State-of-the-art in POPs Analysis: Outcomes of the

State-of-the-art in POPs Analysis: Outcomes of the DIFFERENCE and

DIAC projects

Pim LeonardsNetherlands Institute for Fisheries Research (RIVO)

Dioxins in Food and Feed-Reference Methods and New Certified Reference Materials

“DIFFERENCE”

Dioxin Analysis by Comprehensive Multi-Dimensional Gas Chromatography (GCxGC)

“DIAC”COMPETITIVE AND SUSTAINABLE GROWTH

(GROWTH) PROGRAMME

Background and Societal Needs

Belgian chicken incident 1999Need for dioxin analysis capacityNeed for cheap and reliable screening and confirmatory methodsNew EU MRLs, 1 July 2002

Objectives DIAC

Optimisation of GCxGC-ECD system for dioxin analysisSelection of ‘best’ modulatorOptimisation of quantification comparison with HRMSTest of alternative detection method: ToF-MSSimplification of extraction and clean-up

WP 1, task 2Optimisation of GCxGC

WP 1, task 1Mini-workshop

WP 2, task 1Quantification standardsolution/ cleaned sample

WP 2, task 2Quantification real-life

samples GCxGC vs. GC-HRMS

WP 3, tasks 1, 2Simplification extraction/

clean-up methods

WP 4 Alternative MS detection

method

WP 5Dissemination of results

Workshop for dioxin users

Final report

DIAC Workplan

DIFFERENCE ObjectivesSelection of relevant food and feed matricesPreparation of candidate CRMsFeasibility of certificationOptimisation of bio-analytical and chemical methods for dioxin analysisValidation and standardisationOptimisation of extraction and clean-upStandardised protocols for use in Europe

WP 3. Development& optimisation methods

WP 5. Extraction & clean-up

MEA SFE

PLE

CALUX

GCXGC

Ah-PCR

LR-MS/MS

Interlaboratory tests& optimisation methods

WP 1. Preparation materials

WP 2. Homogeneity & stability test

WP 6. Method validation& standardisation

WP 7. final report

WP 4. Interlaboratory studies(feasibility on certification)

DIFFERENCE Workplan

Partners

PLE

CALUX

GCXGC LR-MS/MS

WHO Dioxins, Furans and dioxin-like PCBs

PCDDs PCDFs Dioxin-like PCBs (+IUPAC nos.)

2,3,7,8-TCDD 2,3,7,8-TCDF 3,3',4,4'-TCB (77) 1,2,3,7,8-PcCDD 1,2,3,7,8-PcCDF 3,4,4',5-TCB (81) 1,2,3,4,7,8-HxCDD 2,3,4,7,8-PcCDF 3,3',4,4',5 -PeCB (126) 1,2,3,6,7,8-HxCDD 1,2,3,4,7,8-HxCDF 3,3',4,4',5,5'-HxCB (169)1,2,3,7,8,9-HxCDD 1,2,3,6,7,8-HxCDF 2,3,3',4,4'-PeCB (105)1,2,3,4,6,7,8-HpCDD 2,3,4,6,7,8-HxCDF 2,3,4,4',5-PeCB (114) OCDD 1,2,3,7,8,9-HxCDF 2,3',4,4',5-PeCB (118) 1,2,3,4,6,7,8-HpCDF 2',3,4,4',5-PeCB (123) 1,2,3,4,7,8,9-HpCDF 2,3,3'4,4',5-HxCB (156) OCDF 2,3,3',4,4',5'-HxCB (157) 2,3',4,4',5,5'-HxCB (167) 2,3,3',4,4',5,5'-HpCB (189)

EU Requirements for Dioxin and dl- PCB Analysis

Screening Methods

Confirmatory Methods

False Negative Rate

<1%

Trueness -20 to +20%

CV <30% <15%

Quality Criteria (2002/69/EC)

r&RHigh accuracy (trueness and precision)

Interferences PCN, PCB,

PCDE

High selectivity (specificity)

Up to 1pgTEQ

High sensitivity and low limits of detection

LOQ at 1pgTEQ/g fat

LOQ (confirmatory method): range 1/5 level of interest

1 – 8 pg TEQPerformance of a method

EU: sensitivity requirements for food (pg diox./g fat)

Veg. oil

0 6

Pork

PoultryFish oil

RuminantsMilk

Eggs

Liver

Fish: 4 pg dioxins/g product

GCxGC studies

RIVO, The NetherlandsJacob de Boer, Peter Korytár, Pim Leonards, Stefan van Leeuwen

Umeå University, SwedenConny Danielsson, Peter Haglund, Mikael Harju, Karin Wiberg

Free University, The Netherlands Udo Brinkman, Maria Kristenson, René Vreuls

University of Bordeaux, FranceHélène Budzinski, Ana Blanc

IQS, Barcelona, SpainJordi Díaz-Ferrero

GCxGC-ECDGCxGC-ToF-MSVarious first and second column combinationsComparison five “modulator” types

Principles of GCxGC

Injector Detector

Modulator

2nd column1st column

•Selection of modulator•5 different modulators:

•SWEEPER•LMCS•Quad N2(l) jet•Dual CO2 jet•Loop CO2

Selection of proper column combination

Cryogenic modulation

11stst columncolumn 22ndnd columncolumn

carrier gascarrier gas

stationary phase 1

COCO22

1. Trapping stationary phase 2

COCO22

2. Release

3. Trapping andSeparation

COCO22

Cryogenic modulator

How does GCxGC work ?


1

1

0 s 10 s



1

23

45

6

1

6

0s 10 s

2D plots

0 s 10 s43

46 min

43 46

10 s

0

1

6

0 10 s2nd dimension

2nd

dim

ensi

on

1st dimension

Cod liver: WHO-PCB separation2n

d di

men

sion

rete

ntio

n tim

e [s

]

1st dimension retention time [min]

85 90 95 100 105 110 115 120 125 130 135

6.0

5.0

4.0

3.0

2.0

1.0

0.0

206

194

207195

189

208

*

**

*202 201

170

**

*

180

*

171

157156

185

167

**

128

187

183

*

*

129

*

138,163

*141

137

*153**

105

*

*

114

149

118

151*

*

*

82

154

124

35 40 45 50 55 60 65 70 75 80

6.0

5.0

4.0

3.0

2.0

1.0

0.0

110

136

7785

87

97

*

I.S.112

**155

99101

92

56,60

84

95*

**

103

66

7061,74

**44

47

49

*

*

28,31

52,69

*

8Cl

7Cl6Cl5Cl

6Cl

3Cl 4Cl

5Cl

DB1 x HT-8

Milk: PCB fraction (DB-XLB x LC-50)

7781

123118

114

105 126

167

156157 169

189


2nddi

men

sion

rete

ntio

n tim

e [s

]

Milk: Dioxins (DB-XLB x LC-50)

4F1 4D15F1

5F2

5D1

6F1

6F2

6F3

6D1

6D2

6D3

7F1

6F4

7D1

7F2

OCDD

OCDF

2nddi

men

sion

rete

ntio

n tim

e [s

]


Improved clean-up and solvent grade


2nd

dim

ensi

on re

tent

ion

time

[s]

4F1 4D1

5F1

5F2

5D1

6F1

6F2

6F3

6D16D2

6D3

7F1

6F4

7D1

7F2

B


2nd

dim

ensi

on re

tent

ion

time

[s]

4F1 4D1

5F1

5F2

5D1

6F1

6F2

6F3

6D16D2

6D3

7F1

6F4

7D1

7F2

B4F1 4D1

5F15F2

5D1

6F1

6F2

6F3

6D1

6D2

6D3

7F1

6F4

7D1

7F2

OCDD


2nd

dim

ensi

on re

tent

ion

time

[s]

A4F1 4D1

5F15F2

5D1

6F1

6F2

6F3

6D1

6D2

6D3

7F1

6F4

7D1

7F2

OCDD


2nd

dim

ensi

on re

tent

ion

time

[s]

A

Sewage sludge with improved clean-up

4F1 4D1

5F1

5F2

5D1

6F1

6F2

6F3

6D16D2

6D3

7F1

6F4

7D1

7F2


2nd

dim

ensi

on re

tent

ion

time

[s]

Modulator comparison and column selection ■ Cryogenic modulation with CO2 had best performance

■ Most suitable column combinations:

• DB-XLB x LC50

• DB-1 x 90% cyanopropyl

• HT5 x BPX 50

■ Other phases less suitable because of:

• high background levels (bleeding of column)

• not all critical congener pairs (with different TEF values) could be separated from each other

Area

: 85.5

097

24.

936

Area

: 230

.815

25.

069

Area

: 20.4

163

25.

204

Area

: 7.74

321

24.

935

Area

: 35.2

081

25.

070

Area

: 16.0

686

25.

202

Integration and identification example: 23478-PeCDF

Sample

Standard

AccuracyD

evia

tion

from

GC

-HR

MS

Pork fat

Hake

Salmon

TunaTro

utCod

Spiked m

ilk

Spiked m

ilk 2

Fly as

hMilk

Feedingstu

ff

Feedingstu

ff 2

Sewag

e sludge

Sewag

e sludge 2

Sedim

ent

Herring oil

Spiked m

ilk

Vegeta

ble oil Eel

Compound feed

Fish oil-70%

-60%

-50%

-40%

-30%

-20%

-10%

0%

10%

20%

30%DIAC 1 DIAC 2 DIAC 3 DIFF CERT

Conclusions GCxGC-ECD

High selectivityHigh sensitivity but:

Multi clean-up/fractionation steps are needed

Integration of peaks is time-consumingvarious retention time markers in GCxGC plane

Improved software requirement

GC-LRMS/MS

GC-ITMS/MS (GCQ/Polaris)MS/MS modeElectron impact (EI)

Department of Analytical ChemistryUniversity of BarcelonaBarcelona (Spain)

UNIVERSITAT DE BARCELONA

BU Mass Spectrometry-

Dioxin LaboratoryDepartment of EcotechnologiesIIQAB-CSIC, Barcelona (Spain)

J. MalaviaM.T. GalceranF.J. Santos

M. ÁbalosE. AbadJ. Rivera

RT:28.98 - 32.31SM:9G

29.0 29.5 30.0 30.5 31.0 31.5 32.0Time (min)

0

20

40

60

80

1000

20

40

60

80

1000

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

10031.10

30.00 30.6229.98

31.08

31.44

31.92

32.0031.7731.32

31.43

NL: 7.85E2m/z= 274.4-275.4+276.4-277.4 F: + c SRM ms2 [email protected] [ 274.90-276.90] MS y030121_13




1,2,3,7,8-PeCDF

1,2,3,7,8-PeCDD

13C12-1,2,3,7,8-PeCDD

2,3,4,7,8-PeCDF

13C12-1,2,3,7,8-PeCDF 13C12-2,3,4,7,8-PeCDF

m/z 339.9 → m/z 274.9+276.9

m/z 351.9 → m/z 285.9+287.9

m/z 355.9 → m/z 290.9+292.9

m/z 367.9 → m/z 301.9+303.9

Clean Fish Extract: Dioxins

Fish oil: HRMS vs. LRMS/MS

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0C

once

ntra

tion

(pg/

g)

2,3,7,8-TCDD

1,2,3,7,8-PeCDD

1,2,3,4,7,8-HxCDD

1,2,3,6,7,8-HxCDD

1,2,3,7,8,9-HxCDD

1,2,3,4,6,7,8-HpCDD

1,2,3,4,6,7,8,9-OCDD

2,3,7,8-TCDF

1,2,3,7,8-PeCDF

2,3,4,7,8-PeCDF

1,2,3,4,7,8-HxCDF

1,2,3,6,7,8-HxCDF

2,3,4,6,7,8-HxCDF

1,2,3,7,8,9-HxCDF

1,2,3,4,6,7,8-HpCDF

1,2,3,4,7,8,9-HpCDF

1,2,3,4,6,7,8,9-OCDF

Fish oil (Herring)PCDD/Fs

GC-HRMS GC-ITMS/MS

Problems with clean-up

Hexa-mono-ortho-PCBs

Rel

ativ

eA

bund

ance

34 35 36 37 38 39 40Time (min)

0

10

20

30

40

50

60

70

80

90

1000

10

20

30

40

50

60

70

80

90

100

39.15

34 35 36 37 38 39 40Time (min)

0

10

20

30

40

50

60

70

80

90

1000

10

20

30

40

50

60

70

80

90

100

39.15

Insufficient clean-up

34 35 36 37 38 39Time (min)

0

10

20

30

40

50

60

70

80

90

1000

10

20

30

40

50

60

70

80

90

100

Rel

ativ

eA

bund

ance

36.78

38.37

34.66

38.71

36.75

34.64

38.6838.35

m/z 359.8 ? 287.8 + 289.8

m/z 371.9 ? 299.9 + 301.9

34 35 36 37 38 39Time (min)

0

10

20

30

40

50

60

70

80

90

1000

10

20

30

40

50

60

70

80

90

100

Rel

ativ

eA

bund

ance

36.78

38.37

34.66

38.71

36.75

34.64

38.6838.35

PCB 167PCB 156

PCB 157

m/z 359.8 ? 287.8 + 289.8

m/z 371.9 299.9 + 301.9

PCB 138

13C12-PCB 156 13C12-PCB 15713C12-PCB 138

13C12-PCB 167

Suitable clean-up

Conclusions GC-ITMS/MS

■ Low detection limits and high selectivity■ Appropriate clean-up and fractionation method

needed ■ Further studies needed in order to prove the

general applicability of the GC-ITMS/MS for the analysis of PCDD/Fs and dioxin-like PCBs

DR-CALUX studiesScientific Institute of Public Health, Beernaert H., Carbonnelle S., Hanot V., Hellebosch L., Roos P., Van Loco J., Van Overmeire I., Van Wouwe N. and I. Windal

Vrije Universiteit Brussel, Baeyens W., Sanctorum H. and C. Schroijen

Université de Liège, De PauwE., Eppe G. and M. Scippo

■Federal Agency for Safety of the Food Chain, Behets S., Fontaine A. and H. Vanderperren

■Flemish Institute for Technological Research KoppenG., Schoeters G. and R. Van Cleuvenbergen

■RIKILT – Institute of Food Safety, Bovee T., HoogenboomR. and W. Traag

■Xenobiotic Detection Systems,Brown D., Chu M., Clark G. and Gordon J.

DR-CALUX assay

CALUX activity

Cells respond to all compounds of the sample extract that activate the AhR (dioxin-like activity)

both the solvent and the sample contaminants

Total dioxin-like activity (Total TEQ of a sample)Separation of dioxins and PCBs from many other compounds

Accuracy of CALUX results

DIOXIN fraction PCB fractionCalux

0

2

4

6

8

10

12

milk

fish

oil

chic

ken

feed pork

chic

ken

tissu

e

clay

egg

herr

ing

0

2

4

6

8

10

12

milk

fish

oil

chic

ken

feed

pork

chic

ken

tissu

e

clay

egg

herr

ing

CaluxGC-HRMSGC-HRMS

REP ≠ TEF

DIOXIN fraction

0

2

4

6

8

10

0 1 2 3 4 5

CALUX-measWHO-TEF

CALUX-REP

PCB fraction

0

2

4

6

8

10

0 1 2 3 4 5

CALUX-measWHO-TEF

CALUX-REP

DR-Calux vs GC-HRMS in fish oil

DIOXIN fraction

y = 2,0943x + 1,6936R2 = 0,9519

0

5

10

15

20

25

0 2 4 6 8

GC-HRMS

CA

LUX

c

PCB fraction

y = 0,0658x + 0,9952R2 = 0,7287

0

1

2

3

4

5

0 10 20 30 40

GC-HRMSC

ALU

X

c

Conclusions DR-CALUX

High sensitivity Accuracy is lower Fast and cheap methodScreening method

Pressurized liquid extraction (PLE)PLE extraction: Dionex ASE 200 and ASE 300Non-selective ASE with external clean-upSelective ASE with on-line clean-up

Sulphuric acid silicaIntegrated carbon fractionation

Christoph von Holst JRCGeel Belgium

Peter Haglund, Umeå University UmeåSweden

Erland BjörklundLund UniversityLundSweden

http://www.umu.se/

http://food.jrc.it/

Selective PLE fat retainer

• H2SO4/Silica gel• Florisil• Basic alumina• Neutral alumina• Acidic alumina

Na2SO4

Fat Retainer

Filter

Matrix / Na2SO4 / Sand

SFE supportFilter

n-heptane

PCBs (+ Fat)

Björklund, Müller, von Holst, Anal. Chem. 2001, 73, 4050Sporring, Björklund, J. Chrom. A 2004, 1040, 155

Fat retainers

Ratio fat/fat retainer: 0.025 Recovery about 100% Coextracted fat: 500 mg fat, 1-3 mg fat leftColour: clear florisil and sulphuric acid silica onlyReaction with H2SO4: No reaction with sulphuricacid silica

Conclusion sulphuric acid silica preferred

Clean-up of fat using selective ASEASE300, 34mL cells, triglycerides 0,5g (n=3, s.e.m.)

0

20

40

60

80

100

120

0,100 0,100 0,075 0,075 0,050 0,050 0,025 0,025

Fat / fat retainer ratio (FFR)

Ret

aine

d fa

t (%

)

PentaneHexaneHeptane

100°C 50°C 100°C 50°C 100°C 50°C100°C 50°C

5-20g silica

Sporring, Björklund, J. Chrom. A 2004, 1040, 155

Traditional extraction/clean-up vs. PLEVegetable oil

01234567

Lab A Lab B PLE

pgTE

Q/g

Oil

Fish oil (n=6)

02468

1012

Lab A Lab B PLE

pgTE

Q/g

Oil

PCDD/F PCB Total-TEQ

Individual congeners

Vegetable Oil(pg/ g oil)Vegetable Oil

(pg/ g oil)

0.1

1

10

100

1000

10000

81 114

167

189

105

118

156

157 77 126

169

TCDDTCDF12

378F

2347

8FHxF

1HxF

2HxF

3HxF

4

0.1

1

10

100

1000

10000

81 114

167

189

105

118

156

157 77 126

169

TCDDTCDF12

378F

2347

8FHxF

1HxF

2HxF

3HxF

4

Lab A Lab B PLE

Fish Oil(pg/ g oil)

Fish Oil: Non-ortho PCBs and PCDD/Fs6 s.

5

4

3

2

1

0

9

20 30 40 50 60 min.

6

2

1

4

3

7

10

8

5

11

12

13

5525 35 4515

6 s.

5

4

3

2

1

0

HxCDD/F

20 30 40 50 60 min.

PeCDD/F

126

77

TeCDD/F

169

10

HpCDD/F

OCDD

5525 35 4515

Selective PLE with carbonC

arbon/Celite

Matrix/Na2SO4

Na2SO4

Na2SO4

Na2SO4

7.9 %

Dionex ASE200; 33 mL cell3g Fish oil mixed in Na2SO4

“Normal” PLE parametersThree consecutive extractions (fractions)

Integrated carbon fractionation of PCBs and Dioxins

Bulk PCBs

Carbon/C

elite

1. Heptane

Mono-ortho-PCBs

Carbon/C

elite

2. Heptane:DCM (1:1)

Non-ortho-PCBs and PCDD/Fs

Car

bon/

Cel

ite

3. Toluene

Elution profile integrated carbon fractionation

Bulk-PCBsMon-ortho

PCBsNon-ortho

PCBs PCDDs andPCDFs

3 .Toluene2A. DCM/heptane (1:1)

2B. Acetone/heptane (2.5:1)1. Heptane

0

20

40

60

80

100

120

%

Fat recovery (%)Fraction 1 98,7Fraction 2 0,7Fraction 3 0,1

Fish oil

Conclusions PLE

Fast methodIntegrated carbon PLE cost efficient methodLess labour intensive than traditional methodAttractive alternative for traditional dioxin method

Validation studies

VITO (Flemish Institute for Technological Research) R. Van Cleuvenbergen

RIVOS. Van Leeuwen, J. de Boer

Other partners from DIFFERENCE and participants outside DIFFERENCE

Method validation – Interlaboratory studiesStandards, quality control solution, clean fish extractVegetable oils with spikes of PCBs, PCNs, PCDE’sFish oil, milk, vegetable oil, vegetable oils with spikes, cereal based feed, chicken, vegetable feed, egg, fish, pork

Verification of calibration curvesVerification of analytical processVerification of matrix effects during quantificationRepeatability, within & between lab reproducibility (ISO 5725)Detection CapabilitySelectivityRuggedness

DIFFERENCE: (Candidate) CRMs for Dioxin Analysis

Fish Oil: z-scoresFISH OIL: dioxin TEQ (upperbound)

-2

-1

0

1

2

3

4

5

6

7

F G G G G F G G F F F C F C C C B B C J B B C J J A B J J J I B I A I A I I I A A ALab

z-sc

ore

Spiked MilkMILK: dioxin TEQ (upperbound)

-3

-2

-1

0

1

2

3

4

5

6

F F F F G F G C C G G A C C F C C G G J J J J J A J A I A I A I A I I I

LAB

z-sc

ore

MILK: PCB TEQ (upperbound)

-5

-4

-3

-2

-1

0

1

2

3

CAL

UX

CAL

UX

CAL

UX

CAL

UX

CAL

UX

CAL

UX

GC

-HR

MS

GC

-HR

MS

GC

-HR

MS

GC

-HR

MS

GC

-HR

MS

GC

-HR

MS

GC

-HR

MS

GC

-HR

MS

GC

*GC

GC

-HR

MS

GC

*GC

GC

*GC

GC

-LR

MS

GC

*GC

GC

*GC

GC

-HR

MS

GC

-LR

MS

GC

*GC

GC

-LR

MS

GC

-HR

MS

GC

-LR

MS

GC

-LR

MS

GC

-HR

MS

GC

-LR

MS

GC

-HR

MS

GC

-HR

MS

GC

-HR

MS

GC

-HR

MS

GC

-HR

MS

GC

-HR

MS

A A A A A A J J J J J J F F I F I I G I I F G I G F G G F G C C C C C C

LAB

z-sc

ore

CALUX:signal suppression due to high spike of mono-ortho CB-118?

GCxGC: overestimation due to combination LOQ and reporting upperbounds

Precision (%) for spiked vegetable oilWithin-Lab Reproducibility

0,0%

10,0%

20,0%

30,0%

40,0%

50,0%

0 4 8 12Conc (pg TEQ/g)

RSD

(%) CALUX

Within-Lab Reproducibility

0,0%

10,0%

20,0%

30,0%

40,0%

50,0%


RSD

(%) GCxGC-ECD


0,0%

10,0%

20,0%

30,0%

40,0%

50,0%


RSD

(%) GC-HRMS


0,0%

10,0%

20,0%

30,0%

40,0%

50,0%


RSD

(%)

GC-LRMS

Maximum residue levelPrecision requirement

Discussion

Technique Sens

.

Acc

ur.

Prec

is.

RemarksGC-HRMS + + + Confirmatory methodGC-LRMS + + + Potential alternative for HRMS?GCxGC-ECD +/- +/- + Improved software requiredCALUX + +/- + Screening technique

ConclusionsGCxGC-ECD (and GCxGC-ToF-MS), and GC-ion-trap MS/MS may serve as alternative (routine) methods for dioxin analysisCALUX is the alternative for times of crisis, but corrections for recovery are essentialThe use of PLE will significantly reduce the extraction and clean-up time

Acknowledgement

All DIFFERENCE and

DIAC partners

Thank you for your attention

www.dioxins.nl

Documents

State-of-the-art in POPs Analysis: Outcomes of the