1. PCB 101 Spokane River Forum May 25, 2011
Dave Hope,
Pacific Rim Laboratories Inc

2. Outline Pacific Rim Laboratories Inc.
PCB – what is it and why are we still interested in it
Toxicity
How do we analyze it – Analytical Methods
What does Not Detected mean?
Limitations to PCB analysis
Closing thoughts

3. Who Am I? Analytical Chemist I am not PChem – ACPBC
President, Canadian Council of Independent Laboratories
Owner, Lab Director Pacific Rim Laboratories
I am not
Toxicologist
Consultant

4. Pacific Rim Laboratories Inc.
Small ultra-trace organic laboratory specializing in the analysis of Persistent Organic Pollutants by High Resolution Mass Spectrometry
Located in Surrey, BC (suburb of Vancouver)
Accredited by WDOE, NELAC and SCC for the analysis of PCB, PCDD/F and PBDE.
Work in the ppt, ppq and sub-ppq range

5. How small is small? $1 bill is approx. 100 cm2
Area of Canada is 10M km2
Finding a given dollar bill anywhere in Canada is 1 ppq (10-15)

6. Polychlorinated Biphenyls (PCB)
Manufactured until early 1970’s (1980’s in Eastern Europe)
Total production M Tonnes
Aroclor 1016, 1242, 1254, 1260
Kanaclor
Byproduct in the manufacture of other chemicals
Copper sulphate from China contaminated with PCB126, 169 and 77.

7. Chemical Structure
209 possible congeners – mono thru decachlorobiphenyl
Twelve are considered dioxin like and have been assigned Toxic Equivalency Factors (TEF)
MonoCB 3 congeners
DiCB 12 congeners
TriCB 24 congeners
TetraCB 42 (2) congeners
PentaCB 46 (5) congeners
HexaCB 42 (4) congeners
HeptaCB 24 (1) congeners
OctaCB 12 congeners
NonaCB 3 congeners
DecaCB 1 congeners

8. PCBs on ECD – Mix of Aroclor 1242, 1254, 1260

9. Global Production 48% produced in USA
50% of remaining prduction imported to USA
97% in Northern Hemi

11. Applications of PCB Closed systems
Insulation and/or cooling fluid in transformers (48%)
Dielectric fluid in capacitors (21%)
Switches
Partially open systems
Heat transfer fluids
Hydraulic fluid in lifting equipment, trucks and high pressure pumps
Vacuum Pumps
Voltage Regulators
Liquid Filled Electrical Cables
Liquid Filled Circuit Breakers
Open systems (21%)
Paints
Lubricating fluid in oils and grease
Water-repellent impregnating agent and fire retardant for wood, paper, fabric and leather
Laminating agent in paper production
Additive in glues, sealants and corrosion protection coatings
Carrier for insecticides
Polymerisation catalyst support for petrochemicals
Immersion oils for microscopy
Pesticide Formulation
Cable coatings/casings

13. Stockholm Convention
In 2001 the Stockholm convention was adopted, restricting the manufacturing and use of twelve Persistent Organic Pollutants – POPs
PCDD/F, PCB
Pesticides – Aldrin, Dieldrin, Endrin, DDT, Heptachlor, Hexachlorobenzene, Mirex, Chlordane, Toxaphene
Ratified in May 2004
Currently 127 countries have ratified, excluding USA
August 2010 – 9 additional chemicals included
Hexachlorocyclohexane (α, β, γ (Lindane)), PBDE (commercial Penta and Octa), hexabromobiphenyl, chlordecone, pentachlorobenzene, PFOS/PFOA
Endosulphan added last month

14. PCB Toxicity
World Health Organization (WHO) defines twelve toxic congeners
Due to positioning of chlorines, the molecule is unable to rotate and forms structure similar to dioxin
2,3,7,8-TCDD considered to have a toxicity of 1, all others are relative to TCDD
Toxic Equivalency Factors (TEF)
PCB126 (0.1), PCB169 (0.03), PCB077 (0.0001), PCB081 (0.0003)
PCB105, 114, 118, 123, 156, 157, 167, 189 ( )
PCB TEQ = ∑Cx x TEFx for all PCB congeners

15. Aroclor Composition (Wt %)
Toxicity of Aroclors
Congener
Chlorine
Aroclor Composition (Wt %)
Number
Substitution
Frame et al. (1996) Data
Pattern (IUPAC)
A1016
A1242
A1254a
A1254g
A1260 77
3,3',4,4'
0.31
0.2
0.03 81
3,4,4',5
0.01
105
2,3,3',4,4'
trace
0.47
7.37
2.99
0.22
114
2,3,4,4',5
0.04
0.5
0.18
118
2,3',4,4',5
0.66
13.59
7.35
0.48
123
2,3',4,4',5'
0.32
0.15
126
3,3',4,4',5
0.02
156
2,3,3',4,4',5
1.13
0.82
0.52
157
2,3,3',4,4',5'
0.3
0.19
167
2,3',4,4',5,5'
0.35
0.27
169
3,3',4,4',5,5'
189
2,3,3',4,4',5,5'
0.1
ppm
Toxicity
0.70 27 14
0.46

16. Regulatory Levels – Food Max intake – 2 pg TEQ/kg bw/day
European Food Regulations (pg WHO-TEQ/g fat)
PCDD/F (&PCB)
Fish (fresh wt): 4 (8)
Pork: 1 (1.5)
Poultry 2 (4)
Beef / mutton: 3 (4.5)
Liver 6 (12)
Milk: (6)
Eggs: 3 (6)
Vegetable oil: (1.5)
Mix Animal fat: 2 (3)
Fish oil: 2 (10)
European Feed Regulations
Feed*: pg WHO-TEQ/g
Pet foods: 2.25 pg WHO-TEQ/g
Animal fat: 2 pg WHO-TEQ/g
Minerals: 1 pg WHO-TEQ/g
Fish oil: 6 pg WHO-TEQ/g
Fish meal: 1.25 pg WHO-TEQ/g
* - based on 12% moisture content

17. Food levels Eggs – 2.1-38 ug/kg fat Meat – 50 – 1960 ng/kg fat
0.11 – 2.86 ng TEQ/kg fat
Meat – 50 – 1960 ng/kg fat
0.001 – ng TEQ/kg fat
Fish – 87 ug/kg
0.1 – 2.4 ng TEQ/kg

18. Fish sample – Cl4-Cl7 70 52
74 66 60
118
110
105
138
153
156
128
182/187

19. Fish sample – Cl7-Cl10
199
203
194

20. PCB Analysis in Blood PCBs in blood 200-800 ug /kg lipid
tetra-PCBs
penta-PCBs
hexa-PCBs 20

21. Analytical Methods Good – Better - Best
GC/ECD Methods
EPA 8081, 8082, 608, 508
GC/MS Methods
EPA 8270
GC/HRMS Methods
EPA 1668

22. Analytes Good – Better - Best
Total PCB – Aroclor equivalents
Homolog Totals
10 levels of chlorination
Congener specific (209)

23. Sample Extraction & Clean-up
Water – liquid/liquid extraction
Soil – soxhlet
Tissue – acid digestion followed by liquid/liquid extraction
Clean-up
Wash with conc. Sulfuric acid
Column on acidic silica gel
Column on basic alumina

24. Quantitation Methods Good – Better - Best
External standard
compare with calibration curve
Requires known final volume and amount injected
Internal Standard
Add standard just before analysis
Relative response factors determined from calibration curve
With area of IS peak and RRF, can calculate concentration
Isotope dilution
Stable isotopes added before extraction used to quantify results
Corrects for loses during work-up

25. EPA 8082 Can be used to identify Aroclors or congeners
19 congeners – MonoCB – NonaCB
PCB1, 15, 18, 31, 44, 52, 66, 87, 101, 110, 138, 141, 151, 153, 170, 180, 183, 187, 206
IS Method: One internal standard and one surrogate standard

26. Pattern Recognition
Must match the pattern in the sample with one of 7 Aroclor standards
Use 3-5 peaks in the standard to quantify sample (External Std)
Detection limits 0.05 ug/L
Problems with pattern recognition in weathered samples

27. GC/MS Methods EPA 8270 Homolog Total Methods Generic catch-all method
Uses GC/MS for ultimate identification, but causes a greatly reduced sensitivity (increased detection limit)
Detection limits 10 ug/L (Aroclor equivalents)
Homolog Total Methods
Uses GC/MS in SIM for greater sensitivity and selectivity
Internal Standard for each level of chlorination
PCB defined as having peaks in QM and RM channels and being within 15% of theoretical mass ratio
Quantifies on congener basis but not identified
Detection limits ug/L on a congener basis

28. EPA Method 1668
HRGC-HRMS congener specific method (first written 1999, revised in November 2008, June 2010)
Uses 27 internal standards (13C12-labeled congeners)
All twelve dioxin-like PCBs
First and last eluters in each level of chlorination (LOC)
Three 13C12-clean-up standards and five 13C12-recovery standards
Five ( ng/mL) or six point ( ng/mL) calibration for dlPCBs and LOC PCBs
Single point calibration for all remaining PCBs
Two methods of quantification
Isotope dilution (for 27 congeners with labeled standards)
Internal standard – everything else

29. HRMS analysis - Issues Coelutions for TeCB, PeCB and HxCB
M-Cl gives significant peak, therefore cannot have coelutions with significant peaks of higher LOC
Each LOC requires four ions
PeCB, HxCB, HpCB and OcCB overlap
Requires 16 ion channels
Need to maximize sensitivity, therefore typical collection rates of 1 hz (no hz that can be found with ECDs or MSDs)

30. GC Columns
1668 is based on SPB-octyl column, or dual column with DB-1. Has been met with mixed reviews.
Resolves 159 peaks (180 with DB-1)
30/42 TeCB
29/46 PeCB
31/42 HxCB
PCB156 and 157 coelute
DB-5ms is another popular column
Performance specs similar to Octyl (165 peaks) however separates 156 & 157

31. New column – SGE HT8 PCB 8% Phenyl (equiv.) Polycarborane-siloxane
60 m column, 0.25 mm i.d,
Initial Temp 90 C for 1 min.
Ramp 20 C/min -> 190 C
Ramp2 1.5 C/min. -> 250 C
Ramp3 4.0 C/min. -> 300 C
Final Temp 300 C for 4.2 min.
Thermo DFS HRMS with dual Thermo Trace GCs, split/splitless injection

33. Results 209 resolvable by ECD @ 100 hz
189 resolvable peak tops by HRMS
37/42 TeCB
37/46 PeCB
40/42 HxCB

35. What is the Detection Limit?
Standard methods for MDL determination
7-10 blank samples spiked at low level with analytes of interest
MDL = 3 * SD (approximately)
Better precision leads to low MDLs
Our MDLs are 1-13 pg/L
Isotope dilution allows for lower theoretical MDLs due to lower SD
Many isotope dilution methods call for the reporting of Instrument Detection Limits based on 2.5 x SN
Can give IDLs <1 pg/L

36. Analyte
PCB #
Blank n t
Average
Stdev (s)
RSD
Recovery
MDL (ts)
pg/L
2-MoCB 1
3.80 9
2.896
12.06
2.00
17%
165%
6.0
2,4'-DiCB 8
7.89
21.15
2.40
11%
265%
7.5
2,2',6-TriCB 19
8.10
0.62 8%
162%
1.8
2,2',5-TriCB 18
9.86
3.00
30%
197%
2,4,4'-TriCB 28
10.54
17.63
4.13
23%
142% 12
2,2',5,5'-TeCB 52
4.93
1.94
39%
99%
5.8
3,4,4',5-TeCB 81
5.68
0.45
114%
1.3
3,3',4,4'-TeCB 77
6.54
0.25 4%
131%
0.75
2,2',4,5,5'-PeCB
101
2.19
9.03
1.81
20%
137%
5.4
2,2',4,4',5-PeCB 99
1.38
6.22
1.22
97%
3.7
2,3,3',4',6-PeCB
110
6.17
11.01
2.07
19%
6.2
2,3',4,4',5'-PeCB
123
5.48
0.47 9%
110%
1.4
2,3',4,4',5-PeCB
118
3.33
9.36
0.99
121%
3.0
2,3,4,4',5-PeCB
114
5.61
0.37 7%
112%
1.1
2,3,3',4,4'-PeCB
105
3.10
7.17
0.50
81%
1.5
3,3',4,4',5-PeCB
126
4.99
100%
2,2',3,4',5',6-HxCB
149
9.76
20.08
4.11
206% 13
2,2',3,4,5,5'-HxCB
141
4.71
10%
94%
2,2',3,3',4,5-HxCB
138
1.02
10.00
14%
180%
4.1
2,3',4,4',5,5'-HxCB
167
5.28
0.56
106%
1.7
2,3,3',4,4',5-HxCB
156
5.59
0.42
2,3,3',4,4',5'-HxCB
157
5.47
0.46
109%
3,3',4,4',5,5'-HxCB
169
4.35
87%
2,2',3,3',4,4',6-HpCB
171
9.60
1.30
13%
192%
3.9
2,3,3',4,4',5,5'-HpCB
189
5.54
0.41
111%
1.2
2,2',3,3',5,5',6,6'-OcCB
202
0.60
2,2',3,3',4,5',6,6'-OcCB
201
3.66
0.34
73%
1.0
2,2',3,3',4,4',5,5',6-NoCB
206
5.63
0.64
113%
1.9
2,2',3,3',4,4',5,5',6,6'-DeCB
209
6.66
1.14
133%
3.4

37. Values >IDL are real Values between IDL and MDL???
IUPAC Name
PCB #
CAS #
Conc.
IDL
Flag
pg/L
2-MoCB
PCB-001
9.9
0.8 J
3-MoCB
PCB-002
5.6
4-MoCB
PCB-003
3.2
2,6-DiCB
PCB-010
3.7 UJ
2,2'-DiCB
PCB-004
30.2
2,5-DiCB
PCB-009
2,4-DiCB
PCB-007
2,3'-DiCB
PCB-006
2,3/2,4'-DiCB
PCB-005/008
7.2
3,5-DiCB
PCB-014
3,3'-DiCB
PCB-011
44.4
3,4/3,4'-DiCB
PCB-012/013
4,4'-DiCB
PCB-015
4.3
2,2',6-TrCB
PCB-019
0.9
2,4,6-TrCB
PCB-030
2,2',5-TrCB
PCB-018
12.6
2,2',4-TrCB
PCB-017
5.4
2,3,6-TrCB
PCB-024
2,3',6-TrCB
PCB-027
2,4',6-TrCB
PCB-032
2,2',3-TrCB
PCB-016
4.4
2,3,5-TrCB
PCB-023
2,3',5'-TrCB
PCB-034
2,4,5-TrCB
PCB-029
2,3',5-TrCB
PCB-026
2,3',4-TrCB
PCB-025
2,4',5-TrCB
PCB-031
8.6
2,4,4'-TrCB
PCB-028
6.8
2,3,4-TrCB
PCB-021
PCB-020/033
4.7
2,3,4'-TrCB
PCB-022
3,3',5-TrCB
PCB-036
3,4',5-TrCB
PCB-039
3,4,5-TrCB
PCB-038
3,3',4-TrCB
PCB-035
3,4,4'-TrCB
PCB-037
Values >IDL are real
Values between IDL and MDL???

38. What is Total PCB?
Sounds simple, add up individual results for all 209 congeners
How do you handle different detection limits? Co-elutions? Low level concentrations?
Should data be reported below EQL?

39. PCB # A B C D E
Blank
Homologs
pg/L
Monochlorobiphenyls
10.8 10
12.3
10.7 11
Dichlorobiphenyls
19.6
74.6
50.5 52
14.4
Trichlorobiphenyls 63
12.6
11.3
Tetrachlorobiphenyls
101
10.4
Pentachlorobiphenyls
152
23.2
15.5
27.6
29.9
Hexachlorobiphenyls
69.2
Heptachlorobiphenyls
Octachlorobiphenyls
Nonachlorobiphenyls
Decachlorobiphenyl
Total PCB
415
110
88.7
102
25.4

40. PCB # A B C D E
Blank
Homologs
pg/L
Monochlorobiphenyls
14.5
18.7
18.5
20.1
17.6
9.3
Dichlorobiphenyls
19.6
86.1
50.5 52
14.4
Trichlorobiphenyls
88.2
42.5
31.5 36
34.7
Tetrachlorobiphenyls
132.8
29.5
33.6
37.7
24.4
Pentachlorobiphenyls
173.4
40.2 24
27.6
15.5
29.9
Hexachlorobiphenyls
119.3
9.2
4.6
5.6
1.8
Heptachlorobiphenyls
30.2
2.3
Octachlorobiphenyls
11.6
2.5
1.4
Nonachlorobiphenyls
Decachlorobiphenyl
4.4
Total PCB
594
226.2
175.1
178
115.9 41
10.8 10
12.3
10.7 11
74.6 63
12.6
11.3
101
10.4
152
23.2
69.2
415
110
88.7
102
25.4

41. What is Blank? PCB is omnipresent
Almost impossible to achieve a complete blank in the lab
Every sample we analyze has detectable PCB levels.
In order to reduce blank levels in the lab
Use carbon filtered water
Disposable glassware (not always possible)
Keep food samples away!
<20 pg/L per congener is excellent

42. What is achievable?
Less than 20 labs in North America can run EPA 1668
No analytical confidence at 3 pg/L for Total PCB
Reductions will be hard to measure
Are they lab caused or real?
Focus on Toxic PCBs rather than 209 congeners

43. dave@pacificrimlabs.com www.pacificrimlabs.com
Thank you!