent-based calibr. std

• Folpet, Captan work perfectly • Dicofol still not fully satisfactory







Analyte Protectants-Reduction of Matrix Induced Enhancement Errors

Errors eliminated if: Response in Matrix/Response in Solvent ~ 1

0,00 0,50 1,00 1,50 2,00 2,50 3,00

methamidophos

mevinphos

acephate

omethoate

vinclozolin

metalaxyl

chlorpyriphos

carbaryl

dichlofluanid

fenthion

cyprodinil

endosulfan I

imazalil

azinphos-methyl

coumaphos

Without AP With AP no AP with AP

1,0

Response in Matrix

Response in Solvent

Overestimations when using

Standards in Solvent

AP was added to both :Sample Extract and Calibration Standard (in pure Solvent)







Analyte Protectants – Examples

Various Compounds Tested for “Protective Potential”. Best Protection : Polyhydroxy-Compounds (sugars, ~derivatives)

O

OH OH

OOH

OHOH O

OH

OHOH

OH

OHOH

OH

Sorbitol

Ethylglycerol

δ-Gulonolactone

Examples:

APs typically give broadly eluting peaks For protection over a broad volatility range use AP-Mixtures as each covers a different volatility range







D-Sorbitol

Degradation products of L-Gulonic acid γ-lactone

First dimension (DB5-ms)

Seco

nd d

imen

sion

(BPX

-50)

3-Ethoxy-1,2-propandiol(end of the tail)

Elution of pesticides

GCGCGCGC/TOF-MS/TOF-MS of Analyte Protectants of Analyte Protectants

By J. Hajslova







0

100

200

300

400

500

Dichlor

vos

Mev

inpho

s

Omet

hoat

e

Pirimica

rb

Vinclo

zolin

Met

alaxy

l

Pirimip

hos-

met

hyl

Met

hioca

rb

Chlorp

yrifo

s

Tolyl

fluan

id

Procy

mido

ne

Thiab

enda

zole

Endos

ulfan

I

Imaz

alil

Parat

hion-

met

hyl

Brom

opro

pyla

te

Phosa

lone

Cyhal

othr

in-l

Deltam

ethr

in

Azoxy

strob

in

Re

lativ

e in

ten

sity

(%

)

Comparison of Generated Data

Accurate data over a broad analyte spectrum Accurate data over a broad analyte spectrum !!

3-ethoxy-1,2-propandiol (4 mg/mL)

L-gulonic acid γ-lactone (1 mg/mL)

D-glucitol (1 mg/mL)

Baby food0.015 mg/kg

Without analyte protectants With analyte protectants

1340%1340% 1450%1450%

By J. Hajslova







Strong interactions with active sites (H-Bond activity)

Similar volatility to analytes to be protected (so that protection extents during entire run)

Soluble in sample extract (not in non-polar solvents)

Not accumulating in GC-system

Not reactive with analytes (not inducing their degradation)

Minimal interference with analyte detection (small m/z)

Not deteriorating GC-column separation performance

Cheap and not hazardous

Analyte Protectants- Desirable properties







Active Site masked by AP

Active Sites

PestizidPestizidePesticides

Aktive StelleAP- Elution-Band









































































































































































55 potential APs have been individually tested

2-Deoxy-D-ribose, 4,6-O-Ethylidene-a-D-glucose, 4-Hydroxybenzoic acid, Adipic acid, Allylacetic acid, Formic acid, Succinic acid, Pyruvic acid, Brij92 (Diethylene glycol oleyl ether), Butyric acid, Citric acid, Caffeine, D-(+)-Galactose, D-(+)-Gluconic acid-δ-lactone, D-(+)-Glucose, D-(+)-Ribonic acid-γ-lactone, D-(+)-Xylose, D-Fructose, Diethylene glycol, DL-Malic acid, D-Mannite, D-Sorbitol, Acetic acid, Ethylene glycerol, Fumaric acid, Polyethylene glycols (mixture of), Gallic acid, Glutaric acid, Glycolic acid, Glycyl-glycine, Hexaethylene glycol, Caffeic acid, L-(+)-Arabinose, L-(+), Gulonic acid-γ-lactone, Lactose, L-Sorbose, Maleic acid, meso-Erythitol, Lactic acid, myo-Inositol, Oxalic acid, Pentaethylene glycol, Pimelic acid, Propyl gallate, Adonitol, Sucrose, Shikimic acid, Sorbic acid, Tetraethylene glycol, Triethylene glycol, Triglycerol, Valeric acid, Vanillic acid, Tataric acid, Xylitol







Ranking of potential APs

The ranking system used the ratio of peak height and peak area for all pesticides covered by GC methods.

Important properties of APs: many hydroxy groups (e.g. sugars) proton donation (e.g. acids) to protect

base-sensitive compounds such as captan, folpet and chlorothalonil

vapor pressure corresponding to pesticides

Ranking Analyte Protectant Score

1 D-(+)-Gluconic acid--lactone 2.06

2 4,6-O-Ethylidene--D-glucose 1.77

3 L-(+)-Arabinose 1.76

4 Triglycerol 1.72 5 D-Fructose 1.72

6 D-Sorbitol 1.72

7 Brij ' 92 1.60 8 D-(+)-Glucose 1.59

9 2-Deoxy-D-ribose 1.57

10 Adonitol 1.56 11 Xylitol 1.56

12 Shikimic acid 1.55

13 D-(+)-Xylose 1.55

14 D-(+)-Ribonic acid--lactone 1.42

15 Polyethylene glycols (mix of) 1.37

16 AP Mix 1 1.36 17 L-Sorbose 1.36

18 L-(+)-Gulonic acid--lactone 1.25

19 D-Mannite 1.24 20 Ethyl glycerol. 1.21

21 meso-Erythitol 1.08

22 Vanillic acid 1.06 23 Citric acid 1.01

24 Gallic acid 0.96

25 Caffeic acid 0.83

40 DL-Malic acid 0.25







Influence of amount of analyte protectants

Beside the type of an analyte protectant also the amount is important!

Example Malic Acid:

increasing protection the more malic acid added

Best at 0.35 mg / mL MeCN(in this experiment)

Carb

aryl

Ditalim

ph

os

Parath

ion

-m

Dico

fol

Cap

tan

Fo

lpet

Ch

loro

thalo

nil

Diclo

fluan

id

To

lyfluan

id

Clo

zolin

ate

TP

P

Pro

cymid

on

pH 4pH 4.5

pH 5pH 5.5

pH 60

1

relative respo

ns

Malic Acid







Latest mixture of analyte protectants

(E)

OH

OH

OH

O

OH

OO

OH

OH

OH

HO

AP 2-Mix

d-Gluconolactone

Shikimic acid

OH OOH

OHOH

OH

OHOH

OH

AP 1-Mix

Sorbitol

Ethyl glycerol







Reduction of matrix-effect related errors using the new AP-mixture

Use of APs to Reduce Matrix-Induced Effect Related Errors

0 10 20 30 40 50 60 70 80 90 100 110

TPP (ISTD)

Vinclozolin

Tebufenpyrad

Quinoxyfen

Pyriproxyfen

Pyrimethanil

Pyridaben

Profenofos

Procymidon

Pirimiphos-methyl

Pirimicarb

Phosmet

Pendimethalin

OPP

Myclobutanil

Mirex

Metalaxyl

Mepanipyrim

Malathion

Kresoxim-methyl

Iprodion

Fludioxonil

Fipronil

Fenarimol

Esfenvalerat

Diethofencarb

Dichloran

Cyprodinil

Chlorpyrifos

Biphenyl

Azoxystrobin

Relative Error [%](absolute deviation between signals from pure-solvent standard to signals from matrix-containing standard)

with APs without APs

Average Error- without APs= 22 %

- with APs= 7 %







Analyte protectants

When using APs, the presence of

matrix is less important.







Literature AP

AP Publications:M. Anastassiades, K. Maštovská, and S.J. Lehotay J. Chromatogr. A, 1015 (2003) 163-184

K. Maštovska, S.J. Lehotay and M. Anastassiades, Anal Chem 77 (2005) 8129

AP-Posters:M. Anastassiades, S. Lehotay, 'Reduction of Analyte Degradation and Peak Tailing During GC Injection by Addition of Protecting Agents' 4th European Pesticide Residue Workshop in Rome/Italy Awarded as best Poster

M. Anastassiades, E. Scherbaum, B. Tasdelen: ‘Investigations on the use of analyte protectants for multiresidue GC analysis‘6th European Pesticide Residue Workshop in Corfu/Greece







LC-MSLC-MS







Why do we need LC-MS if GC-MS is available?

GC requires evaporation of analytes at high temperatures!

Consequently, analytesmust be volatile at 220 – 300°Cshould not be thermo labile

GC MS is not applicable for approximately 25 – 30% of all pesticides!

Some chemical classes cause problems in GC:

• aryloxyalkanoic acids (e.g. 2,4-D)

• benzoylureas (e.g. diflubenzuron)

• carbamates (e.g.desmedipham)

• neonicotinoids (e.g. imidacloprid)

• sulfonylureas (e.g. amidosulfuron)

Slide by Lutz Alder







Comparison: sensitivity of GC-MS and LC-ESI-MS/MS

0%

20%

40%

60%

80%

100%

organo-P Carbamate organo-Cl Sulfonylurea Triazole Triazine Urea Pyrethroide Alkanoicacid

Other all

GC-MS better LC-MS/MS better equal sensitivity

L. Alder et al., Mass Spectrometry Reviews 25, 838-65 (2006)







APCI INTERFACEAPCI INTERFACE

N2

SHEATH GAS

NEBULIZER GAS

LCELUENT

HEATER500 ºC

CORONA DISCHARGENEEDLE4-6 KV N2







Atmospheric pressure Chemical ionization (APCI)

– evaporation of neutral droplets– ionization of neutral molecules induced by hot corona needle– competition for charge in gas phase

Declustering

3. Solvent/Analyte inteaction

Spraygas

Vacuum

2. Ionisation of solvent

Atmospheric pressure

Curtaingas

1. Corona-discharge

Solventl

Analyte

(~120°C)

LC







IONSPRAYIONSPRAY

ATMOSPHERE VACUUMNEBULIZER GAS

LIQUID SHEATH

STAINLESS STEEL CAPILLARY HOLDER 3-8 KV

STAINLESS STEELHEATED CAPILLARY

LC FLOW

TOANALYZER







Electrospray Ionization (ESI)

– Formation of droplets containing ions– Shrinking of droplets as solvent evaporates with time (heating

gas)– Ion release to gas phase by coulomb explosion (at high ion

density) or spontaneous evaporation from droplet surface – Competition for charges with matrix in solution– Works best with polar analytes which are either charged in

solution or easily ionized coming out of solution







LCZ

Off axis

Orthogonal

LC-MS – Spray Types







XIC of +MRM (175 pairs): 334.0/145.0 amu from Sample 12 (15740 S Ingwer R160 Ho) of DataMonitoring_2008_A_Q.wiff (Turbo Spray) Max. 4.8e5 cps.

12.0 12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 17.0 17.5 18.0 18.5 19.0Time, min

0.0

2.0e4

4.0e4

6.0e4

8.0e4

1.0e5

1.2e5

1.4e5

1.6e5

1.8e5

2.0e5

2.2e5

2.4e5

2.6e5

2.8e5

3.0e5

3.2e5

3.4e5

3.6e5

3.8e5

4.0e5

4.2e5

4.4e5

4.6e5

4.8e5

In

te

ns

ity

, c

ps

15.30

QuEChERSextract of

organic ginger

MRMs of Tebufenpyrad(m/z)

334 > 145334 > 117

Limitations of LC-MS/MSBy M. Jezussek LGL-Erlangen

XIC of +MRM (175 pairs): 334.0/145.0 amu from Sample 4 (Monitoring_2008 0,10 ug/ml Matrix: Spinat) of DataMonitoring_2008_A_Q.wiff (Turbo ... Max. 9.2e4 cps.

12.0 12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 17.0 17.5 18.0 18.5 19.0Time, min

0.0

5000.0

1.0e4

1.5e4

2.0e4

2.5e4

3.0e4

3.5e4

4.0e4

4.5e4

5.0e4

5.5e4

6.0e4

6.5e4

7.0e4

7.5e4

8.0e4

8.5e4

9.0e4

In

te

ns

ity

, c

ps

15.39

Tebufenpyrad standard sol.

0.1µg/mLMRMs (m/z)334 > 145334 > 117

XIC of +MRM (175 pairs): 334.0/145.0 amu from Sample 13 (15740 ZV S Ingwer + 1 mg/kg Tebufenpyrad R160 Ho) of DataMonitoring_2008_A_... Max. 4.4e5 cps.

12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 17.0 17.5 18.0 18.5Time, min

0.0

2.0e4

4.0e4

6.0e4

8.0e4

1.0e5

1.2e5

1.4e5

1.6e5

1.8e5

2.0e5

2.2e5

2.4e5

2.6e5

2.8e5

3.0e5

3.2e5

3.4e5

3.6e5

3.8e5

4.0e5

4.2e5

4.4e5

In

te

ns

ity

, c

ps

15.33

14.56 15.02

QuEChERSextract of

organic ginger fortified with

Tebufenpyrad

MRMs (m/z)334 > 145334 > 117

LC-MS/MS: “probably yes”

Is it Tebufenpyrad?







By M. Jezussek LGL-Erlangen

organic ginger extract+ Tebufenpyrad

Reference MS-spectrum of Tebufenpyrad

QuEChERS extract of organic ginger

Background-substracted MS-spectrum

Limitations of LC-MS/MS

GC-MS : “No”







Matrix Effects in LC-MS applicationsMatrix Effects in LC-MS applications

-







Matrix effects in LC/MS - Example

0

50

100

150

200

250

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7

2 g/ml

1 g/ml

0,5 g/ml

0,25 g/ml

0,125 g/ml

ohne Matrix

TPP Konzentration [µg/ml]

Fläche

Matrix

FIA mode, no column used

1998







Unsaturated Just saturated Over-saturated

Cross-sectional views of droplets with increasing analyte concentration

Effect of Analyte Concentration on Analyte Response in ESI

Signal saturation is due to crowding at the droplet surface. Some analyte ions (AH+) are unable to reach the surface.

AH+

AH+

A

AH+AH+

AH+

AH+

A

AH+

AH+AH+

AH+

AH+

AH+

AH+

A

AH+

AH+

AH+ AH+A

AH+AH+

A

AH+AH+

AH+

AH+ AH+

A

AAH+

AH+

A

AH+ AH+AH+

AH+

AAA

AH+

A







COMPETITION between analyte and electrolyte ions for conversion to gas-phase ions decreases analyte response.

1. Surface competition

2. Charge competition

OAc- + AH+ HOAc + A

Electrolyte concentration

Analyte response

Matrix Effects on Analyte Response in ESI

NH4+

NH4+

NH4+

NH4+OAc-

NH4+OAc-

AH+

AH+NH4+

AH+OAc-

NH4+

A

A

NH4+

NH4+NH4+

AH+

AH+

A







Matrix effects in LC-API-MS(/MS)

Ionization efficiency in the API source is influenced usually suppression

Effect takes not place during injection or elution (unlike GC)

matrix components and analytes must co-elute

Differences in column retention or deviations in the elution profile may shift position of matrix peak relative to analyte and alter the effect

Differences from commodity to commodity

Spurious matrix effect when matrix from previous injections coelutes







Compare the calibration curves 1-5-10-25-50-100-250-500 µg/l

If the Slope matrix/Slope solvent: >1 signal enhancement

=1 no matrix effect

<1 signal supression

Standard in SOLVENT; in TOMATO; in PEAR and in ORANGE matrix

Spiroxamine

y = 1085,3x + 2745,3R2 = 0,9984

y = 1270,6x + 1482,3R2 = 0,9999

0

100000

200000

300000

400000

500000

600000

700000

0 100 200 300 400 500 600

Concentration (ng/ml)

Inte

nsi

ty (

cps)

solvent

matrix

Matrix effect: - 15 %

Signal supression in pear

Fipronil

y = 9,5105x - 13,16R2 = 0,997

y = 5,9001x + 47,606R2 = 0,9915

0

1000

2000

3000

4000

5000

6000

0 100 200 300 400 500 600


Inte

nsi

ty (

cps)

matrix

solvent

Matrix effect: + 61 %

Signal enhancement in orange

Matrix Effects in LC/MS – How to detect?







Metalaxyl

y = 255,98x + 1671,8

R2 = 0,9992

y = 256,1x + 1418,6

R2 = 0,9984

0

20000

40000

60000

80000

100000

120000

140000

0 100 200 300 400 500 600


Inte

ns

ity

(c

ps

)

solvent

matrix

Matrix effect: 0 %

No matrix effect in tomato

Hexythiazox

y = 22,47x - 114,88R2 = 0,9981

y = 89,566x + 764,82R2 = 0,9957

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

0 100 200 300 400 500 600

Concentration (ng/ml)In

ten

sity

(cp

s)

solvent

matrix

Strong signal supression in orange

Matrix effect: -75 %

Matrix Effects in LC/MS - Examples







Agüera et al. J Chromatogr A 1045 (2004) 125-135

Determination of triflumizole by LC-ESI-MS (m/z 346):Extreme suppression in the case of pepper extracts








Tomato Pear Orange

9

24

109

71

19

31

5047

3

22 24

5145

8

0

20

40

60

80

100

120

<-50% -50-(-20)% -20 - 0 % 0 - 20 % 20 - 50 % >50 %

Tomate Pera Naranja

fuerte moderado suave suave moderado fuerte

supresión de la señal aumento de la señal

Signal supression Signal enhancement

strong moderate soft strongmoderatesoft

Strong matrix effect

~ 13% of compounds

TomatoPear

Orange








Intensity ratio matrix standards / standards in solvent

5 matrices x 7 levels x 100 pesticides (= 3500 data)

0%

10%

20%

30%

40%

50%

60%

70%

0-2

0%

20

-40

%

40

-60

%

60

-80

%

80

-10

0%

10

0-1

20

%

12

0-1

40

%

14

0-1

60

%

16

0-1

80

%

18

0-2

00

%

>2

00

%

ESI response in matrix / response in solvent

Perc

en

tag

e o

f to

tal

cases

Tomato

Lemon

Raisin

Wheat flour

Avocado

No problem for screeningMatrix matched standards needed for exact quantification








Effects of salts on ionization efficiency

Carbamates typical MRM transitions:

[M+H]+ [M - CH3-N=C=O]+

In the presence of Na+ salts :[M+Na]+ [Na]+

Pyrethroides Significant ionization in the presence of NH4

+ salts only







Dealing with Matrix Effects in LC

Use of isotopically labelled internal standardsExpensive but applicable for single residue methods (see SRM-presentation)

Standard additions ApproachMost used when MRL violation has to be checked

DilutionSimple way as long as instrument sensitivity is sufficientBe careful to avoid precipitations

e.g. non-polar compounds when diluting with water

Better Cleanupeffect only partly eliminated

Better Chromatographic SeparationCan be very helpful (e.g. UPLC) but not always achievable







• Cleanup• Dilution• Lowering the injection volume …only reduce Matrix effects

Matrix-matched calibration is better but not always practicable!

A good chromatographic separation remains mandatory

Important to know which analytes are significantly affected in which commodity (database can help)!

Dealing with Matrix Effects in LC







Reduction of supression by cleanup

-1 1 3

-4

215

-1

-2

400

662 1

-14

-3

10

-4-4-5

1

-1

57 6 75

-1 -1 -2-9 -10 -9

-18 -17

-9 -8-9

-50

-40

-30

-20

-10

0

10

201 2 3 4 5 6 7 8 9 10 11 12 14 15 16 17 18 19 20 22 23 24 25 26 27 28 29 30 31 32 34 34 35 36 37 38 39 40 41 42

Extract of Raisins (1 g/ mL) - Raw

Deviation on average: 5 %

-50

-40

-30

-20

-10

0

10

201 2 3 4 5 6 7 8 9 10 11 12 14 15 16 17 18 19 20 22 23 24 25 26 27 28 29 30 31 32 34 34 35 36 37 38 39 40 41 42

-1 -2-6

1

-7-2-3

0

-11

1

-7-11

10 63

-6-6 -5

0 -1

-1 -1

-1

-17

-4

-1

711 63 23 202 3332

-4

-50

-40

-30

-20

-10

0

10

201 2 3 4 5 6 7 8 9 10 11 12 14 15 16 17 18 19 20 22 23 24 25 26 27 28 29 30 31 32 34 34 35 36 37 38 39 40 41 42

Extract of Raisins (1 g/ mL) - after D-SPE


Suppression in raisins sample is low, thus cleanup has minimal effect.







Reduction of supression by cleanup

-11

7

-26

-49

-14

-67

-4

-61

-27

3

-16

-6

-17-21

-29 -34-27-22

-31-29

-4

15

-17-22

-6-10

-27

-36

-15-16

-21-25-28

-65 -44

11

-16

5

-43-44-50

-40

-30

-20

-10

0

10

201 2 3 4 5 6 7 8 9 10 11 12 14 15 16 17 18 19 20 22 23 24 25 26 27 28 29 30 31 32 34 34 35 36 37 38 39 40 41 42

Extract of Oranges (1 g/ mL) - Raw


0

-6

-15 -15

1 3

-2

-29

-5 -4-6

05 4 3 3 5 5

-3

4

-36

-10

1114 15

-64

-7-12

9

-17

-59

-21

-8-9

69 8

-11

-22-21

-50

-40

-30

-20

-10

0

10

201 2 3 4 5 6 7 8 9 10 11 12 14 15 16 17 18 19 20 22 23 24 25 26 27 28 29 30 31 32 34 34 35 36 37 38 39 40 41 42

Extract of Oranges (1 g/ mL) - after freeze-out / D-SPE


Citrus are notoriously problematic as regards matrix effects in LC-MS/MS- Effects are significantly reduced on average but still remain unacceptably high for various compounds! - Analysts should be aware of this! Dilution, Matrix-matching, Std-addition…







The use of ECHO technique

Eluent AInjectorPump

Detector

HPLC-column

43

2

Waste

Aux. pump-

Eluent A

16

5

Sam

ple

Standard

Pre column

1st injection: standard

By Lutz Alder







Eluent A

InjectorPumpDetector

HPLC-column

1

43

2

65

Eluent B

Sta

nd

ar

d

Sample

Waste

Pre column

Aux. pump-

Eluent A

The use of ECHO technique

2nd injection: sample (50 seconds later)

By Lutz Alder







Definition:

Significant matrix effects if signal difference (reduce or enhance) at least 30%

Occurrence of significant matrix effects with 58 standards in solvents

with external stds. with ECHO stds. cucumber 7 4 lemon 40 14 raisin 22 10 wheat flour 6 6

Summary: ECHO standards reduced effects only by 50%.

The use of ECHO technique - Results

Reason: suppression changes virtually from second to second







Postcolumn infusion to study matrix effects

pump A

pump B

no injection

syringe pumpwith analyte

T-piece

ESI interface

Mass spectrometer

column

0

50000

100000

150000

200000

250000

300000

350000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

time [min]

inten

sity [

cps] Suppression area







By H. Stahnke

Postcolumn infusion to study matrix effects







Comparison of matrix effects in different matrixes

Str

on

g e

ffec

tsW

eake

r ef

fect

s

By H. Stahnke







Matrix-Effects – Comparison GC vs. LC-MS

GC applications LC-MS applications (ESI-mode)

Effects take place in inlet and column Effects take place in ion source (after the chromatogr.run)

Most commonly signal enhancement Most commonly signal suppression

Virtually not influenced by chromatographic resolution Greatly influenced by chromatographic resolution

May change as sequence is progressing due to surface contamination with low volatility

components (matrix induced diminishment effect)Typically not changing as sequence is progressing

Matrix (e.g. electrolyte composition) may influence MS-spectra and ion-yield

Different from analyte to analyte

Different from matrix to matrix

Can be overcome using matrix-matched calibration

Can be overcome using standard additions approach

Can be overcome using isotopically labeled ISTDs

Can be overcome using AP approachAnalyte Protectants (AP) approach does not work

Echo-peak approach does not work Echo-peak approach does work (with limitations)

Ion-yield and MS-spectra are constant







THE HUMAN FACTOR!!

(Ensure: training possibilities, pleasant working environment)

of a successful lab is…

Don’t forget: The…

(SKILLED and MOTIVATED PEOPLE)

AΩ&







Thank you very much for your Attention

Documents

Michelangelo Anastassiades