Poulain M; Geffroy Rodier C; Canaff C; Ruban V; Ambles A

Research Article

Austin Chromatogr. 2014;1(3): 5.

Quantification of Polynuclear Aromatic Hydrocarbons in Retention Pond Waters Using Gas Chromatography-Tandem Mass Spectrometry

Poulain M¹, Geffroy Rodier C¹*, Canaff C¹, Ruban V² and Ambles A¹

¹Institute of Chemistry of Poitiers : Materials and Natural Resources, University of Poitiers, France

²Ifsttar Nantes, French Institute of Science and Technology, France

*Corresponding author: Geffroy Rodier C, Institute of Chemistry of Poitiers : Materials and Natural Resources, UMR 7285, B 27, 4 rue Michel Brunet, B27-TSA51106, 86073 Poitiers, France

Received: September 25, 2014; Accepted: October 21, 2014; Published: October 30, 2014

Abstract

The individual quantification of 16 Polynuclear Aromatic Hydrocarbons (PAHs) in French retention pond water samples is reported. The full analytical protocol including Solid Phase Extraction (SPE) and tandem mass spectrometry detection allowed the detection limit in the range of 0.5–5 ng L^-1. PAHs quantification is achieved using a seven point calibration plot with internal calibration. Real sample analysis proved that the procedure is convenient for PAHs determination in environmental water. Due to the high levels of PAHs in retention pond waters care must be taken when considering the reuse of such samples.

Keywords: Triple quadrupole mass spectrometry; PAHs; Water; Retention ponds

Introduction

During the last decades, tremendous improvements in analytical instrumentation have allowed a better detection of pollutant in diluted systems such as water. However monitoring of pollution of ecosystems is still an ongoing challenge. Due to their mutagenic and carcinogenic properties [1,2], PAHs are included in the US Environmental Protection Agency (EPA) priority pollutant list [3]. PAHs are derived from natural or anthropogenic sources. Natural sources include forest and prairie fires [4] or post-depositional transformations of biogenic precursors [5]. Anthropogenic sources include combustion of fossil-fuel [6], long-range atmospheric transport of PAHs [7,8]; urban runoff [5] and spillage of petroleum and its refined products [9]. These organic pollutants are ubiquitous. They are found, for example, in food [10], water [11], soils, sediments [12] and air particulates [13]. In recent years, they have received considerable attention as, in addition to their toxicity, they are highly persistent and can accumulate in environmental aqueous systems. Presence of PAHs in water is now part of environmental concern.

Over the past few decades, retention ponds have been built alongside highways and motorways with the initial role of controlling water flow during rainstorms. Urban and road runoff waters contribute to pollution of receiving ponds, streams and lakes. Previous studies show that PAHs are highly concentrated in sludge of some retention ponds, i.e. 400-900 times the Dutch threshold value for polluted soils [14,15]. Thus, concentrations of such compounds in water retention ponds must be evaluated.

Many analytical techniques have been developed for the determination of organic compounds in water samples. PAHs can be extracted from aqueous matrixes by various methods such as liquid/ liquid extraction [11-16], Solid Phase Extraction (SPE) [17,18], Solid- Phase Micro Extraction (SPME) [19,20] or Stir Bar Sorptive Extraction (SBSE) [21,22]. Detection is performed by Gas Chromatography (GC) or High Performance Liquid Chromatography (HPLC). Detection limits for the EPA–PAHs depend on the selected concentration method and the detection. SPE coupled to HPLC / fluorescence is the most common procedure to quantify PAH traces in water with a detection limit in the range of ngL^-1. Compared to single stage MS modes, tandem Mass Spectrometry (MS–MS) offers a higher degree of sensitivity. MS–MS enables the analysis of organic molecule trace levels in the presence of interfering compounds without losing identification capability due to a drastic reduction of the background signal [23,24]. Triple quad technology directly enables the recovery of three different mass spectra (precursor ions, product ions and neutral fragments).

This paper presents GC-MS-MS as a sensitive alternative to HPLC for the analyses of PAHs in environmental samples. Validation parameters such as linearity, reproducibility, limits of detection and quantification are determined. Finally, the procedure is applied to the quantification of PAHs from two French retention pond waters.

Material and Methods

Chemicals and reagents

HPLC grade solvents (acetonitrile, dichloromethane and methanol) are purchased from Merck (Darmstadt, Germany) and are re-distilled before use. Water is purified on a Milli-Q SP reagent water system (Millipore, Bedford, MA, USA). Standard mixture of the 16 priority PAHs, at 10 mgL^-1 in acetonitrile, is purchased from Sigma- Aldrich. The surrogate internal standard is a mixture containing perpetuated PAHs ([²H₈] naphthalene (Np-d₈), [²H₁₀] acenaphthene (Ace-d₁₀), [²H₁₀] phenanthrene (Ph-d₁₀), [²H₁₂] chrysene (Chry-d₁₂), [²H₁₂] perylene (Pe-d₁₂) purchased from Sigma-Aldrich.

SPE is performed onto Waters Oasis HLB Sorbent cartridges (6cc/200mg) purchased from Waters. Stir bars (Twisters), 10 mm long, coated with a 0.5 mm film thickness layer (24 L) of PDMS, are from Gerstel (Mulheim, Germany).

Sample description

The Wissous pond was created in 1999. Located in the industrial area of Villemilan (France), it drains the A6 motorway area. The Saint Joseph pond is located at the North of Nantes (on the French West coast). It was created in the midth of the 20^th century; it is of urban type, draining the St Joseph district.

Solid phase extraction

SPE cartridges are conditioned with 5mL dichloromethane and 5mL methanol. Before loading on cartridges acetonitrile is added to avoid PAHs adsorption upon glassware. Standard solutions and water samples (50mL) are loaded at a flow rate of 1 mL.min^-1. Then the cartridge is rinsed with 5 mL Milli-Q water. Organics are extracted using 8 mL dichloromethane, evaporated and reconstituted in 200 μL dichloromethane. Pure dichloromethane processed through SPE units is used as the procedural blank.

Stir bar sorptive extraction

Stir bars extraction is performed according to Garcia Falcon et al. (2004) on PDMS coated bars. PAHs desorption is performed with an ultrasonic device (Sonorex Digital 10P, Bandelin GmbH, Germany) for 10 minutes in 200 μL acetonitrile.

Gas chromatography-mass mass spectrometry

GC-MS-MS analyses are carried out on a Varian 3800 GC gas chromatograph coupled to a Varian 1200 L triple Quadrupole mass spectrometer (Varian, Les Ulis, France). The GC system is equipped with a PTV injector with a SGE Liner. All analysis is carried out in split less mode at the temperature of 300 °C. The split less valve is closed for 0.75 min, and then operated in the split mode at 20 mL/ min. All injections volumes are 1μL.

Separation is achieved on a 30 m x 0.25 mm i.d.VF-5ms (Factor Four, Varian, France) coated with 5% phenyl, 95 % dimethylpolysiloxane (film thickness 0.25μm). The temperature program is 1 min at 50 °C, 10°C/min up to 100 °C, 9°C/min up to 310 °C and temperature is finally hold for 4 minutes at 310 °C. The carrier gas is Helium and the column head pressure is 15 psi to reach a constant flow rate of 1.2 ml/min.

The Quadrupole mass spectrometer is operated under electron impact ionization (70eV). The filament emission current is 500μA. The source and transfer line temperatures are at 250°C and 310°C, respectively. The electron multiplier is set at 1880 V. The acquisition mode chosen for mass spectrometry analysis is MRM mode (Multiple Reaction Monitoring). The scantime is 0.3 s/scan. Peak detection and integration are carried out using Varian Workstation version 6.3. Precursor ions are isolated using 0.7 amu isolation window and subjected to Collision Induced Dissociation (CID) in the second Quadrupole (Q2). The dissociation is induced by collisions with neutral target species of the collision gas: Argon. In the third step, ions are transmitted in the third Quadripole (Q3) for separation. In MRM mode the fragment ions are also selected in the third quadripole. Different collision energies and multiple transitions are studied for each compound.

PAHs quantization is achieved using a seven point calibration plot with internal calibration over the linearity GCMS-MS range (1- 100 μg/L) established with 40 μg/L internal perpetuated standards and 1,5,10,25, 50, 75 and 100μg/L PAH solutions. Limit of Detection (LOD) and Limit of Quantification (LOQ) are estimated at the lowest concentration of analytes having clear discerned peaks with Signal to Noise ratio (S/N) of 3 and 10 respectively.

Results and Discussion

Mass spectrometry and quantification

In order to improve sensitivity and selectivity, the MRM acquisition mode is used. In a triple quadripole instrument, the ions generated in the EI source are transmitted to the first Quadripole (Q1) where precursor ions are selected. A preliminary study is carried out in MS-MS mode under full scan conditions on Q3 in order to determine the fragment ions to be used for further MRM experiments. For each PAH from one to three ions are observed. In most cases these ions are molecular ions (M)⁺, (M-H)⁺, (M-H₂)⁺ or (M-C₂H₂)⁺ fragment ions. The collision energy is selected for each PAH to obtain the maximum signal dissociation of the product ions (table 1). The optimization of the CID voltage of each transition is performed using multiple scan function in different channel for each retention time window (segment). The influence of CID voltage is studied from 10V to 40 V at 5V steps. The example for chrysene is given in Figure 1: 10eV for (M-H)⁺ (227), 35eV for (M-H₂)⁺ (226) and 20 eV for (M-C₂H₂)⁺ (202). The argon collision gas pressure is set at 2.2 mtorr. The higher pressure of 4 mtorr is discarded because of the lower sensibility obtained at each transition (Figure 1).

Figure 1: Influence of collision gas pressure on the intensity of the three chrysene’s transitions at different CID voltage.



Figure 1:  Influence of collision gas pressure on the intensity of the three chrysene’s transitions at different CID voltage.

LOD and LOQ of GCMSMS analysis are in the range of HPLC/ Fluorescence limits and are better than those obtained with ion trap tandem mass spectrometry [25].

Table 1: GC-MS-MS segment program with fragment ions chosen for each PAH.



  
    Segment
      time (min)
    Compound
    Retention time (min)
    Precursor ion (m/z)
    Channel
    Product ion
    CID voltage
      (eV)
  
  
    0-12
    Naphthalene
    9.62
    128
    3
      2
      1
    127
      102
      78
    10
      20
      20
  
  
    12-15
    Acenaphthylene
    13.63
    152
    1
    151
    10
  
  
    12-15
    Acenaphthene
    14.09
    154
    2
    153
    10
  
  
    15-16
    Fluorene
    15.45
    166
    1
    165
    10
  
  
    16-20
    Phenanthrene
    17.96
    178
    2
      1
    177
      152
    10
      20
  
  
    16-20
    Anthracene
    18.10
    178
    2
      1
    177
      152
    10
      20
  
  
    20-24
    Fluoranthene
    21.11
    202
      202
    2
      1
    201
      200
    10
      35
  
  
    20-24
    Pyrene
    21.70
    202
    2
      1
    201
      200
    10
      35
  
  
    24-26.50
    Benz[a]anthracene
    24.89
    228
    3
      2
      1
    227
      226
      202
    10
      35
      35
  
  
    24-26.50
    Chrysene
    24.99
    228
    3
      2
      1
    227
      226
      202
    10
      35
      20
  
  
    26.50-30
    Benzo[b]fluoranthene
    27.55
    252
    2
      1
    250
      226
    35
      35
  
  
    26.50-30
    Benzo[k]fluoranthene
    27.61
    252
    2
      1
    250
      226
    35
      35
  
  
    26.50-30
    Benzo[a]pyrene
    28.29
    252
    2
      1
    250
      226
    35
      35
  
  
    30-33
    Indeno[1,2,3-cd]pyrene
    30.79
    276
    1
    275
    10
  
  
    30-33
    Dibenzo[a,h]anthracene
    30.85
    278
    4
      3
    277
      276
    35
      10
  
  
    30-33
    Benzo[ghi]perylene
    31.47
    276
    1
    275
    10



Table 1:  GC-MS-MS segment program with fragment ions chosen for each PAH.

SPE and SBSE extractions

To study environmental samples from urban and motorway retention ponds we first evaluate the extraction technique on a wide concentration range. SPE and SBSE are compared by spiking 50 mL Milli-Q water samples with the standard PAH solution in order to study concentrations ranging from 5 to 100 ng/L. Stir Bar Sorptive Extraction is a technique which has been introduced by Baltusen et al. [21], using stir bars (magnetic stirring rod incorporated in a glass jacket) coated with polydimethylsiloxane (PDMS). This phase is previously used for Solid Phase Micro Extraction (SPME) but due to a larger amount of PDMS relative to the SPME fiber, SBSE enables to increase recovery of analytes. Temperatures (room temperature, 40°C, 60°C, 80°C and 100°C) and adsorption time (from 1 to 10 hours) are studied (Figure 2). Compared to SPE, SBSE avoids the evaporation step which could explain the better recoveries (40%) for the lower molecular weight compounds at the lower concentration. For 5 ngL^-1 SPE does not allow a recovery higher than 15% whereas 45% are recovered at 100 ngL^-1. However the best conditions achieved after optimization are time and energy consuming steps (60°C and 6 hours). For the high molecular weight compounds SBSE recoveries are similar to those previously described in literature (from 40 to 50% with RSD (n=9) <10%) [22]. But are far lower than those achieved in present SPE experiments (average recovery by SPE 90%, Table 2). Regarding the diagnostic values of high molecular weight compounds we choose the SPE concentration method.

Figure 2: Influence of exposure time (from 1 to 10 hours) on SBSE recoveries at 60°C mesured at 25 ngL^-1.



Figure 2:  Influence of exposure time (from 1 to 10 hours) on SBSE recoveries at 60°C mesured at 25 ngL^-1.

Table 2: Validation parameters of the SPE-GC-MS-MS method.



  
    Compound 
    SPE 
    SBSE 
    Detection limit including SPE 
    Quantitation limit including SPE 
    Calibration curves y=ax+b
       
    R²
  
  
    Recovery^* % 
    Recovery^* % 
    (ngL^-1) 
    (ngL^-1) 
     
     
  
  
    Naphthalene 
    30 
    40 
    1.40 
    3.50 
    10.3558x
    0,997 
  
  
    Acenaphthylene 
    30 
    40 
    1.40 
    6.60 
    4.0393x
    0,999 
  
  
    Acenapthene 
    30 
    40 
    1.40 
    6.60 
    2.7364x+0.0688
    0,995 
  
  
    Fluorene 
    27 
    53 
    1.40 
    3.70 
    0.8649x
    0,998 
  
  
    Phenanthrene 
    72 
    63 
    1.50 
    1.50 
    6.3735x+0.3451
    0,999 
  
  
    Anthracene 
    75 
    62 
    0.50 
    5.50 
    4.7588x+0.4
    0,999 
  
  
    Fluoranthene 
    82 
    75 
    1.40 
    2.50 
    1.9916x+0.32
    0,972 
  
  
    Pyrene 
    85 
    75 
    0.50 
    2.50 
    2.2094x
    0,973 
  
  
    Benzo[a]anthracene 
    94 
    55 
    1.00 
    4.20 
    1.1434x
    0,993 
  
  
    Chrysene 
    95 
    47 
    1.00 
    4.20 
    1.3334x
    0,996 
  
  
    Benzo[b]fluoranthene 
    91 
    53 
    3.00 
    5.50 
    2.3022x
    0,983 
  
  
    Benzo[k]fluoranthene 
    94 
    48 
    3.00 
    5.50 
    2.6742x
    0,997 
  
  
    Benzo[a]pyrene 
    91 
    45 
    4.00 
    6.60 
    2.4684x
    0,995 
  
  
    Indeno[1,2,3-cd]pyrene 
    90 
    44 
    3.00 
    5.80 
    4.5376x
    0,996 
  
  
    Dibenzo[a,h]anthracene 
    94 
    37 
    4.00 
    6.40 
    1.3937x
    0,987 
  
  
    Benzo[ghi]perylene 
    91 
    32 
    2.00 
    4.40 
    7.3390x
    0,998



Table 2:  Validation parameters of the SPE-GC-MS-MS method.

For the study of natural samples, the SPE concentration procedure thus is performed. For natural waters, including SPE concentration step, LOD ranges between 0.50 and 4.00 ng/L (Table 2) which is ten times better than SPE GC FID validated method for similar water samples [26]. All standard solutions are injected 5 times to estimate intra-day (RDS<4%) and inter-day (RSD <6%) precisions of the SPE-GC-MS-MS method (Table 2).

Analysis of samples from water retention ponds

In water retention ponds, PAHs may be present as very complex mixture with many isomers. The use of a sensitive and selective method is then needed. The Wissous and St Joseph water samples are analyzed by the present SPE-GC-MS-MS method. Results are presented in Table 3 and illustrated on Figure 3. Low molecular weight PAHs are not detected in the two studied samples. Overall PAH concentration (sum of the 16 EPA-HAPs) is 3 times higher in Wissous sample than in Saint Joseph one (0.31 and 0.98 μg/L respectively). Ratios listed in table 3 indicate a pyrogenic origin for Saint Joseph PAHs whereas some petroleum origin is inferred for the Wissous ones [9]. Thus, chrysene (Chry) and benz [a] anthracene (B[a] A), often looked upon as the most toxic of the PAHs, are present in both samples. Regarding the value of the ratio Bz[a]A / (Bz[a]A + Chry) the dominant source of these pyrogenic PAHs is combustion (burning of wood or coal, or natural forest fires) (0.79>0.5) in St Jospeh sample but is more debatable in Wissous one (0.56) [27]. The traffic pollution is however confirmed by the B[a]Pyr / B[ghi]Pe ratio close to the 0.3-0.44 range, characteristic of traffic emission [27]. This pollution is already described in Wissous sludge [14]. Part of the PAHs found in Wissous water samples could come from the two motorways located in the neighborhood. Indeed, in such retention ponds, urban run-off could contain PAHs from gasoline and oil drips or spills exhaust products, tyre particles, and bitumen’s from road surfaces [5]. This result implies that heavy metal pollution linked to traffic could also be assumed. Furthermore, high concentration of B[a]Pyr (strongly carcinogen) in Wissous could rise serious ecological and human health risks.

Download PDF

Citation: Poulain M, Geffroy Rodier C, Canaff C, Ruban V and Ambles A. Quantification of Polynuclear Aromatic Hydrocarbons in Retention Pond Waters Using Gas Chromatography-Tandem Mass Spectrometry. Austin Chromatogr. 2014;1(3): 5. ISSN: 2379-7975

Instruction for Authors

Submit Your Article

	Saint Joseph	Wissous
Compound	ngL^-1	ngL^-1
Naphtalene	*ND	ND
Acenaphtylene	ND	ND
Acenaphtene	ND	ND
Fluorene	58	60
Phenanthrene	12	ND
Anthracene	ND	ND
Fluoranthene	14	18
Pyrene	ND	ND
Benzo(a)anthracene	22	94
Chrysene	6	74
Benzo(b)fluoranthene	22	110
Benzo(k)fluoranthene	28	130
Benzo(a)pyrene	22	96
Indeno(1,2,3 c,d)pyrene	64	134
Dibenzo(a,h)anthracene	ND	104
Benzo(g,h,i)perylene	66	158
*Σ 16 PAH*	314	978
*Bz[a]A / (Bz[a]A + Chry)*	0.79	0.56
*B[a]Pyr / B[ghi]Pery*	0.33	0.61
*ΣB[b]Fl, B[k]Fl,� In and B[g,h,i]Pery*	0.18	0.53