Polycyclic Aromatic Hydrocarbon

(PAH) Analysis in Fish by GC/MS

Using Agilent Bond Elut QuEChERS
dSPE Sample Preparation and a High
Efficiency DB-5ms Ultra Inert GC
Column
Application Note

Food Testing and Hydrocarbon Processing

Authors Abstract
Doris Smith and Ken Lynam This application note details a quick and effective analytical method for the determi-
Agilent Technologies, Inc. nation of low and trace level polycyclic aromatic hydrocarbons (PAHs) in fish samples
2850 Centerville Road as an alternative to procedures involving more time consuming and complex tech-
Wilmington, DE 19809-1610 niques. A simplified QuEChERS (Quick, Easy, Cheap, Effective, Rugged, and Safe)
USA method with dispersive solid phase extraction (dSPE) provided sufficient sample
matrix cleanup while preserving low level analyte detection. A pressure controlled tee
was installed post column to implement the use of automated backflush to diminish
residual sample carryover and reduce instrument cycle times.

The Agilent J&W DB-5ms Ultra Inert (UI) 20 m × 0.18 mm , 0.18 µm column effectively
resolved the 16 targeted PAHs. A GC/MS method employing selective ion monitoring
(SIM) was calibrated at the 10, 25, 50, 100, 250, 500, and 1000 ng/mL PAH levels,
yielding excellent linearity and reproducibility. The spiked levels for recovery determi-
nations were 25, 250, and 500 ng/mL. Recoveries ranged from 80% and 139% with
RSD below 6%.
Introduction The DB-5ms column has been widely used for US-EPA PAHs
and offers suitable resolution of the sixteen targeted PAHs
Oil spills can introduce massive amounts of crude oil and the investigated by this method. The high efficiency DB-5ms Ultra
myriad of components found in the oil, including Polycyclic Inert column was chosen for this application because of its
Aromatic Hydrocarbons (PAHs). PAHs are a broad class of added benefits for trace level analysis in more complex sam-
molecules that are well known, well characterized, and a ples matrixes while retaining the same selectivity [7]. A 0.18
worldwide focal point for regulations as both food contami- mm id or high efficiency GC column can also help by providing
nants and environmental pollutants [1-2]. PAHs have known faster sample analysis than a 0.25 mm id column typically
toxicity and in some cases have been implicated as carcino- used for GC/MS analysis .[8-9].
gens. These molecules are persistent in the environment and
bio-accumulate in fatty tissue of fish species throughout the The GC/MS system used was also equipped with backflush
food chain [3]. Fortunately, from an analyst’s prospective, capability. This capability enables faster instrument cycle time
there are historical accounts of large oil spills and their by backflushing late eluting matrix components back through
impact on aquatic species and the environment that serve as the inlet purge valve. Long bake out times between injections
case studies for investigation of future spills [4]. are avoided by using this technique. Backflushing has the
additional benefit of increasing the time intervals for source
When large scale oil spills occur, regulatory agencies spring cleaning effectively clearing deleterious matrix components
to action to minimize the risk of human exposure, contain the from the system [10-11].
leak at its source when possible, and protect the environment
as much as feasible. In the US, the National Oceanic and
Atmospheric Administration (NOAA) in coordination with the
Experimental
US Food and Drug Administration (FDA) and other federal and An Agilent 7890 GC and Agilent 5975B GC/MS System
state agencies set policies for both fishery closure and the re- equipped with a multimode inlet (MMI) and Agilent 7693
opening of affected fisheries. As part of the reopening automatic liquid sampler were used for this series of experi-
process for a closed fishery NOAA and the FDA have estab- ments. The GC was also fitted with a pressure controlled tee
lished testing procedures for PAH analyses as a requirement (PCT) post-column for automated backflush. Table 1 lists the
to reopen a fishery. Levels of concern for eight targeted PAHs chromatographic conditions used for these analyses. Table 2
and their alkylated homologues have been defined for finfish, lists flow path consumable supplies used in these experi-
oysters, shrimp and crab species. The established levels of ments.
concern range from a high of 233 ppm for
Table 1. Chromatographic Conditions
anthracene/phenanthrene in shrimp and crabs to a low of
0.03 ppm in finfish for benzo(a)pyrene [5]. GC/MSD Agilent 7890 GC/Agilent 5975B GC/MS System
Sampler Agilent 7693 automatic liquid sampler, 5.0 µL syringe
Analysis at these levels can be reliably achieved using a cap- (p/n 5181-1273)
illary GC/MS approach. Resolution of anthracene and PCT Device Purged Ultimate Union (p/n G3186-60580)
phenanthrene can also be routinely achieved using an Agilent Carrier Helium, constant flow 1.7 mL/min
J&W DB-5ms column and appropriate instrument conditions. Restrictor 0.7 m × 0.15 mm id deactivated silica tubing
Use of the Ultra Inert column in this case assures less inter- PCM 1 3.8 psi constant pressure
action between the column and the sample matrix. MMI 0.5 µL splitless; 320 °C, Purge flow 50 mL/min at
0.8 min
Sample preparation to execute the NOAA method has histori- Gas saver 30 mL/min at 2 min
cally been labor intensive involving two successive steps of Column Agilent J&W DB-5msUI 20 m × 0.18 mm, 0.18 µm
preparative chromatography. In this note the use of a (p/n 122-5522UI)
QuEChERs (Quick Easy Cheap Effective and Rugged) approach Oven 50 °C (0.4 min), 25 °C/min to 195 °C (1.5 min),
8 °C/min to 265 °C, 20 °C/min to 315 °C
to sample preparation is presented for use as a screening (1.25 min)
method for PAHs. Typically sample preparation using the
Postrun backflush 7 min at 315 °C, backflush pressure 70 psi,
NOAA method takes 12-14 hours per sample. Using a 2 psi inlet pressure during backflush
QuEChERs sample preparation approach, up to 60 samples MSD 340 °C transfer line, 340 °C source, 150 °C quad
can be processed in one eight-hour shift [6]. Sample prepara-
tion time savings are substantial for the QuEChERs approach
enabling higher laboratory throughput for the screening of
PAHs in seafood.

2
Table 2 Flow Path Supplies rpm for 1 minute. The samples were centrifuged at 4000 rpm
Vials: Amber screw top glass vials (p/n 5183-2072) for 5 min.
Vial Caps: Blue screw caps (p/n 5182-0717)
An 8 mL aliquot of the upper layer was transferred to an
Vial inserts: 100 µL glass/polymer feet (p/n 5181-8872)
Agilent Bond Elut QuEChERS fatty sample dispersive SPE 15
Syringe: 5 µL (p/n 5181-1273)
mL tube (p/n 5982-5158). The dSPE tube was vortexed for
Septum: Advanced green (p/n 5183-4759)
1 minute and then centrifuged at 4000 rpm for 5 minutes to
Inlet liners: Deactivated dual taper Helix liner complete the sample extraction. The liquid from the dSPE
(p/n G5188-5398)
tube was transferred to a GC vial and analyzed by SIM
Ferrules: 0.4 mm id short; 85/15 vespel/graphite
(p/n 5181-3323) GC/MS using the chromatographic conditions listed in
PCT fittings: Internal nut (p/n G2855-20530) Table 1.
PCT ferrules: SilTite ferrules, 0.25 mm id (p/n 5188-5361)
Extractions of water and acetonitrile aliquots were prepared
20x magnifier : 20x Magnifier loop (p/n 430-1020) in the same manner as the samples and served as a reagent
blanks.

Reagents and Chemicals Agilent Bond Elut QuEChERS Extraction Procedure for PAHs in Fish
All reagents and solvents were HPLC or Ultra Resi grade.
Acetonitrile (ACN) was from Honeywell (Muskegon, MI, Weigh 3 g sample (± 0.05g) in 50 mL centrifuge tube.
USA), and acetone was from VWR International (West
Chester, PA, USA). The 16-component PAH standard used Add surrogate/IS solution, and QC spike solution, if necessary. Vortex 1 min.
was obtained from Agilent (p/n 8500-6035).
Add 12 mL of DI water and 2 ceramic bars to the sample
Solutions and Standards (p/n 5982-9313).

The PAH stock standard solution (502 µg/mL of 16 polyaro-

matic hydrocarbons) was diluted in acetone to yield spiking Add 15 mL of ACN, vortex 1 min.
solutions of 0.5,1, 5, 10 and 50 µg/mL. The spiking solutions
were used to prepare the calibration curves in the matrix Add original Agilent Bond Elut QuEChERS extraction salt packet for 15 g
blank extract by appropriate dilution. samples (p/n 5982-6555)

Sample Preparation Shake vigorously for 1 min on Geno/Grinder at 1500 rpm.

A red snapper fish sample was purchased from a local gro-
cery store. The fish was chopped into small cubes and frozen Centrifuge at 4000 rpm for 5 min.
at –80 °C overnight. The samples were then comminuted
thoroughly to achieve sample homogeneity. The sample Transfer 8 mL of the ACN layer to Agilent AOAC fatty sample
extraction method used the QuEChERS method followed by dSPE 15 mL tube (p/n 5982-5158).
dSPE [5]. Figure 1 illustrates the sample preparation proce-
dure graphically in a flow chart. Vortex 1 min, Centrifuge at 4000 rpm for 5 min.

Samples containing 3.0 grams of fish were weighed into cen-

trifuge tubes. QC samples were spiked with an appropriate Analyze extract by GC/MS.
amount of PAH spiking solution to yield QC samples with con-
centrations of 25, 250, and 500 ng/mL. Each sample received
12.0 mL aliquot of deionized water and 15 mL aliquot of ACN. Figure 1. Flow chart of the Agilent Bond Elut QuEChERS modified extrac-
Two ceramic bars (p/n 5982-9313) were added to each sam- tion procedure for fish sample.
ple to aid in sample extraction. The samples were vortexed for
1 minute. An original Agilent Bond Elut QuEChERS extraction
salt packet (p/n 5982-6555) containing 6 grams of MgSO4 and
1.5 grams sodium chloride was added to each centrifuge tube.
The capped tubes were shaken on a Geno Grinder at 1500

3
Discussion of Results Table 3. The R2 Values for the PAH Calibration Standards Over the
10 ng/mL to 1000 ng/mL Range of this Study

The sixteen PAH analytes were resolved on the Agilent J&W

Analyte R2
DB-5msUI 20 m × 0.18 mm, 0.18 µm analysis column in less
Naphthalene 0.9997
than 20 minutes. Figure 2 shows the separation of a
500 ng/mL PAH standard solution. Quantitative GC/MS Acenaphthylene 0.9999
analysis of the target PAHs was performed using selective ion Acenaphthalene 0.9999
monitoring (SIM), which gave excellent sensitivity down to 10 Fluorene 0.9999
ppb for all the analytes. Phenanthrene 0.9999
Anthracene 0.9999
The performance of the DB-5msUI high efficiency column
Fluoranthene 0.9999
yielded excellent linearity and recovery over the calibration
Pyrene 0.9998
range of this study. The linearity of the column as defined by
the R2 values of the PAH standard curve ranged from 0.9988- Benz[a]anthracene 0.9993
0.9999. The individual PAH analyte values are shown in Chrysene 0.9994
Table 3. Excellent signal-to-noise ratios were also seen for the Benzo[b]fluoranthene 0.9990
lowest calibration standard on the column as shown in Benzo[k]fluoranthene 0.9988
Figure 3. The method limit of quantitation (MLQ) of 10 ppb for Benzo[a]pyrene 0.9992
benzo[a]pyrene is well below the level of concern of 30 ppb Indeno[1,2,3-c,d]pyrene 0.9992
set by the NOAA and FDA.
Dibenz[a,h]anthracene 0.9991
Benzo[g,h,i]perylene 0.9995

Separation of 16 PAHs on Agilent J&W DB-5msUI column

Abundance
3
130000 1. Napthalene
2. Acenaphthylene
120000 2 3. Acenaphthene
110000 5 4. Fluorene
1 5. Phenanthrene
100000 6. Anthracene
90000 4 7. Fluoranthene
6 8. Pyrene
80000
9. Benz[a]anthracene
70000 10. Chrysene
11. Benzo[b]fluoranthene
60000
12. Benzo[k]fluoranthene
50000 13. Benz[a]pyrene
7 8
40000 14. Indeno[1,2,3-c,d]pyrene
10 15. Dibenz[a,h]anthracene
30000 9 12 16 16. Benzo[g,h,i]perylene
11 13 14
20000 15
10000
0
4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00
Time

Figure 2. GC/MS chromatogram of the 500 ng/mL PAH standard prepared in sample matrix analyzed on an Agilent J&W DB-5msUI 20 m × 0.18 mm, 0.18 µm
capillary GC column (Agilent p/n 122-5522UI). Chromatographic conditions are listed in Table 1.

4
 The extraction process using the QuEChERS method followed
Excellent Signal-to-Noise Ratios at Trace Levels by dispersive SPE was effective in retaining the PAHs in the
spiked fish sample and providing sufficient cleanup of the
Benzo[a]pyrene
sample matrix for GC/MS analysis. Figure 4 shows the sepa-
S/N = 12.6 ration of the extracted PAHs in a spiked fish sample on the
DB-5ms UI column.
Abundance 1
2 3
1500 The recoveries were determined at the 25, 250, and 500 ng/mL
1400 PAH levels. Recoveries for the individual PAHs are shown in
1300 1. Benzo[b]fluoranthene
1200 2. Benzo[k]fluoranthene Table 4. The recoveries ranges (80% to 139%) and RSDs were
3. Benzo[a]pyrene
1100 excellent with the DB-5ms UI column for all PAHs investigated.
1000
900
800
700
600 12 3
500
400
300
15.00 16.00 17.00
Time

Figure 3. Enlarged view chromatogram of the benzo[a]pyrene peak in the

10 ng/mL PAH calibration standard prepared in sample matrix ana-
lyzed on the Agilent J&W DB-5msUI capillary column
(p/n 122-5522UI). Chromatographic conditions are listed in Table 1.

GC/MS SIM Chromatogram of Red Snapper Fish Extracts Blank Relative to Spiked Sample
after Agilent Bond Elut QuEChERS Extraction and Dispersive SPE

Abundance 1. Napthalene
2. Acenaphthylene
6000
3 3. Acenaphthene
5500 4. Fluorene
5. Phenanthrene
5000 6. Anthracene
2
4500 7. Fluoranthene
1 8. Pyrene
4
4000 5 9. Benz[a]anthracene
10. Chrysene
3500 11. Benzo[b]fluoranthene
6
3000 12. Benzo[k]fluoranthene
13. Benz[a]pyrene
2500 7 8 14. Indeno[1,2,3-c,d]pyrene
15. Dibenz[a,h]anthracene
2000 10 16. Benzo[g,h,i]perylene
14 16
1500 9 12 13 15
11
1000 25 ppb fortified fish extract
500
Blank fish extract
0
4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00
Time

Figure 4. GC/MS SIM chromatogram of the fish extract blank and the 25 ng/mL spiked fish extract analyzed on Agilent J&W DB-5msUI capillary column
(p/n 122-5522UI). Chromatographic conditions are listed in Table 1.

5
Table 4. Recovery and Repeatability of PAHs in Spiked Red Snapper Fish with Agilent J&W DB-5msUI Column (p/n 122-5522UI)

` 25 ng/mL fortifield QC 250 ng/mL fortifield QC 500 ng/mL fortifield QC

Analyte % Recovery RSD (n=6) % Recovery RSD (n=6) % Recovery RSD (n=6)
Naphthalene 80.35 3.29 96.77 4.23 98.64 1.88
Acenaphthylene 95.28 2.30 103.36 2.80 101.02 2.27
Acenaphthalene 92.28 2.51 101.18 2.87 100.69 2.34
Fluorene 95.98 2.99 105.94 2.82 105.00 1.28
Phenanthrene 100.51 3.46 104.93 2.71 103.25 1.70
Anthracene 107.38 3.51 105.95 3.45 105.38 1.74
Fluoranthene 113.27 3.87 105.76 3.33 103.64 1.81
Pyrene 113.55 3.51 103.99 3.24 102.29 1.94
Benz[a]anthracene 129.79 3.41 101.45 3.91 100.61 3.24
Chrysene 116.75 4.01 98.55 4.17 95.95 5.61
Benzo[b]fluoranthene 131.20 3.70 98.77 4.08 98.08 3.24
Benzo[k]fluoranthene 139.45 2.52 99.13 3.98 95.31 4.54
Benzo[a]pyrene 125.30 3.68 95.33 3.89 96.82 1.80
Indeno[1,2,3-c,d]pyrene 119.51 3.47 94.57 3.23 93.71 2.55
Dibenz[a,h]anthracene 126.35 3.54 98.55 3.50 98.85 2.24
Benzo[g,h,i]perylene 114.91 4.93 97.30 3.37 95.63 1.83

Conclusions The Agilent J&W DB-5ms UI High Efficiency column was

effective at analyzing 16 PAHs in a fish matrix following sam-
This application note successfully shows a quick and efficient ple matrix cleanup by QuEChERS and dSPE. The DB-5ms UI
analytical method for monitoring low and trace level PAHs in provided satisfactory resolution of the four known critical
fish samples suitable to address current food safety concerns. pairs; phenanthrene/anthracene, benz[a]anthracene/chry-
This method demonstrates the feasibility of a simplified sene, benzo[b]/[k]fluoranthene, and indeno[1,2,3-c,d]pyrene/
approach for routine fish screening as an alternative to more dibenz[a,h]anthracene. The performance of the DB-5ms UI
time consuming and complex techniques. yielded excellent linearity over the range of concentrations
studied with R2 values between 0.9988-0.9999 for the PAH
The Agilent Bond Elut QuEChERS method followed by dSPE compounds. Recovery and reproducibility was shown to be
for fatty samples was effective at providing enough sample greater than 80% with an RSD below 6.0. The method limit of
cleanup to avoid matrix interferences while still maintaining quantitation of 10 ppb for the benz[a]pyrene using this
low level analyte detection. The simple QuEChERS extraction approach was significantly lower than current levels of
method allows faster sample prep, facilitating higher sample concern.
throughput. Any residual sample matrix carryover is removed
through use of backflush, which eliminates the need for a
bakeout cycle, and significantly reduces analytical run times.

6
Acknowledgements 7. Kenneth Lynam and Doris Smith, “Polycyclic Aromatic
Hydrocarbon (PAH) Analysis Using an Agilent J&W DB-
The authors wish to thank Joan Stevens for her help and sug- 5ms Ultra Inert Capillary GC Column,” Agilent
gestions with the Agilent Bond Elut QuEChERS extraction Technologies publication 5989-9181EN
procedure. 8. Kenneth Lynam and Mike Szelewski, “Analysis of
Semivolatiles using High Efficiency Capillary GC
References Columns,” Agilent Technologies publication 5989-
7500EN
1. Environmental Protection Agency (EPA) of the United
9. To download Agilent Method Translation software please
States of America, Compendium Method TO-13A, EPA,
visit the link below:
Cincinnati, OH, USA, 1999.
http://www.chem.agilent.com/cag/servsup/usersoft/fi
2. European Union, Commission Recommendation les/GCTS.htm
2005/108/EC, Off. J. Eur. Comm. L34 (2005) 433.
10. Chin Kai Meng, “Improving Productivity and Extending
3. Haruhiko Nakata, Yasufumi Sakai, Takashi Miyawaki, and Column Life with Backflush,” Agilent Technologies publi-
Akira Takemura “Bioaccumulation and Toxic Potencies of cation 5989-6018EN
Polychlorinated Biphenyls and Polycyclic Aromatic
11. Harry Prest, “User Quick Guide to Pressure Controlled
Hydrocarbons in Tidal Flat and Coastal Ecosystems of the
Tee (PCT) Operation – Post Run Backflushing,” Agilent
Ariake Sea, Japan,” Environmental Science and
Technologies publication 5990-5484EN
Technology, 2003,37, 3513-3521.
4. Zhendi Wand and Scott A. Stout, “Oil Spill Environmental
For More Information
Forensics: Finger Printing and Source Identification,” For more information on our products and services, visit our
Academic Press, Burlington, MA 01803, USA, 2007. Web site at www.agilent.com/chem.
5. NOAA Cleared Document Dated 5/29/2010 “Protocol for
Interpretation and Use of Sensory Testing and Analytical
Chemistry Results from Re-Opening Oil-Impacted Areas
Closed to Seafood Harvesting.”
6. AOAC Method 2007.01

7
www.agilent.com/chem
Agilent shall not be liable for errors contained herein or
for incidental or consequential damages in connection
with the furnishing, performance, or use of this material.

Information, descriptions, and specifications in this

publication are subject to change without notice.

© Agilent Technologies, Inc., 2012

Printed in the USA
January 6, 2012
Publication Number 5990-6668EN