Simultaneous Determination of PCBs, PBDEs, and Organochlorine Pesticides Using GC‐Tandem Mass Spectrometry
Sabrina Crispo SmithApril 23, 2014ECL Seminar
Background• Three classes of compounds
– Polychlorinated biphenyls (PCBs)– Polybrominated diphenyl ethers (PBDEs)– Organochlorine pesticides (OCPs)
• Persistent – Lipophilic – Bio‐accumulative
• Health Effects– Endocrine disruptors– Reproductive effects
Outline
• POP Serum Analysis Method• Method Development
– SGE HT8‐PCB– Agilent DB‐5ms– Results from Studies
• Method Updates• Future Work• Conclusions
Steps for Serum Extraction
Sample Preparation:Defrost, add surrogate spike, and
denature with formic acid
Automated Sample Extraction using Waters HLB Column
Sample Evaporation to ~1mL
Sample Clean-up with manually packed acid silica column
Matrix change and sample concentration to ~80uL
Internal Standard spike and transfer to GC vial for analysis
Analysis Methods
Biomonitoring CA• Contemporary Samples• Analysis on GC‐HRMS• Two GC columns
– PCBs/OCPs (~ 1hr/sample)– PBDEs (~ ½ hr/sample)
• Total time: ~ 1.5 hrs/sample
Federally Funded Projects• Archived Samples (1960s)
and Contemporary Samples• Analysis on GC‐ECD• Dual Column Method
– Run concurrently– PCBs/OCPs
• Total time:~ 2.5 hrs/sample
• Upgraded to GC‐tandem MS
Compounds of InterestMeasured on GC-ECD (two columns) or
GC-HRMS (one column)Measured on GC-HRMS
(one column)Organochlorine
Pesticides (OCPs)Polychlorinated
Biphenyls (PCBs)Polybrominated Diphenyl
Ethers (PBDEs)β-BHC PCB66 BDE17HCB PCB74 BDE28
Oxychlordane PCB99 BDE47Trans-nonachlor PCB101 BDE66
p,p’-DDE PCB105 BDE85o,p’-DDT PCB118 BDE99p,p’-DDT PCB138 BDE100
PCB153 BDE153PCB156 BDE154PCB170 BDE183PCB180 BDE196PCB183 BDE197PCB187 BDE201PCB194 BDE202PCB203 BDE203
BDE206BDE207BDE208BDE209
Method Development
• New Method– Transfer method from GC‐ECD to GC‐triple quadrupolemass spectrometer (GC‐QqQ)
• Goals for Method Transfer:– All analytes from GC‐ECD (PCBs/OCPs)– Shorter analysis time than GC‐ECD– Additional analytes
• PCB28, PCB167• “major” PBDEs
GC‐ECD method
• External surrogates – TCMX, PCB14, PCB65, PCB166
mV
0.00
50.00
100.00
150.00
200.00
250.00
300.00
350.00
400.00
450.00
500.00
550.00
600.00
650.00
700.00
750.00
800.00
Minutes0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00 110.00 120.00 130.00 140.00 150.00
Restek RTX‐5ms J&W DB‐XLB
GC‐ECD method10
5.69
7
4,4'
-DD
T/P
CB
138
- 106
.186
106.
996
PC
B18
7 - 1
08.2
43
108.
896
PC
B16
6/18
3 - 1
09.2
44
109.
780
mV
0.00
20.00
40.00
60.00
80.00
100.00
120.00
140.00
160.00
180.00
200.00
220.00
240.00
260.00
280.00
300.00
320.00
340.00
360.00
380.00
Minutes104.60 104.80 105.00 105.20 105.40 105.60 105.80 106.00 106.20 106.40 106.60 106.80 107.00 107.20 107.40 107.60 107.80 108.00 108.20 108.40 108.60 108.80 109.00 109.20 109.40 109.60 109.80 110.00 110.20
PCB166+
PCB183p,p’-DDT
+ PCB138
40.2
45
PC
B14
- 40
.478
41.4
64
42.2
01
b-B
HC
/PC
B30
- 42
.790
mV
0.00
50.00
100.00
150.00
200.00
250.00
300.00
350.00
400.00
450.00
500.00
Minutes40.20 40.40 40.60 40.80 41.00 41.20 41.40 41.60 41.80 42.00 42.20 42.40 42.60 42.80 43.00 43.20 43.40 43.60 43.80 44.00
β-BHC+
PCB30
GC‐QqQ
Agilent 7890A GC– Injector
• Pulsed Splitless• 250°C
– Column• Flow Rate: 1mL/min• Carrier Gas: Helium
Agilent 7000B Series– Source
• Electron Impact Extractor (EIEX)
• 70eV• 275°C
– Collision Cell• Nitrogen (1.5mL/min)• Helium (2.25mL/min)
– Q1 and Q3• 150°C• MS1 resolution: 1.2amu• MS2 resolution: 1.2amu
– Detector:• Time filter peak width: 0.7s
SGE HT8‐PCBColumn Length: 60mColumn I.D.: 0.25mmTransfer Line: 300°CRun Length: ~ 61 min
050100150200250300350
0 20 40 60 80
Tempe
rature (C
)Time (minutes)
SGE HT8‐PCBColumn Length: 60mColumn I.D.: 0.25mmTransfer Line: 300°CRun Length: ~ 61 min
050100150200250300350
0 20 40 60 80
Tempe
rature (C
)Time (minutes)
SGE HT8‐PCBColumn Length: 60mColumn I.D.: 0.25mmTransfer Line: 300°CRun Length: ~ 61 min
HT8‐PCB28 Peaks(two overlap)
050100150200250300350
0 20 40 60 80
Tempe
rature (C
)Time (minutes)
SGE HT8‐PCBColumn Length: 60mColumn I.D.: 0.25mmTransfer Line: 300°CRun Length: ~ 61 min
HT8‐PCB28 Peaks(two overlap)
R.T. Compound 1 Compound 2
15.1 PCB14 HCB
28.2 PCB101 t‐nonachlor
050100150200250300350
0 20 40 60 80
Tempe
rature (C
)Time (minutes)
MRM Transitions
• PCBs– Precursor Ion: [M+]– Product Ion: [M – 2Cl]+– Qualifier uses 37Cl loss
+ +
Loss of 2 x ClIn Collision Cell
MRM Transitions• OCPs
– Depends on compound
Precursor Product Loss of
TCMX M+ [M ‐ 1Cl]+ 1 x Cl
HCB M+ [M ‐ 2Cl]+ 2 x Cl
p,p'‐DDE M+ [M ‐ 2Cl]+ 2 x Cl
t‐nonachlor [M ‐ 1Cl]+ [M ‐ 3Cl]+ 2 x Cl
β‐BHC [M ‐ 3Cl]+ [M ‐ 4Cl]+ 1 x Cl
DDT (o,p' and p,p') [M ‐ CCl3]+ [M ‐ CCl3 ‐2Cl]+ 2 x Cl
Oxychlordane
HT8‐PCB: Pedigree Study
• Breast Cancer Study– Samples from 1960’s– N = 195
• Limits of Detection– 0.005 ‐ 0.02 ng/mL for PCBs – 0.05 ‐ 0.15 ng/mL for OCPs
• Linearity Ranges– 0.01 – 3 ng/mL for PCBs– 0.05 – 80 ng/mL for OCPs
Pedigree Study QC: % Recoveries
0
20
40
60
80
100
120
140
160
TCM
X*
PC
B14
*H
CB
β-B
HC
PC
B28
PC
B65
*O
xych
lord
ane
PC
B74
PC
B66
t-non
achl
orP
CB
101
PC
B99
p,p'
-DD
Eo,
p'-D
DT
PC
B11
8P
CB
153
PC
B10
5p,
p'-D
DT
PC
B13
8P
CB
187
PC
B16
6*P
CB
183
PC
B16
7P
CB
156
PC
B18
0P
CB
170
PC
B20
3P
CB
194
Matrix Spike NIST SRM% Recovery: 50 – 120% % Error: 3 – 20%
* External Surrogate
Detection Frequencies for BDEsBEST MIEEP (M) MIEEP (CB) Zota FA Study FOX Study
N = 110 N = 77 N = 63 N = 38 N = 25 N = 101BDE17 2% 0% 3% 0% 0% 1%BDE28 92% 36% 13% 32% 84% 97%BDE47 99% 92% 67% 74% 84% 99%BDE66 9% 4% 0% 0% 0% 6%BDE85 20% 6% 3% 5% 16% 35%BDE99 81% 61% 25% 63% 92% 90%BDE100 89% 90% 56% 74% 100% 98%BDE153 99% 90% 37% 68% 100% 100%BDE154 17% 9% 0% 0% 12% 20%BDE183 4% 4% 2% 3% 8% 5%BDE196 3% 3% 0% 0% 0% 2%BDE197 43% 36% 0% 3% 60% 62%BDE201 5% 3% 0% 3% 0% 5%BDE202 1% 0% 0% 0% 0% 1%BDE203 1% 1% 0% 0% 0% 1%BDE206 5% 4% 3% 0% 0% 12%BDE207 18% 23% 3% 0% 0% 50%BDE208 3% 9% 2% 0% 0% 8%BDE209 38% 53% 17% 0% 36% 29%
“Major” BDEsBEST MIEEP (M) MIEEP (CB) Zota FA Study FOX Study
N = 110 N = 77 N = 63 N = 38 N = 25 N = 101BDE17 2% 0% 3% 0% 0% 1%BDE28 92% 36% 13% 32% 84% 97%BDE47 99% 92% 67% 74% 84% 99%BDE66 9% 4% 0% 0% 0% 6%BDE85 20% 6% 3% 5% 16% 35%BDE99 81% 61% 25% 63% 92% 90%BDE100 89% 90% 56% 74% 100% 98%BDE153 99% 90% 37% 68% 100% 100%BDE154 17% 9% 0% 0% 12% 20%BDE183 4% 4% 2% 3% 8% 5%BDE196 3% 3% 0% 0% 0% 2%BDE197 43% 36% 0% 3% 60% 62%BDE201 5% 3% 0% 3% 0% 5%BDE202 1% 0% 0% 0% 0% 1%BDE203 1% 1% 0% 0% 0% 1%BDE206 5% 4% 3% 0% 0% 12%BDE207 18% 23% 3% 0% 0% 50%BDE208 3% 9% 2% 0% 0% 8%BDE209 38% 53% 17% 0% 36% 29%
MRM Transitions
• PBDEs– Precursor Ion: [M+]– Product Ion: [M+] – 2Br– Qualifier uses 81Br loss
Loss of 2 x BrIn Collision Cell
+ +
HT8‐PCB: PBDE ResultsHT8‐PCB34 Peaks(two overlap)
HT8‐PCB: PBDE Results
BDE 153BDE 154
BDE 99 BDE 100
BDE 47
HT8‐PCB: PBDE Results
0%
20%
40%
60%
80%
100%
120%
140%
BDE47 BDE99 BDE100 BDE153 BDE154
% R
ecov
ery
Whole Blood Spike NIST SRM % Recovery: 86 – 112% % Error: 1 – 10%
CA Childhood Leukemia Study• Case‐control study started in 1995
• Seeks to identify genetic and environmental risk factors for childhood leukemia
• Children’s Whole Blood (Case)
GOAL: To measure PCBs, PBDEs, and OCPs in 100uL of whole blood using a GC‐MS/MS with a
single injection
HT8‐PCB: PBDE Results
0%
25%
50%
75%
100%
125%
0
0.1
0.2
0.3
0.4
0.5
BDE47 BDE99 BDE100 BDE153 BDE154
Det
ectio
n Fr
eque
ncy
IDL
(pg/
uL)
New Column
SGE HT8‐PCB(60m, 0.25mm)
Transfer Line: 300°CRun Length: 75min
Agilent DB‐5ms*(30m, 0.25mm, 0.25um)Transfer Line: 280°CRun Length: 32min
* Based on Lin et al, 2013 Talanta 113: 41–48
0
100
200
300
400
0 20 40 60 80
Tempe
rature (C
)
Time (minutes)
HT8‐PCB DB‐5ms
HT8‐PCB and DB‐5ms
BDE47
BDE99
BDE100
BDE154BDE153
BDE47 BD-99
BDE100
BDE154 BDE153
HT8‐PCB
DB‐5ms
HT8‐PCB and DB‐5ms
BDE47
BDE99
BDE100
BDE154BDE153
BDE47 BD-99
BDE100
BDE154 BDE153
HT8‐PCB
DB‐5ms
RRF HT8-PCB DB-5msBDE47 0.014 0.015
BDE99 0.009 0.010
BDE100 0.005 0.014
BDE153 0.001 0.002
BDE154 0.002 0.003
HT8‐PCB vs DB‐5ms (TIC %)
HT8‐PCB34 Peaks(two overlap)
DB‐5ms32 Peaks(four overlap)
HT8‐PCB vs DB‐5ms (TIC %)
HT8‐PCB34 Peaks(two overlap)
DB‐5ms32 Peaks(four overlap)
R.T. Compound 1 Compound 2
13.45 PCB74 Oxychlordane
17.0 PCB165* o,p’‐DDT
18.9 PCB138 p,p’‐DDT
23.9 PCB194 BDE100
R.T. Compound 1 Compound 2
15.1 PCB14 HCB
28.2 PCB‐101 t‐nonachlor
Column Comparison Results
0.0
0.5
1.0
1.5
2.0
2.5
0.0 0.5 1.0 1.5 2.0 2.5
30m
DB
-5M
S on
GC
-QqQ
60m HT8-PCB on GC-QqQ
PCB118PCB138PCB153PCB180
GC‐HRMS vs GC‐QqQ Results
0.0
0.5
1.0
1.5
2.0
0.0 0.5 1.0 1.5 2.0
30m
DB
-5m
s on
GC
-MS/
MS
15m DB-5MS High Resolution GC-MS
BDE47
BDE99
BDE100
BDE153
N = 10
0.00
0.05
0.10
0.15
0.20
0.25
0.30
BDE47 BDE99 BDE100 BDE153 BDE154
IDL
(pg/
uL)
HT8-PCB DB-5ms
Instrument Detection Limits based on Column Selection
IDL = S/N ≥ 5
IDL and Method Detection Limits
0.00
0.05
0.10
0.15
0.20
0.25
0.30
BDE47 BDE99 BDE100 BDE153 BDE154
pg/u
L
HT8-PCB DB-5ms MDL using DB-5ms
MDL = 3 * σ(Blank)
IDL and MDL and Detection Frequency
0%
25%
50%
75%
100%
125%
0
0.1
0.2
0.3
0.4
0.5
BDE47 BDE99 BDE100 BDE153 BDE154
Det
ectio
n Fr
eque
ncy
IDL
(pg/
uL)
HT8-PCB DB-5ms MDL using DB-5ms
DB‐5ms QC Recoveries
0.0
20.0
40.0
60.0
80.0
100.0
120.0
140.0
160.0TC
MX
*H
CB
PC
B14
*β-
BH
CP
CB
28P
CB
65*
PC
B74
Oxy
chlo
rdan
eP
CB
66P
CB
101
PC
B99
Tran
s-…
p,p'
-DD
EP
CB
118
PC
B16
5*o,
p'-D
DT
PC
B15
3P
CB
105
p,p'
-DD
TP
CB
138
PC
B18
7P
CB
183
PC
B16
7P
CB
156
PC
B18
0B
DE
47P
CB
170
PC
B20
3B
DE
100
PC
B19
4B
DE
99B
DE
154
BD
E15
3
% R
ecov
ery
Matrix Spike NIST SRM% Recovery: 48 – 124% % Error: 4 – 17%
* External Surrogate
CCLS Santa Rosa Three Generation StudyChildren Mothers BBCC Study Mothers Daughters
N 194 50 100 492 300Vol Serum 100uL 1mL 2mL 1mL 1mL
D.F. > 50% PCB105 PCB99 PCB28 PCB28 PCB28PCB118 PCB105 PCB74 PCB66 PCB66PCB138 PCB118 PCB99 PCB74 PCB74PCB153 PCB138 PCB101 PCB99 PCB99PCB180 PCB153 PCB118 PCB101 PCB101p,p'-DDE PCB170 PCB138 PCB105 PCB118p,p'-DDT PCB180 PCB153 PCB118 PCB138
Trans-nonachlor PCB187 PCB156 PCB138 PCB153BDE47 PCB194 PCB170 PCB153 PCB156BDE99 β-BHC PCB180 PCB156 PCB167
BDE100 HCB PCB187 PCB167 PCB170BDE153 Oxychlordane PCB203 PCB170 PCB180
o,p'-DDT β-BHC PCB183 PCB183p,p'-DDE HCB PCB187 PCB187p,p'-DDT Oxychlordane PCB180 PCB194
Trans-nonachlor p,p'-DDE PCB194 PCB203BDE47 p,p'-DDT PCB203 HCBBDE99 Trans-nonachlor β-BHC Oxychlordane
BDE100 BDE47 HCB o,p'-DDTBDE153 BDE99 Oxychlordane p,p'-DDE
BDE100 o,p'-DDT p,p'-DDTBDE153 p,p'-DDE Trans-nonachlor
p,p'-DDT BDE47Trans-nonachlor BDE99
BDE100BDE153
Current Work ‐ Surrogates
0%
20%
40%
60%
80%
100%
120%
140%H
CB
β-B
HC
PC
B74
Oxy
chlo
rd…
PC
B66
PC
B10
1P
CB
99t-n
onac
hlor
p,p'
-DD
EP
CB
118
o,p'
-DD
TP
CB
153
PC
B10
5p,
p'-D
DT
PC
B13
8P
CB
187
PC
B18
3P
CB
156
PC
B18
0B
DE
47P
CB
170
PC
B20
3P
CB
194
BD
E10
0B
DE
99B
DE
154
BD
E15
3
Matrix Spike NIST SRM% Recovery: 69 – 111% % Error: 1 – 10%
GC‐HRMS vs GC‐QqQ: Samples
0
5
10
15
20
0 5 10 15 20GC
-HR
MS
(pg/
uL)
GC-QqQ Results (pg/uL)
PCBs
050
100150200250300350
0 100 200 300GC
-HR
MS
(pg/
uL)
GC-QqQ Results (pg/uL)
OCPs
0
5
10
15
20
0 5 10 15 20GC
-HR
MS
(pg/
uL)
GC-QqQ Results (pg/uL)
PBDEs
Current Work: Additional PBDEs
0%
20%
40%
60%
80%
100%
120%
140%
BDE17 BDE28 BDE85 BDE183
Matrix Spike NIST SRM% Recovery: 91 – 109% % Error: 3 – 11%
Issue To Address• Instrument detection limits for new PBDEs
– Especially BDE85 and BDE183
0.000
0.025
0.050
0.075
0.100
BDE17 BDE28 BDE47 BDE99 BDE100 BDE85 BDE154 BDE153 BDE183
Conc
(pg/uL) w
ith S/N
= 5 IDL: S/N = 5
Future work
• Address New PBDE instrument detection limits
• Implement DB‐5ms column method on GC‐HRMS– Will increase sample throughput– No high MW BDEs (e.g. BDE209)
• Process Expanded BEST study on GC‐HRMS with new method
Conclusions
• GC‐MS/MS method allows for the analysis of PCBs, PBDEs, and OCPs in a single 32 min run– No detected interferences for serum analysis
• When method is transferred to GC‐HRMS should reduce sample analysis time and increase sample throughput by ~ 3‐fold.
Acknowledgements
• Hilda Barry• Greg Yeh and Hyoung Gee Baek• Todd Whitehead, Warren Li, PraphopphatAdhatamsoontra, and Lea Pearlman
• Weihong Guo• DTSC staff