Environ. Sci. Technol. 2009, 43, 2871–2877
Fluorescence Spectrophotometer Analysis of Polycyclic Aromatic Hydrocarbons in Environmental Samples Based on Solid Phase Extraction Using Molecularly Imprinted Polymer R E D D I T H O T A J . K R U P A D A M , * ,† BHAGYASHREE BHAGAT,† SATISH R. WATE,† GHANSHYAM L. BODHE,† BORJE SELLERGREN,‡ AND YERRAMILLI ANJANEYULU§ National Environmental Engineering Research Institute, Nagpur 440 020, India, INFU, Dortmund University, 44221 Dortmund, Germany, and Department of Chemistry, Jackson State University, Jackson, Mississippi 39217
Received September 9, 2008. Revised manuscript received January 23, 2009. Accepted February 3, 2009.
A molecularly imprinted polymer (MIP) was synthesized using a polycyclic aromatic hydrocarbon (PAH) standard as a template, methacrylic acid as a functional monomer, ethylene glycol dimethacrylate as a cross-linker, and acetonitrile as a porogen. This polymer was used as a solid phase adsorbent for the quantitative enrichment of PAHs in coastal sediments, atmospheric particulates, and industrial effluents. The MIP selective adsorption capacity for PAHs started reducing when the chemical oxygen demand (COD) and total dissolved solids (TDS) was more than 800 mg L-1 in the targeted environmental samples. The adsorption stability of the MIP was tested by the consecutive contact of environmental samples, and it was shown that the performance of the MIP did not vary after 10 enrichments and desorption cycles. Recoveries of eight PAH compounds, extracted from 10 g of coastal sediments and 1 L of industrial effluent spiked with 10 µL of standard PAHs, showed recoveries between 85 and 96%. The fluorescence spectrophotometer limit of detection of PAHs varied from 10 to 30 ηg L-1 in industrial effluent and from 0.1 to 2.9 ηg kg-1 in solid samples (coastal sediment and atmospheric particulates), and this indicates that the environmental analytical method is significantly sensitive, when compared with other commonly used methods such as gas chromatography-mass spectrometry and liquid chromatography-mass spectrometry.
Introduction Polycyclic aromatic hydrocarbons (PAHs) usually occur as complex mixtures consisting of two or more condensed benzene rings, and the resulting molecules contain only carbon and hydrogen atoms. PAHs are among the group of * Corresponding author phone: +91-712-2249885; fax:+91-7122249896; e-mail:
[email protected]. † National Environmental Engineering Research Institute. ‡ Dortmund University. § Jackson State University. 10.1021/es802514c CCC: $40.75
Published on Web 03/13/2009
2009 American Chemical Society
compounds defined as persistent organic pollutants (POPs). Because of their low vapor pressures, PAHs with four rings or more readily adsorb onto particles present in combustion emissions (1, 2). Discovery of the carcinogenic properties of benzo(R)pyrene in the 1930s first stimulated interest in PAH chemistry. Since then many PAHs have been identified as potent carcinogens and mutagens, and the U.S. Environmental Protection Agency (EPA) has designated 16 PAHs as priority pollutants (3). PAHs are present in gasoline, diesel engine exhaust, cigarette and wood smoke, and natural gas and oil-fired burner emissions; therefore, in urban areas, there are countless individual point sources of PAHs, and approximately 90% of PAH emissions are estimated to be anthropogenic (4). In cities, vehicular traffic is the most significant contributor to the atmospheric PAH load. In an effort to reduce PAH emissions from industry and transportation, some countries have adopted stringent air quality standards for selected PAHs; for example, the United Kingdom has a proposed annual average standard for benzo(R)pyrene (BaP) of 0.25 ηg m-3 (5). In the 1980s, the U.S. EPA was very concerned with analyzing PAHs in different environmental matrices and developing analytical methods for their study, especially for the 16 EPA priority PAHs (6). There are several problems associated with PAH detection and quantification. First is their nonpolar structure and extremely low water solubility, making their concentration in natural water extremely low. In order to increase the sensitivity of the analyses, preconcentration and extraction techniques are required. Not only do these steps increase analysis time, cost, and use of reagents, but they also introduce the possibility of quantitative error due to analyte loss. Second is the case of wastewaters and sediments that usually contain many organic species, often at relatively high concentrations, which can interfere with PAH detection and quantification. Complicated, multicomponent environmental matrices thus require multistep, time-consuming chemical analyses that preclude real-time, in situ measurements. Because of the quite low concentration of PAHs, as little as 0.01% (mass), in the organic content (7), separation techniques are often applied prior to analysis to achieve selectivity. At present, PAHs are separated as they are extracted from environmental samples such as soils, sludge, and wastewater, following methods such as Soxhlet extraction (SE), automated SE, supercritical fluid extraction (SFE), and pressurized liquid extraction (PLE) (8, 9). The most commonly adopted U.S. EPA Methods 8100 and 8310 for industrial sludge and effluents are also sometimes not able to provide clean extract for analysis (10, 11). The sample cleanup depends very much on the selectivity of the applied extraction technique. Interestingly, when nonpolar organic solvents are used for analyte extraction, humic acids and aliphatic hydrocarbons dissolved in analyte pose matrix implications in trace analysis. Adoption of solid phase extraction techniques, on the other hand, has been successfully applied to clean up environmental samples (12). Commercially available activated carbon and carbon nanotubes have been tested using microscale SPE for determination of PAHs in water samples and show promise in significantly lowering limits of detection (LOD) (13). Molecularly imprinted polyurethanes were used as sensitive coatings for detection of PAHs in water where PAHs were detected in the parts per thousand range with little matrix effect by humic acid (14). Dickert et al. reported double molecular imprinting to prepare MIPs for PAH detection in water. Measurements of PAHs in water were performed with a quartz crystal microbalance, which shows that humic acid interference is not found, and hence, VOL. 43, NO. 8, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2871
the fluorescence property of the molecularly imprinted polymer (MIP) had no change (15). However, the breakthrough thresholds and recovery percentages reported have been poor. To achieve the goal of preparing clean extract, the adsorbent must have a high adsorption capacity, rapid adsorption/desorption kinetics, selectivity, and low fouling. Because of their high mechanical and thermal stability, high adsorption capacity, and attractive selectivity, molecularly imprinted polymers are gaining increasing attention as adsorbent materials (16, 17). In this article, a molecularly imprinted polymer (MIP) adsorbent was synthesized via template-directed molecular imprinting, where copolymerization of methacrylic acid-ethylene glycol dimethcarylate was used. A PAH standard consisting of 16 PAH compounds was used as a template so that we can hypothesize the formation of various PAH compounds imprinted in the matrix of polyacrylate. It is important in environmental analysis to measure a group of pollutants together, rather than analyze a single pollutant, which saves analysis time and reduces chemicals and reagents used for sample processing. The three types of environmental samples chosen for the study are (i) sediment (coastal) samples (BS) collected from the Bombay coast, where commercial port activities and offshore oil production are a major source of PAHs; (ii) wastewater samples (WW) (industrial effluent) collected from a petrochemical refinery effluent treatment plant (ETP) outlet; and (iii) industrially originated dust (particulate) samples (IOD) from a petrochemical industry. Because of their aromatic structures, PAHs are inherently fluorescent. Here, we used fluorescence spectrophotometry to analyze PAHs extracted from various environmental matrices, using a molecularly imprinted polyacrylate as an adsorbent. The adsorption capacity of the MIP sorbent was compared with commercially used activated carbon; this explained the merits of the newly developed fluorescence spectrophotometric method for environmental PAH analysis.
Experimental Section Chemicals and Materials. PAH-imprinted polymer precursors, methacrylic acid (MAA) as a functional monomer, ethylene glycol dimethacrylate (EGDMA) as the cross-linking agent, and 2,2’-azobisisobutyronitrile (AIBN) as a polymerization reaction initiator were purchased from Sigma-Aldrich (Buchs, Switzerland) and stored in a N2 atmosphere prior to use. The cross-linker EGDMA was washed with 15% NaOH, saturated with NaHCO3 and NaCl solutions, dried with CaH2, and distilled before use. Methacrylic acid was distilled under vacuum prior to use; all other chemicals were used as procured. Acetonitrile was obtained from Merck (Darmstadt, Germany) and used as received. An analytical stock standard of PAHs was obtained from AccuStandard, Inc. (AE-00025, 10 ML, New Haven, CT), and the solution was stored at -18 °C. Deionized water (20 MΩ) was obtained from a Milli-Q water purification system (Millipore, UK). Preparation of the PAH-Imprinted Polymer Sorbent. By adopting bulk polymerization, PAH-imprinted polymer sorbent was synthesized via photoinitiated polymerization. The 2,2′-azobisisobutyronitrile (AIBN) was used as an initiator (0.05 g), whereas EGDMA was a cross-linking agent. One milliliter of a standard PAH solution was used as a template. The PAH standard contains 16 compounds, from which eight compounds are efficiently fluorescently active (Table SI-1 of the Supporting Information). In the polymerization, a ratio of 1:4 of template/monomer (MAA) was maintained, whereas the monomer/cross-linker (EGDMA) ratio is 4:40. The polymerization was carried out in a 50 mL glass vessel, using 15 mL of acetonitrile as a solvent. Sealed glass tubes containing the reaction mixture were freeze-thaw-degassed by submerging the tubes in liquid nitrogen and holding the 2872
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 8, 2009
frozen tubes under a vacuum of 100 mTorr for a period of at least 15 min for each cycle. Each sample was freeze-thawdegassed in this manner for three cycles. The tubes were thawed, charged with a positive pressure of argon, placed in ice-filled dewars, and temperature equilibrated. These dewars were then placed under a water-cooled, medium-pressure mercury vapor lamp (550 W) for a period of 24 h. During this time the tubes were turned, and the ice in the dewars was changed three times to maintain a constant temperature of 0 °C during the entire polymerization. Upon completion of polymerization, the tubes were taken out of the dewars and crushed. The polymers were ground in a ball mill and dry sieved to a size between 100 mesh (149 µm) and 200 mesh (74 µm). The PAHs were extracted in a batch mode, using chloroform containing 10% acetonitrile as the solvent (10 times × 10 min, room temperature), followed by washing three times with the respective imprinting solvent, i.e., acetonitrile. The percent template recovery from MIP showed 99.99 ( 0.01% after the 10th washing (see data in Table SI-2 of the Supporting Information). The corresponding nonimprinted polymer (NIP) was prepared in parallel in the absence of a template and treated in the same manner. Characterization. The pore parameters and surface area of the samples were measured using a N2 porosimeter (Model JEOL JSM-6400). A 100 mg quantity of dry MIPs was used and degassed at 170 °C under a nitrogen flow for approximately 6 h prior to measurement. The nitrogen adsorption-desorption data were recorded at the liquid nitrogen temperature (77 K). The specific surface area was calculated using the Brunauer-Emmett-Teller (BET) equation. The mesopore size distribution was determined by the Barrett-Joyner-Halenda (BJH) method, and the micropore distributions were determined by microspore analysis method. A fluorescence spectrophotometer (Hitachi Model F-4500, Japan) was used for trace level detection of PAHs. The fluorescence emission peaks were recorded in an emission-excitation matrix (EEM) mode between 290 and 550 nm, using an integration time of 90 s. For better analysis, the spectra were cut down to excitation and emission ranges of 290-420 and 421-550 nm, respectively, for easy quantification of individual PAH compounds in environmental samples. The fluorescence intensity of the peak represents the concentration of the PAH compound in the sample. Environmental Sample Preparation. One of the aims of this study is to develop a selective MIP adsorbent for PAH extraction from environmental samples. A sediment (coastal) sample (BS) was collected from the Bombay Coast near Mumbai Port. The sample was sun dried and then ground in a mortar to a fine powder before use. One gram of the sediment sample was spiked with 1 mL of a standard PAH solution. The spiked sample was allowed to stand overnight before extraction. The extraction of PAHs from the sediment sample was performed according to Dinelli’s method with a minor modification. A 2 mL sample of sodium hydrocarbonate solution (0.1 M, pH 7.8) was added to 1.0 g of a spiked sediment sample, and the suspension was shaken for 1 h. The slurry was then centrifuged at 5000 rpm for 10 min. The extraction procedure was repeated three times, and the liquid extracts were combined. After the addition of 0.3 g of EDTA, the extract was centrifuged, and the supernatant was used as a sediment sample. A blank sediment sample, without PAHs, was also prepared by the same procedure described above. A wastewater (WW) sample (industrial effluent) was collected from the outlet of a petroleum refinery effluent treatment plant (ETP) after tertiary treatment. The sample was filtered using a glass fiber filter (MN 85/90 BF, MacheryNagel, Du ¨ ren, Germany) to remove particles larger than 0.5 µm. The filtrate was collected in a brown bottle and kept at 4 °C in the dark until analysis. Before being spiked with
TABLE 1. Adsorption Capacity of MIP, NIP, and AC spiked PAH standard to 5 mL of water/acetonitrile solution (99:1%, v/v) adsorption (ηg mL-1) amount of adsorbent MIP used
amount of adsorbent NIP used
amount of adsorbent AC used
PAH compound
initial concentration (ηg mL-1)
5 mg
10 mg
50 mg
5 mg
10 mg
50 mg
5 mg
10 mg
50 mg
acenaphthene benzo(R)anthracene benzo(R)pyrene chrysene fluoranthene fluorene phenanthrene pyrene ∑PAH
1.25 0.49 0.98 0.99 2.01 2.01 1.48 1.98 11.19
0.05 0.04 0.09 0.11 0.23 0.21 0.15 0.22 1.1
0.16 0.25 0.55 0.82 1.35 1.16 1.19 1.45 6.93
0.18 0.29 0.59 0.86 1.39 1.22 1.27 1.51 7.31
0.01 0.02 0.05 0.09 0.01 0.04 0.12 0.14 0.48
0.03 0.07 0.16 0.39 0.03 0.09 0.21 0.57 1.53
0.09 0.12 0.19 0.41 0.07 0.13 0.24 0.59 1.84
0.03 0.02 0.05 0.08 0.11 0.09 0.06 0.07 0.51
0.13 0.18 0.36 0.75 0.91 0.86 0.84 0.88 5.07
0.15 0.20 0.38 0.79 0.96 0.89 0.84 0.92 5.13
TABLE 2. Recoveries (%), Precision, and Limits of Detection (LOD) for Eight Fluorescently Active PAHsa wastewater (industral effluent) samples (WW) (1 L)
sediment (coastal) samples (BS) (10 g)
acenaphthene benzo(R)anthracene benzo(R)pyrene chrysene fluorenthrene fluorene phenanthrene pyrene
industrially originated dust (particulate) samples (IOD) (10 g)
PAH standard (10 µL)
recovery (%)
RSD
LOD
recovery (%)
RSD
LOD
recovery (%)
RSD
LOD
recovery (%)
RSD
LOD
73 71 91 95 82 86 83 89
2.8 3.1 3.5 1.2 2.4 3.5 2.5 3.8
12 14 7 14 11 8 10 7
79 74 84 98 87 83 91 86
4.8 5.3 5.2 4.4 6.1 4.3 4.8 2.3
15 16 10 17 12 19 14 12
81 89 92 91 90 87 90 93
2.3 1.7 3.1 2.4 1.9 2.5 3.1 2.5
8 78 34 97 68 38 7 9
99 101 102 98 97 107 100 96
6.5 8.7 4.3 5.0 4.4 4.7 4.2 4.8
5 7 3 5 5 6 6 5
a Data collected after adsorption onto MIP spiked with 10 µL of a standard PAH solution. LOD defined as three times the standard deviation calculated at the spike level considered (RSD, n ) 3). The unit LOD of the liquid sample is ηg L-1 and of solid sample (sediment and particulates) is µg kg-1. Recovery of environmental samples was determined by spiking the PAH standard and then desorbing from the MIP. A standard addition plot was constructed, and linear regression was performed on the data points. The slope of the regression line corresponds to the recovery value.
standard PAH, the free metal ions in the water samples were complexed by the addition of EDTA by adding 9 mg per 100 mL level. This mixture was centrifuged at 15000 rpm for 10 min to remove turbidity. The supernatant was used for adsorption studies. The industrially originated dust (particulate) samples (IOD) were collected using high-volume samplers (13.8 m3 h-1, Environtech, India) from industrial ambient air samples. The air sampler was operated on the industrial premises for 170 ( 2 h, and IOD were collected on Whatmann GF/B-100 nm glass fiber filters (Buchs, Switzerland). Each filter (approximately 1-2 g of IOD) was cut into four identical parts. Each cut filter was cleaned by ultrasonication in dichloromethane for 15 min, and then in methanol for 15 min. Afterward, they were dried at 150 °C for one night and finally stored in a glovebox at constant humidity until use. One gram of IOD was placed in a 10 mL glass vial, and to it was added 5 mL of acetonitrile. The content in the glass vial was agitated in a water bath for 3 h at 70 rpm. Later, the glass vial was centrifuged, and supernatant was decanted. The decanted sample solution was spiked with 0.1 mL of a PAH standard, and this was called the “real sample”. The real sample (5 mL) was spiked with 0.1 mL of a standard PAH solution, and then an adsorption-desorption study was carried out using MIP as a sorbent for PAH enrichment prior to fluorescence spectrophotometry. These environmental samples, before and after spiking of PAHs, were also analyzed for significant parameters such as chemical oxygen demand (COD) and total dissolved solids (TDS) (18). PAH Analysis. The fluorescence spectrophotometer was calibrated for qualification of the 16 EPA PAH compounds
and the two internal standards with triplicate measurements of standard solutions at four different concentration levels in the range from a few µg L-1 up to 100 µg L-1. Emission excitation matrix (EEM) spectra were measured for eight PAH-water standards and water blanks, using an imaging fluorescence spectrophotometer (Figure SI-1 of the Supporting Information). The spectral peak areas were integrated with the software package CLASS-VP (Hitachi, Japan), and linear calibration functions of peak area versus concentration with correlation coefficients R > 0.999 were obtained (number of data points, n ) 9). The detection limit of 0.3-1.5 µg L-1 corresponds to 0.1-0.5 ηg of PAHs in environmental samples. Three repeated measurements were performed for every sample, and the obtained peak areas were corrected for extraction efficiency and losses during cleanup and analysis by multiplication with the inverse recovery of the internal standards. The recovery of water/acetonitrile (99:1%, v/v) was used for the three analytes because the water soluble portion of PAHs is significant in an environmental perspective. Recovery of the PAH standard solution was determined by using different PAH standard concentrations desorbed from MIP. A standard addition plot was constructed, and linear regression was performed on the data points. The slope of the regression line corresponds to the recovery value. The experiment was performed in duplicate, and the mean recovery values are presented in Table 2. The overall uncertainty of the determination of the PAH content of the investigated environmental samples is estimated to be (10% or less. Adsorption Study. PAH adsorption onto MIP and other adsorbents was performed in batch mode. The dry adsorbents VOL. 43, NO. 8, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2873
FIGURE 1. (a) Nitrogen adsorption-desorption isotherm, (b) pore size distribution for PAHs-MIP, (c) nitrogen adsorption-desorption isotherm, and (d) pore size distribution for NIP. (MIP, NIP, or AC; 10 mg) were weighed into 5 mL glass vessels, and to this were added 0.1, 0.2, 0.3, 0.4, and 0.5 mL of a standard PAH solution mixed with a water/acetonitrile (99: 1%, v/v) solvent in separate glass vessels. The samples were then shaken in a water bath at 25 °C for 3 h. After sedimentation of the adsorbents, the concentration of PAHs was measured using a fluorescence spectrophotometer. The amount of PAH adsorbed was calculated by substraction, using a calibrating curve obtained from the same experiment leaving out the adsorbent. Commercially activated carbon (AC) was used for comparison with MIP and NIP for their adsorption capacity and selectivity. For each adsorbent, these experiments were repeated at least twice. Amounts of PAH compounds bound to the imprinted polymer [SA] were calculated by the following equation [SA] ) (C0-Ct)V/W
(1)
where, C0 and Ct are the PAH compound concentrations (mg mL-1) measured at initial and after interval time (h) for equilibrium, respectively. Symbols V and W are the volume of the PAH solution (10 mL) and the weight of the dry polymer (10 mg) used for the binding experiments, respectively.
Results and Discussion Surface Properties of MIP. MIP particles were prepared by bulk polymerization of MAA with a PAH mix template in the presence of a large excess of cross-linker and small amount 2874
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 8, 2009
of initiator. The PAH mix was leached out leaving cavities that were molecular-size dependent. Nitrogen sorption measurements performed on the PAH imprinted MIP (BET surface area, 785 m2 g-1; cumulative pore volume, 0.579 cm3 g-1) displayed a type IV isotherm with large adsorption capacities of 0.42-0.98 cm3 g-1. A total of 85-91% of the total surface area (674 ( 1.2 m2 g-1) of the MIP had a distribution of nanocavities with diameters in the range of 0.5-0.75 nm. The isotherm displayed a large step in a nanoporous region and then flattened to indicate a lack of macropores (Figure 1a). The MIP showed a hysteresis loop, where the desorption curve closed but leveled above the adsorption curve. This was pronounced for the MIP exhibiting no shrinking, when subjected to increasing pressure at a liquid N2 temperature. This represents that the MIP has low swelling and low solvent uptake. During desorption, the release of N2 gas from the pores of certain size ranges (25-75 Å) at a lower relative pressure is required, rather than be calculated from the Kelvin equation (19). Possibly, the formation of pores can be reduced by a gas with a lower saturation pressure (Po), which would allow measurements at a higher temperature. The adsorption average pore diameter was in closer agreement with values obtained from the BET calculation (Table SI-3 of the Supporting Information). The wide pore size distribution and lower average pore diameter derived from the adsorption plot (Figure 1b) indicate the existence of a heterogeneous continuum of “bottle shape”
pores (20). This represents that the size of the cavities formed in the MIP matrix had an important role in binding capacity. These results are in agreement with our previous results (21), but in this study, we achieved more surface area in MIP, which is an indication of a better imprinting effect. Also, different PAH molecules were imprinted in a single polymer matrix, leading to the formation of a narrow range pore size distribution. MIPs showed higher surface area than NIPs. The N2 adsorption-desorption isotherms (BET, 98.5 m2 g-1; single-point total pore volume, 0.023 cm3 g-1) also yielded type IV isotherms (Figure 1c) with the adsorption branch exhibiting a linear region from P/Po of ∼0.2-0.5, which is characteristic of broad pore size distribution in the mesoporous range (Figure 1d). NIP was nonporous with essentially no surface area or internal pore volume. The roughness of the particle surface should be considered as a factor providing an increase in the surface area of MIP that reduces diffusion resistance and facilitates mass transfer through a high internal surface area (22), which is a desirable property of an efficient adsorbent. Binding Capacity of the MIP. Binding properties of the MIP were evaluated on the basis of equilibrium binding experiments in water/acetonitrile (99:1%, v/v). Batch rebinding experiments were performed with a PAH mix, where individual compound concentrations were found in the range of 0.49-2.01 µg mL-1. The rebinding amounts increased with time and became constant after 3 h. To estimate the capacity of the MIP, 10 µL of water/acetonitrile (99:1%, v/v) was spiked with a PAH mix standard contacted with 5, 10, and 50 mg of a MIP separately for 3 h. After attaining equilibrium (3 h), the content in the glass vial was centrifuged at 15000 rpm for 5 min, and then the supernatant was analyzed with a fluorescence spectrophotometer. Out of the eight PAHs studied, only 10-20% of the initial concentration of acenaphthene (Ace), benzo(R)anthracene (BaA), and aenzo(R)pyrene (BaP) were adsorbed onto 5 mg of MIP (Table 1). At 10 mg of MIP, 70% BaP, 93% chrysene (Chr), 73% fluoranthene (Flu), 82% phenanthrene (Phe), 73% pyrene (Pyr), and 50% BaA adsorbed indicate that the interaction between PAHs and MIP was saturated. An exception was Ace, which was still not reasonably adsorbed. This may be due to its lower hydrophobic property (low Kow value compared to those of other PAHs), which explains its lower binding capacity in nonspecific (NIP/MIP) and specific (MIP) interactions. It is observed that the binding capacity of Chr to MIP is highest. The saturated binding amounts of [SA] increased from 5 to 10 mg of MIP used as an adsorbent. For example, the [SA] for the total PAHs was 1.53 and 6.93 µg mg-1 of polymer for NIP and MIP, respectively. The binding of PAH compounds onto NIP was not similar to that observed with MIP. This happens when the imprinted process of PAHs encoded the PAH shape, although the sites slightly showed reasonable nonspecific binding for NIP and AC. Batch rebinding experiments conducted for AC showed nonspecific binding for all of the PAH compounds with a similarity in binding kinetics and capacity of MIP (Figure SI-2 of the Supporting Information), but the samples obtained from AC are not clean. Not only does AC adsorb PAHs, but chemical oxygen demand (COD), heavy metals, and even dissolved inorganics also cause interference during analysis with GC-MS or fluorescence spectrophotometry. Matrix Effect. The presence of high dissolved organic matter (COD) and high total dissolved solid (TDS) causes a strong interference in trace level detection of specific environmental pollutants. The COD produces a strong background for the determination of polar compounds because of its broad range of molecular weights and intense UV-vis adsorption. It is almost difficult to quantify the target environmental pollutants because their signals may be completely overlapped by the interfering background. The
effect of organic matter on the absorption of PAHs onto MIP was studied for environmental samples. In this study, humic acid was not chosen as a representative of organic matter because industrial effluent (WW) and coastal sediment (BS) samples need not be only humic acid. It was found that the solubility of humic acid is not linearly reflecting the concentration of COD. Hence, we used COD as an indicator of matrix effect. Industrial effluent (WW) with a COD value of 1200 mg L-1 was diluted with Milli-Q water to obtain 600, 300, 150, and 75 mg L-1 before spiking PAH compounds, and then the PAH adsorption onto MIP was studied. The fluorescence signal of the MIP-adsorbed samples was largely reduced when COD is more than 800 mg L-1. Interestingly, the samples tested for COD after adsorption of PAHs onto MIP found only a (5 mg L-1 reduction in concentration of initial COD (Figure 2), while AC had shown a greater COD reduction, i.e., (260 mg L-1. These results indicate that MIP is selective enough to limit COD interference, while AC had not produced a clean sample. A COD higher than 800 mg L-1 reduces the PAH adsorption up to 10-15% and can be explained on the basis of masking the surface leads fouling of MIP. As per pollution regulations in India, the industrial effluent discharge standards for COD and TDS are 100 and 500 mg L-1, respectively. During treatment of industrial effluents (WW), it is quite easy to remove dissolved organics (COD) by aerobic and/or anaerobic reactions, following tertiary treatment using AC or resin adsorption columns to meet the regulatory standard set at 100 mg L-1. In the case of TDS, there are no practicable methods available to remove dissolved inorganic solids. The effect of inorganic common ions such as Ca2+, Mg2+, and Al3+ on the adsorption capacity of MIP was studied because their existence in environmental samples is quite common. The inorganic ions present in sediment (BS), industrial effluents (WW), and air particulates (IOD) were represented as total dissolved solids (TDS), where the soluble organic fraction was removed by oxidation of samples with K2Cr2O7 in acidic conditions at 150 °C. The interference of TDS was studied by diluting the environmental samples 2-5 times with Milli-Q water before evaluating the adsorption capacity of MIP for PAHs. It was found that all PAH compounds were adsorbed onto MIP without any deviation in their adsorption capacity and selectivity up to 800 mg L-1 TDS. The reason for this could be that major ions such as Ca2+, Mg2, and Al3+ could not form complexes with PAHs because no functional groups or functional monomers (MAA) having -COOH groups begins to form complexes when the TDS is greater than 800 mg L-1. NIP and AC showed diminished adsorption of PAHs when increasing TDS in environmental samples. When the PAH extracts of MIP, NIP, and AC are compared, the matrix interference in the case of MIP is almost completely removed, and consequently, the specific PAHs were analyzed free of complex matrix implications using fluorescence spectrophotometry. MIP has many advantages when compared to traditionally used adsorbents for enrichment of environmental pollutants. Because of the existence of specific size pores in the polymer, selectively adsorbed targeted PAHs remove matrix complications in environmental analysis. This is an important issue in environmental analysis where very low concentrations need to be quantified. Also, MIP has a selective adsorption capacity for a group of molecules (PAHs). Whenever we use selectivity, it represents adsorption of a molecule onto the adsorbent; however, we demonstrated adsorption of a group of environmental contaminants (PAHs) onto a single polymer, which is an innovative approach. In addition to the above two distinct features, the use of a small quantity of MIP (high adsorption capacity) and its reusability (easy to regenerate) make this material environmentally benign. These features VOL. 43, NO. 8, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2875
FIGURE 2. Interference of TDS and COD in the adsorption of PAHs onto MIP. are not commonly seen in other commercial adsorbents such as activated carbon, etc. Fluorescence Spectrophotometry Analysis of Environmental Samples. To demonstrate the applicability and reliability of the fluorescence method for detection of PAHs in environmental samples, the standard PAHs and the spiked environmental samples were cleaned with MIP. After recovery of PAHs from MIP, the fluorescence spectrophotometry analysis was carried out, and the data are summarized in Table 2. As observed, for analysis of PAH spiked samples, the PAH recoveries were higher than 82% and were unaffected by the nature of the aqueous matrix in which analytes were dissolved. However, for Ace and BaA, it could not be detected in more complex matrices such as coastal sediment (BS) and industrial effluent (WW) samples because its signal is completely masked by the matrix background. The relative standard deviation (n ) 5) for quantification was between 1.2 and 4.1% for coastal sediment (BS), and 2.3 and 6.1% for industrial effluents (WW), which are acceptable values for real sample analysis. The limit of detection (LOD) calculated for coastal sediment (BS) and air particulates (IOD) was between 7 and 14 ηg kg-1, which is lower than the stipulated standard value of 100 ηg kg-1 set by pollution control boards and agencies. The LOD for the industrial effluent (WW) samples was very similar and varied from 10 to 19 ηg L-1. It is very interesting to see that the existence of PAHs is more 2876
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 8, 2009
dominant in sediment and soils (BS), followed by atmospheric particulates (IOD). Because of low water solubility and limited volatility, PAHs are present in very low concentrations in effluents (WW) (Figure SI-3 of the Supporting Information). The very similar fluorescence spectra of effluents (WW) and air particulates (IOD) show that the PAHs can undergo a phase transfer from water to air and vice versa, and this is quite significant. In conclusion, the newly developed MIP proved to be an efficient selective adsorbent for cleaning environmental samples for PAH analysis, which improves the detection limits of the fluorescence spectrophotometer. PAHs prominently include 16 compounds, but in this study, fluorescently active compounds were analyzed in the range of 0.01-20 µg mL-1. In this way, our objective to analyze a series of PAH compounds in a single attempt was proven successful in our environmental analysis.
Acknowledgments We acknowledge financial support by the Council of Scientific & Industrial Research (CSIR), India, under its Supra Institutional Project scheme SIP-16 (3.3).
Supporting Information Available Figure SI-1 shows the fluorescence emission-excitation matrix of PAH compounds. Figure SI-2 shows the adsorption capacity of different adsorbents used for adsorption of total
PAH from environmental samples. Figure SI-3 displays the fluorescence spectra of environmental samples and standard PAHs. Table SI-1 lists molecular properties and physichemical characteristics of PAHs. Table SI-2 lists recovery of PAHs from MIP during washings. Table SI- 3 lists SEM and N2 porosimetry data of MIP, NIP, and AC. This information is available free of charge via the Internet at http:// pubs.acs.org.
Literature Cited (1) Hertel, G. R.; Rosner, G.; Kielhorn, J. Selected non-heterocyclic aromatic hydrocarbons. International Program on Chemical Safety (IPCS), Environmental Health Criteria, No. 202; World Health Organization: Geneva, Switzerland, 1998. (2) Harvey, R. R. G. Polycyclic Aromatic Hydrocarbons; John Wiley & Sons: New York, 1997. (3) Go¨tze, H. J.; Schneider, J.; Herzog, H. G. Determination of polycyclic aromatic hydrocarbons in diesel soot by highperformance liquid chromatography. Fresenius J. Anal. Chem. 1991, 340, 27–30. (4) Chetwittayachan, T.; Shimazaki, D.; Yamamoto, K. A comparison of temporal variation of particle-phase polycyclic aromatic hydrocarbons (PAHs) concentration in different urban environments: Tokyo, Japan, and Bangkok, Thailand. Atmos. Environ. 2002, 36, 2027–2037. (5) Lohmann, R.; Northcott, G. L.; Jones, K. C. Assessing the significance of domestic burning to the atmospheric concentrations of PCDD/Fs, co-planar PCBs and PAHs. Environ. Sci. Technol. 2000, 34, 2892–2899. (6) Wenzl, T.; Simon, R.; Kleiner, J. E. Analytical methods for polycyclic aromatic hydrocarbons (PAHs) in food and the environment needed for new food legislation in the European Union. TrAC, Trends Anal. Chem. 2006, 25, 716–725. (7) Prevedouros, K.; Brorstrom-Lunden, E.; Halsall, C. J.; Jones, K. C.; Lee, R. G. M.; Sweetman, A. J. Seasonal and long-term trends in atmospheric PAH concentrations: Evidence and implications. Environ. Pollut. 2004, 128, 17–27. (8) Knopp, D.; Niessner, R. Biomonitors Based on Immunological Principles. In Solid Waste: Assessment, Monitoring and Remediation; Twardowska, T., Allen, H. E., Kettrup, A. A. F., Lacy, W. J., Eds.; Elsevier Science BV: Amsterdam, 2004; Chapter 4, pp 505-537. (9) Schauer, C.; Niessner, R.; Po¨schl, U. Polycyclic aromatic hydrocarbons in urban air particulate matter: Decadal and seasonal trends, chemical degradation, and sampling artefacts. Environ. Sci. Technol. 2003, 37, 2861–2868.
(10) Sauvain, J. J.; Duc, T. V.; Huynh, C. K. Development of analytical method for the simultaneous determination of 15 carcinogenic polycyclic aromatic hydrocarbons and polycyclic aromatic nitrogen heterocyclic compounds: Application to diesel particulates. Fresenius J. Anal. Chem. 2001, 371, 966–974. (11) Namiesnik, J.; Zabiegala, B.; Kot-Wasik, A.; Partyka, M.; Wasik, A. Passive sampling and/or extraction techniques in environmental analysis: A review. Anal. Bioanal. Chem. 2005, 381, 279– 284. (12) Koester, C. J.; Moulik, A. Trends in environmental analysis. Anal. Chem. 2005, 77, 3737–3741. (13) Wang, J. J.; Fraser, D. G.; Law, B.; Lewis, M. D. Identification and quantification of urinary benzo(R)pyrene and its metabolites from asphalt fume exposed mice by microflow LC coupled to hybrid quadrupole time-of-flight mass spectrometry. Analyst 2003, 128, 864–871. (14) Dickert, F. L.; Tortschanoff, M. Molecularly imprinted sensor layers for the detection of polycyclic aromatic hydrocarbons in water. Anal. Chem. 1999, 71, 4559–4563. (15) Dickert, F. L.; Achatz, P.; Halikias, K. Double molecular imprintingsa new sensor concept for improving selectivity in the detection of polycyclic aromatic hydrocarbons (PAHs) in water. Fresenius J. Anal. Chem. 2001, 371, 11–15. (16) Ouyang, G.; Chen, Y.; Pawliszyn, J. Time-weighted average water sampling with a solid-phase microextraction device. Anal. Chem. 2005, 77, 7319–7325. (17) Lai, J. P.; Niessner, R.; Knopp, D. Benzo(R)pyrene imprinted polymers: synthesis, characterization and SPE application in water and coffee samples. Anal. Chim. Acta 2004, 522, 137– 144. (18) American Water Works Association (AWWA)/American Public Health Association (APHA): Standard methods for wastewater analysis. New York, 2001. (19) Gregg, S. J.; Sing, K. S. W. Adsorption, Surface Area, and Porosity; Academic Press: London, 1982. (20) Sellergren, B.; Shea, K. J. Influence of polymer morphology on the ability of imprinted network polymers to resolved enantiomers of special interest. J. Chromatogr., A 1993, 635, 31–49. (21) Krupadam, R. J.; Ahuja, R.; Wate, S. R. Benzo(R)pyrene imprinted polyacrylate nanosurfaces: Adsorption and binding characteristics. Sens. Actuators, B 2007, 124, 444–451. (22) Sellergren, B., Ed. Molecularly Imprinted Polymers-Man-Made Mimics of Antibodies and Their Applications in Analytical Chemistry; Elsevier: Amsterdam, 2001.
ES802514C
VOL. 43, NO. 8, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2877
