- Email: [email protected]
Rapid fluorometric determination of perfluorooctanoic acid by its quenching effect on the fluorescence of quantum dots
Journal of Luminescence 161 (2015) 374–381
Contents lists available at ScienceDirect
Journal of Luminescence journal homepage: www.elsevier.com/locate/jlumin
Full Length Article
Rapid fluorometric determination of perfluorooctanoic acid by its quenching effect on the fluorescence of quantum dots Qi Liu, Aizhen Huang, Nan Wang n, Guan Zheng, Lihua Zhu College of Chemistry and Chemical Engineering, Huazhong University of Science and Technology, Wuhan 430074, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 27 August 2014 Received in revised form 26 December 2014 Accepted 17 January 2015 Available online 28 January 2015
Analysis of perfluorooctanoic acid (PFOA) usually requires a combination of high-performance liquid chromatography and mass spectrometry, which is expensive and time-consuming. In the present work, water-soluble CdS quantum dots (QDs) were employed to develop a simple and rapid fluorometric method for the determination of PFOA. Strongly fluorescent CdS QDs were prepared by using 3-mercaptopropionic acid (MPA) as a stabilizer. It was observed that PFOA strongly quenched the fluorescence emission of the MPA-CdS QDs because PFOA promotes the aggregation of MPA-CdS QDs through a fluorine–fluorine affinity interaction. Under optimum conditions, the fluorescence intensity of MPA-CdS QDs was observed to decrease linearly with an increase in the concentration of PFOA from 0.5 to 40 μmol L 1, with a limit of detection of 0.3 μmol L 1. This new method was successfully implemented for the analysis of PFOA-spiked textile samples, with recoveries ranging from 95% to 113%. & 2015 Elsevier B.V. All rights reserved.
Keywords: Perfluorooctanoic acid CdS quantum dots Fluorescence Quenching
1. Introduction Perfluorocarboxylic acids (PFCAs) have been widely used as emulsifying agents in fluoropolymer manufacturing and as surfactants in paints, lubricants, photolithography, polishers, food packaging and fire-fighting foams. Within the PFCA group, perfluorooctanoic acid (C7F15COOH, PFOA) is the most commonly detected compound for two reasons: PFOA finds use in numerous applications and PFOA is a stable degradation product of precursor perfluorinated chemicals. PFOA has been widely found in sediment, municipal wastewater, coastal water, and even tap water [1–3]. It was reported that PFOA was detectable in the coastal seawaters of the Pearl River Delta, including the South China Sea, and in Korea, with concentration ranges of 0.24 to 16 and 0.24 to 320 pg L 1, respectively [4]. Much higher concentrations were found in the wastewaters of factories involved in the manufacture or use of PFOA, e.g., concentrations of up to 3.35 mmol L 1 were detected in the untreated wastewater from a fluoropolymer manufacturing plant [5]. PFOA exhibits bioaccumulation in wildlife and humans, and it is potentially carcinogenic. Therefore, it is important to develop analytical methods for the determination of PFOA. The main analytical methods used for PFOA determination are gas chromatography (GC) and high-performance liquid chromatography (HPLC). Due to the high polarity of PFOA, the direct injection of PFOA into a GC system will result in rather severe tailing of the peaks. Therefore, GC determination of PFOA requires appropriate
n
Corresponding author. Tel.: þ 86 27 87543032; fax: þ 86 27 87543632. E-mail address: [email protected] (N. Wang).
http://dx.doi.org/10.1016/j.jlumin.2015.01.045 0022-2313/& 2015 Elsevier B.V. All rights reserved.
derivatizations [6], which can be carried out with tetrabutylammonium hydrogen sulfate [7], diazomethane [8], isobutyl chloroformate [9] or 2,4-difluoroaniline [10]. Due to a lack of chromophores, PFOA is not easily amenable to traditional HPLC methodologies. Among the reported HPLC methods for PFOA determination, mass spectrometry (MS) detection is employed most frequently. However, HPLC–MS requires expensive instrumentation, a professional operator, complicated sample pretreatment and considerable analysis time [11]. To develop new HPLC methods for simpler detection, Ohya et al. derivatized PFCAs by using laboratory-synthesized 3-bromoacetyl7-methoxycoumarin for HPLC analysis with fluorescence detection [12]. Based on the use of the commercially available fluorophore 3-bromoacetyl coumarin as a derivatization reagent, Poboży developed another HPLC method with fluorimetric detection [13]. However, the derivatization procedures are time-consuming and the formed derivatives exhibit limited stability, producing a substantial source of uncertainty in PFOA analysis [14]. Recently, Takayose et al. developed a colorimetric method for the detection of PFOA utilizing polystyrene-modified gold nanoparticles (Au NPs) [15]. PFOA with carboxylate groups can be bound to the surface of Au NPs, whereas fluorine–fluorine interactions between PFOA-modified Au NPs encourage interparticle aggregation, causing the color of Au NP suspensions to change from red to blue-purple. However, this colorimetric sensor has very limited sensitivity, and it cannot clearly detect PFOA at concentrations below 250 μmol L 1. It is known that fluorescent sensors usually provide a considerably higher sensitivity than colorimetric methods. The purpose of the present work was to develop a rapid fluorometric method for the
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
determination of PFOA. To this end, we considered another group of NPs, quantum dots (QDs). It is worth noting that QDs have sizedependent and interparticle-distance-dependent optical and electronic properties. By varying the size of QDs, the emission wavelength and fluorescence quantum yield can be tuned. For example, differentsized CdS QDs emit blue to near-UV light [16], whereas the fluorescence quantum yield of CdSe QDs can be changed by varying the particle size (i.e., 18% for 2.8 nm, 38% for 3.3 nm, and 2% for 4.0 nm) [17]. In addition, the luminescence intensity of QDs is also sensitive to the interparticle distance; therefore, it is expected that fluorine–fluorine affinity-induced QDs aggregation may cause a change in fluorescence intensity, which makes QDs-based fluorescence analysis of PFOA feasible. To the best of our knowledge, thus far, no reports have employed the photoluminescence of QDs to detect PFOA. Thus, in the present work, the direct involvement of PFOA in the control of aggregation-dispersion of MPA-CdS QDs was studied and a simple and sensitive MPA-CdS QDs system for the fluorescence detection of PFOA was developed.
375
having been washed 5 times, the MPA-CdS QD deposits were dried at 60 1C by a vacuum drier. 2.3. Apparatus Transmission electron microscopy (TEM) images of MPA-CdS QDs were acquired on an FEI TECNAI G2 20U-TWIN at an accelerating voltage of 200 kV. The colloidal solution of the NPs in water was dropped onto a 0.1-nm-thick carbon-coated copper grid, with the excess solution immediately removed. FTIR spectra were recorded on an Equinox 55 Fourier Transform Spectrometer (Bruker Germany). UV–vis absorption spectra were recorded on a Cary 60 UV–vis spectrophotometer (Agilent Technologies). Fluorescence measurements were performed on an FP6200 fluorescence spectrophotometer (Jasco, Japan). pH measurements were conducted using a Sartorius PB-10 pH meter. All optical measurements were performed at room temperature and under ambient conditions. Zeta potentials and hydrodynamic diameter measurements were performed on a Malvern ZEN 3690.
2. Experimental section
2.4. Measurements
2.1. Reagents
Unless specified otherwise, the basic reaction conditions for the determination of PFOA were as follows: 1.0 mL of MPA-CdS QDs stock dispersion (200 mg L 1) and a given volume of PFOA standard stock solution were rapidly mixed in a 10-mL test tube, followed by the addition of 1.0 mL of Na2HPO4–NaOH buffer to maintain the pH at 10. Then, the mixture solution was diluted to 10 mL with water. After 10 min, fluorescence emission spectra were obtained at an excitation wavelength of 365 nm. Both the excitation and emission slit widths were set to 10 nm, and the photomultiplier tube voltage was 400 V.
3-Mercaptopropionic acid (MPA, HS–CH2–CH2–COOH) was obtained from Aladdin Chemical Reagent Company, Cd(ClO4)2 6H2O was obtained from Alfa Aesar Company, and sodium sulfide nonahydrate was obtained (Na2S 9H2O) from Shanghai LingFeng Chemical Reagent Company. PFOA (96%), heptafluorobutyric acid (PFBA, C3F7COOH, 99%) and pentafluoropropionic acid (PFPrA, C2F5COOH, 97%) were purchased from Acros (New Jersey, USA), perfluoroheptanoic acid (PFHpA, C6F13COOH, 98%) from Alfa Aesar (Lancs, UK), undecafluorohexanoic acid (PFHeA, C5F11COOH, 98%) and perfluoropentanoic acid (PFPeA, C4F9COOH, 98%) from Tokyo Kasei Kogyo Co., Ltd. (Tokyo, Japan). Other chemicals were purchased from Shanghai Chemical Reagent Company (Shanghai, China). All reagents were of analytical reagent grade or the highest purity available and directly used without further purification. All aqueous solutions were prepared with double-distilled deionized water. A PFOA stock standard solution (0.5 mmol L 1) was prepared and then stored at 4 1C. A series of pH buffer solutions was prepared by mixing 0.1 mol L 1 NaH2PO4 and 0.1 mol L 1 H3PO4 to maintain a pH range of 3.0 to 4.0, 0.1 mol L 1 NaH2PO4 and 0.1 mol L 1 Na2HPO4 for a pH range of 5.0 to 8.0, and 0.1 mol L 1 Na2HPO4 and 0.1 mol L 1 NaOH for a pH range of 8.0 to 11.0.
Untreated textiles were purchased from a local supermarket. Samples of fabric ( 5 cm 5 cm) were placed into a 50-mL beaker, followed by the addition of 25 mL of water. The samples were then fortified by adding PFOA stock solution, if needed. For example, a recovery study was carried out on samples spiked with 10–30 μmol L 1 PFOA to evaluate the developed method. After the mixture had been left at 40 1C under ultrasound irradiation for 40 min, the sample was filtered under vacuum and washed with water. The collected water sample was then diluted to 100 mL for analysis.
2.2. Preparation of CdS QDs
3. Results and discussion
Water-soluble CdS QDs were prepared with MPA, a capping agent, through the procedures described by Hosseini [18], with some modifications. Cd(ClO4)2 6H2O (0.84 g, 2 mmol) was dissolved in 100 mL of water to obtain a 0.02 mol L 1 Cd2 þ solution. MPA (1.6 mL, 20 mmol) was added to the Cd2 þ solution, resulting in a Cd2 þ /MPA molar ratio of 10:1. After the solution pH had been adjusted to 10 by adding a 2.0 mol L NaOH solution, the obtained mixture was transferred to a 500-mL three-neck, round-bottom flask. Under purging with pure nitrogen gas, the mixture was heated to the boiling point and maintained at that temperature for 30 min. Next, 100 mL of 0.02 mol L 1 freshly prepared Na2S solution was added to the solution to reach a Cd2 þ /S2 molar ratio of 1:1. The bright-yellow colloid was sealed for 2 h of incubation at its boiling point. All steps were carried out under continuous magnetic stirring and purging with pure nitrogen gas. After having been cooled to room temperature, the colloid solution was washed with an equal volume of alcohol and centrifuged to remove excess precursors and contaminants. After
3.1. Characterization of MPA-CdS QDs
2.5. Analysis of textile samples
Hydrophilic MPA molecules were adopted as capping agents to prepare water-soluble CdS QDs. Fig. 1a showed the FTIR spectra of the bare CdS QDs, MPA and MPA-CdS QDs. The characteristic IR absorption bands occurred at 3430 cm 1 (O–H stretching vibration, vOH (adsorbed H2O)), 1623 cm 1 (O–H bending vibration, δOH(H2O)), 1117 cm 1(vS–S), 1013 cm 1(vS ¼ O), and 626 cm 1 (vCd–S) for the bare CdS QDs (curve 1) and at 3595 (vOH of free COOH), 3400– 3099 cm 1 (vOH of COOH with intermolecular hydrogen bonding), 2947 cm 1 (vCH(CH2)), 2661 cm 1 (vS–CH), 2569 cm 1 (vS–H), 1710 (vC ¼ O), 1426 cm 1 (δOH of C–OH) and 1250 cm 1 (vC–O) for MPA (curve 2). In the FTIR spectrum of the MPA-CdS QDs, a broad absorption band at approximately 3376 cm 1 was a characteristic peak of vOH(H2O), the strong absorption peaks at 1565 cm 1 and 1399 cm 1 could be assigned to the asymmetric stretching vibration (vas(COO )) and symmetric stretching vibration (vs(COO )) of
376
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
800
Absorbance (a.u.)
Transmittance (%)
1
2
3
600
0.8
400 0.4
1
2
200
1:CdS 2:MPA 3:MPA-CdS QDs
4000
3000
1800
1200
600
Fluorescence intensity (a.u.)
1.2
0
0.0 200
300
-1
Wavenumber O cm P
400 500 Wavelength (nm)
600
Fig. 1. (a) FTIR spectra of naked CdS QDs, MPA and MPA-CdS QDs. (b) The UV–vis absorption (1) and fluorescence emission (2) spectra of MPA-capped CdS solution (20 mg L 1) at pH 10.0. The fluorescence excitation wavelength was 365 nm and the slit widths were 10 nm for both excitation and emission. 40
(c)
Number (%)
30
20
10
0 0.1
1
10 Size (nm)
100
1000
Fig. 2. TEM images of (a) bare and (b) PFOA-adsorbed MPA-CdS QDs. The PFOA-adsorbed MPA-CdS QDs samples were prepared by pre-immersing MPA-CdS QDs in PFOA solution for 2 h. (c) The hydrodynamic diameter and size distribution of MPA-CdS QDs.
carboxylate anion of MPA, respectively [19–20]. The prominent appearance of the vas(COO ) and vs(COO ) bands and the disappearance of the vC ¼ O band indicated that the carboxyl group of MPA dissociated into carboxylate species and that MPA was successfully coated as carboxylate on the CdS QDs surface [20]. However, the peaks corresponding to vS–H were absent in the spectra of the MPACdS QDs, indicating covalent bonding between the thiol group and the Cd atoms on the QDs surface [21]. The optical properties of MPA-CdS QDs were investigated by both UV–vis absorption spectrometry and fluorescence spectroscopy. As shown in Fig. 1b, the aqueous solution of MPA-CdS QDs exhibited a characteristic absorption profile below 440 nm (curve 1) and a strong fluorescence emission peak at 560 nm when excited by radiation at 365 nm with 10-nm slit widths for both excitation and emission (curve 2). Fig. 2a showed a TEM image of the MPA-CdS QDs. As shown in Fig. 2a, the MPA-CdS QDs were spherical with a primary diameter of 6 nm. After the MPA-CdS QDs had been dispersed in water at pH 10, their hydrodynamic diameter and size distribution were studied by a laser scattering particle size distribution analyzer. The MPA-CdS QDs had a number-weighted hydrodynamic diameter of 8.071.9 nm (Fig. 2c), similar to the particle sizes observed by TEM. This result suggested that the MPA-CdS QDs were highly dispersible in water. 3.2. PFOA-induced quenching of fluorescence of MPA-CdS QDs Fig. 3a illustrated the emission spectra of MPA-CdS QDs in the absence and presence of PFOA. When mixed with PFOA, the fluorescence intensity of MPA-CdS QDs decreased dramatically, and the quenching became stronger with increasing PFOA concentration. For
example, the relative fluorescence intensity (F/F0) of MPA-CdS QDs was 0.971 for 2 μmol L 1 PFOA and 0.200 for 40 μmol L 1 PFOA, where F0 and F are the fluorescence intensity of the MPA-CdS QDs in the absence and presence of PFOA, respectively. This result indicates that MPA-CdS QDs may act as fluorescence probes for sensing PFOA. The fluorescence quenching data were analyzed by both the Stern–Volmer equation and the Lineweaver–Burk equation, which are usually employed to describe the dynamic and static quenching behavior, respectively. These two equations are expressed as follows: F0 ¼ 1 þ K SV ½PFOA F
ð1Þ
ðF 0 F Þ 1 ¼ ðF 0 Þ 1 þ K LB 1 ðF 0 Þ 1 ½PFOA
ð2Þ
where [PFOA] is the concentration of PFOA, KSV is the Stern–Volmer quenching constant, and KLB denotes the binding constant of PFOA and MPA-CdS QDs. The plot of F0/F versus [PFOA] showed a positive deviation (R2 ¼0.8722), whereas the plot of (F0 F) 1 versus [PFOA] showed good linear behavior (R2 ¼0.9953). This finding suggests that the quenching was not initiated by dynamic collision but mainly by the formation of a complex, namely static quenching. From the slope and intercept of the Lineweaver–Burk curve, KLB was calculated to be 1.7 104 L mol 1. The quenching rate constant, Kq, could be calculated according to the following equation: Kq ¼
K LB
τ0
ð3Þ
where τ0 represents the emission lifetime of QDs. Generally, the lifetimes of various QDs fall between one and a few tens of
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
377
0.16
0.12
4
9
400
(F0-F)-1
600
F 0 /F
Fluorescence intensity (a.u.)
5
1
800
3
0.08
2 0.04
200 1
0 450
0.00
0 500
550
600
650
0
10
20
30
0.0
40
c PFOA (µmol L ) -1
Wavelength (nm)
0.5
1.0
c PFOA-1(L
1.5
2.0 -1
µmol )
1
Fig. 3. (a) The fluorescence emission spectra of MPA-CdS solution (20 mg L ) in the absence (1) and presence (2–9) of PFOA at pH 10.0. From 2 to 9: 0.5, 2, 5, 15, 25, 30, 35 and 40 μmol L 1 PFOA. (b) The Stern–Volmer and (c) Lineweaver–Burk curves for analyzing the fluorescence quenching data of MPA-CdS QDs by PFOA at pH 10.0.
1800
1400
1000
600
1
-30
Transmittance (%)
0.4
Zeta potential (mV)
1
2
3 1: MPA-CdS QDs 2: MPA-CdS QDs+PFOA 3: PFOA
1800
1400
1000 -1 Wavenumber O cm P
600
-34
2 -38 PFOA octoic acid
-42 0
20
40
60
c (µmol L ) -1
Fig. 4. (a) FTIR spectra of MPA-CdS QDs before (1) and after (2) the adsorption of PFOA. The FTIR spectrum of PFOA was also given as a reference in (a) (curve 3). (b) Zeta potentials of MPA-CdS QDs solution in the presence of (1) PFOA and (2) octanoic acid at different concentrations.
nanoseconds (approximately 30 ns) [22]. Therefore, Kq ranges from approximately 5.7 1011 to 1.7 1013 L mol 1 s 1, values that are much greater than those (2 1010 L mol 1 s 1), through the maximum scatter collision mechanism [23]. This finding further supports the notion that the fluorescence quenching mainly arises from the formation of a ground-state complex between MPA-CdS QDs and PFOA. To better understand the formation of the complex, we compared the FTIR spectra of the MPA-CdS QDs before and after the addition of PFOA. The FTIR absorption bands of the CF2 and CF3 groups appeared at the same positions for both PFOA and PFOA-adsorbed MPA-CdS QDs, i.e., at 1203 cm 1 and 1147 cm 1 (curves 2 and 3 in Fig. 4a), which supports the binding of PFOA on the MPA-QD surfaces. The complex interaction between PFOA and MPA-CdS QDs could be recognized from the stretching band shift of the carbonyl group of PFOA, as shown in Fig. 4a. The C¼ O stretching bands of pure PFOA powder appeared at 1770 and 1727 cm 1, corresponding to the free carbonyl group and the carbonyl group with intermolecular hydrogen bonding, respectively [24]. In contrast, two vibrational peaks at 1694 and 1659 cm 1 were observed in the spectrum of PFOA-adsorbed
CdS QDs, presumably due to the associated or dimerized carboxylic groups of PFOA [25]. This shift in the C¼O band to lower wavenumbers also suggests that the associated carboxylic groups persisted in the hydrogen-bonded form with MPA. Moreover, the vas(COO ) and vs(COO ) bands in the case of the PFOA-adsorbed MPA-CdS QDs were shifted to lower wavenumbers of 1555 cm 1 and 1415 cm 1, respectively, compared with those of the MPA-CdS QDs (vas(COO ), 1565 cm 1; vs(COO ), 1399 cm 1), further suggesting a strong interaction between the anionic carboxylate group of MPA-CdS QDs and PFOA molecules. The interaction between PFOA and the MPA-CdS QDs was also indicated by the change in the surface charge of the MPA-CdS QDs before and after the addition of PFOA. Fig. 4b shows that the zeta potential of the MPA-CdS QDs depended on the concentration of PFOA. As the concentration of PFOA increased from 0 to 40 μmol L 1, the zeta potential of the MPA-capped CdS QDs at pH 10.0 varied from 41.3 to 30.7 mV. Prior to the addition of PFOA, the MPA-CdS QDs showed a negatively charged surface at pH 10.0 because the MPA-CdS QDs were protected by the carboxylate anion, as indicated by FTIR analysis (Fig. 1a). In the presence of PFOA, PFOA
378
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
can be adsorbed on the MPA-CdS QDs surface through the interaction between the proton of PFOA and the anionic carboxylate group, which thereby offsets the contribution of the carboxylate anions to the net surface charge of the MPA-CdS QDs, resulting in a shift in the zeta potential to less-negative values. However, it should be noted that octanoic acid, which possesses a similar carboxylic structure, only induced a slight change in the zeta potential of MPA-CdS QDs from 41.3 to 37.3 mV. This disparity suggests that there is another essential factor that caused the adsorption-induced variations in the zeta potential of PFOA-adsorbed MPA-CdS QDs. In particular, the fluorine–fluorine interaction between PFOAadsorbed MPA-CdS QDs encourages MPA-CdS QDs to aggregate with each other, causing the negatively charged carboxylate anions of MPA-CdS QDs to become embedded within aggregates. Consequently, the net charge was further reduced. The TEM images in Fig. 2a and b clearly showed that, in the absence of PFOA, welldispersed MPA-CdS QDs tended to aggregate after the addition of PFOA. Takayose also observed a similar phenomenon: the addition of PFOA caused the detachment of the polystyrene layer from the surface of gold nanoparticles, which resulted in interparticle aggregation [15]. Once the aggregation occurred, the collision between MPA-CdS QDs themselves would quench the fluorescence emission. Thus, the principle of detecting PFOA was based on the aggregationinduced fluorescence quenching of MPA-CdS QDs, as shown in Fig. 5. PFOA was bound on the MPA-CdS QDs through the carboxylate group, and consequently, the fluorine–fluorine affinity induced MPA-CdS QDs aggregation, which decreased the fluorescence emission via collisional self-quenching.
3.3. Optimization of assay conditions 3.3.1. Effect of pH The effect of pH was studied over a pH range between 3 and 11. The fluorescence intensity of the MPA-CdS QDs themselves increased with pH and then became nearly constant when the solution pH was greater than 10 (curve 1 in Fig. 6a). An increase in pH promoted the dissociation of the carboxyl groups of the MPACdS QDs to negatively charged carboxylate species, which increased the absolute zeta potential value of the MPA-CdS QDs (Fig. 6b), as shown in Fig. 6b. This increase in absolute potential resulted in an increase in the electrostatic repulsion between particles, promoting better water dispersibility and stronger fluorescence emission of the MPA-CdS QDs. In the presence of PFOA, except for the effect on fluorescence measurements, the solution pH also affected the adsorption of PFOA on the MPA-CdS QDs, which is a primary requirement for fluorescence quenching. In this study, the pH was adjusted in two steps: the pH of PFOA was adjusted first and then the pH values of the mixtures of PFOA and the MPA-CdS QDs were adjusted prior to the fluorescence measurements. One set of samples was prepared by varying the pH of the first step (pH1st) from 3 to 10 and keeping the pH of the second step (pH2nd) fixed at 10.0. It was observed that the value of F/F0 reached an approximately constant value at pH 3.0–4.5 (0.173–0.200) and then increased to 0.352 at pH 7, to 0.635 at pH 8.5, and to 0.994 at pH 10. This difference was observed because the carboxyl group of PFOA provides a proton to an anionic carboxylate-modifying group of CdS QDs, as indicated by FTIR analysis (see Section 3.2). In turn,
Fig. 5. Schematic illustration of the fluorescent analysis of PFOA using MPA-CdS QDs. MPA-CdS QDs are stably dispersed and show strong fluorescence, while the adsorption of PFOA induces MPA-CdS QDs aggregations via fluorous affinity, leading to a significant fluorescence quenching.
1000
1.0
2
0.6 500 0.4 CdS CdS+40µM PFOA 250
1
Zeta potential (mV)
750
F/F0
Fluorsecencec intensity (a.u.)
0
0.8 CdS CdS+40µM PFOA
-15
-30 2
0.2 1
-45
0
0.0 3
5
7 pH
9
11
3
5
7 pH
9
11
Fig. 6. Effects of pH on (a) fluorescence intensity and (b) zeta potential of MPA-capped CdS (20 mg L 1) in the absence (1) and presence (2) of 40 μmol L 1 PFOA.
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
this donation of a proton is due to the dissociation of carboxylic acid groups in both PFOA and MPA-CdS QDs at high pH, which would have been harmful to the interaction between PFOA anions and negatively charged CdS QDs. Another set of samples was prepared by changing pH2nd at a given pH1st of 4.0. As shown in Fig. 6a, PFOAinduced fluorescence quenching was strongest at approximately pH 9 11 (curve 2). Accordingly, the PFOA-induced change in the zeta potential of the MPA-CdS QDs increased roughly with increasing solution pH and then decreased after reaching a maximum at pH 10 (curve 2 in Fig. 6b). Therefore, 4.0 and 10.0 were chosen as the optimum pH levels for the formation of PFOA-adsorbed MPA-CdS QDs first and subsequent fluorescence measurements, respectively.
3.3.2. Effect of MPA-CdS QDs concentration Fig. 7 illustrated the effect of MPA-CdS QDs concentration on fluorescence intensity in the absence and presence of PFOA. With an increased loading of MPA-CdS QDs, the fluorescence intensity of the MPA-CdS QDs increased initially (Fig. 7a). Then, the intensities increased very slightly after the MPA-CdS QDs concentration reached 40 mg L 1 (Fig. 7a), possibly due to the self-quenching of fluorescent MPA-CdS QDs at higher concentrations [21]. In addition, MPA-CdS QDs clusters easily
aggregated at higher concentrations, which caused a reduction in the fluorescence intensity. After the addition of PFOA, the fluorescence intensity of the MPA-CdS QDs decreased, and the fluorescence quenching efficiency was also found to depend on the MPA-CdS QDs concentration. As shown in Fig. 7b, as the MPA-CdS QDs concentration decreased, the F/ F0 value became more sensitive to changes in the PFOA concentration. However, as the MPA-CdS QDs concentration decreased to 10 mg L 1, F/ F0 was not linearly dependent on the PFOA concentration if the PFOA concentration was greater than 20 mg L 1, indicating a decrease in the linear relationship between F/F0 and the PFOA concentration. To reach a compromise between the sensitivity and the linear range of the calibration function, 20 mg L 1 was selected as the optimum concentration of MPA-CdS QDs.
3.3.3. Effect of reaction time By monitoring the fluorescence spectra at different reaction time intervals, we observed that the fluorescence of the MPA-CdS QDs was immediately quenched immediately after the addition of PFOA. As shown in Fig. 8a, the fluorescence intensity of the MPA-CdS QDs in the presence of PFOA reached an approximately constant value within 5 min, which could be maintained for at least 30 min. Thus,
1500
1.0 0.8
1000 F / F0
Fluorescence intensity (a.u.)
379
0.6
cQDs -1
10 mg L -1 20 mg L -1 30 mg L -1 40 mg L -1 50 mg L
0.4
500
0.2 0
0.0 0
20
40 cQDs (mg L-1)
0
60
10
20
30
40
cPFOA(µmol L-1)
Fig. 7. Effect of the QDs concentration on (a) the fluorescence intensity of MPA-CdS QDs and (b) the relative fluorescence intensity (F/F0) in the presence of PFOA.
1.0
0.8
0.8
0.6
0.6
F/F0
F/F0
1.0
0.4
0.4
0.2
0.2
F/F0=-0.0209c+0.9978 R2=0.9950
0.0
0.0 0
10
20 Time (min)
30
0
10
20
30
40
50
cPFOA(µmol L-1 )
Fig. 8. (a) Effect of the reaction time on the relative fluorescence intensity (F/F0) of MPA-CdS (20 mg L 1) in the presence of 40 μmol L 1 PFOA at pH 10. (b) Calibration plot for the PFOA determination with MPA-CdS QDs.
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
100
75
75
0
PF Pr A
0
PF BA
25
PF Pe A
25
PF H xA
50
PF H pA
50
Relative error (%)
100
Fo rm ic ac Pr id op an oi ca cid O ct an oi ca cid M et ha no l Et ha no Is l op ro pa no l Bu ta no l H ex an ol
Relative error (%)
380
Fig. 9. The relative error due to added interferents on the estimated PFOA concentration through the present fluorescence method. The concentration of added PFOA and interfering compound in (a) was 20 and 40 μmol L 1, respectively. Because PFOA account for over 80% of the PFCAs group, the relative concentration ratio of PFOA to other interfering PFCAs was set as 4:1 in (b), i.e., 32 μmol L 1 of PFOA and 8 μmol L 1 of C3-C7 PFCAs.
fluorescence measurements were conducted 10 min after mixing the reactants. Under optimal conditions, the calibration curve for PFOA determination was obtained by establishing the correlation between the relative fluorescence intensity of MPA-CdS QDs (F/F0) and the PFOA concentration (c). Fig. 8b showed that F/F0 was linearly dependent on c over the range of 0.5 to 40 μmol L 1 with a regression equation of F/F0 ¼ 0.0209cþ0.9978 (R2 ¼0.9950). The detection limit was calculated to be 0.3 μmol L 1 (S/N¼3). The limit of detection for this method was far superior to that of the Au NP-based colorimetric method (250 μmol L 1) [15]. Moreover, the relative standard deviations of 11 replicate measurements for 5 and 30 μmol L 1 PFOA were 1.69% and 1.40%, respectively, indicating that the present method provided high reproducibility.
Table 1 Measurements of PFOA spiked textile samples using the proposed fluoremetric method (n ¼3). Sample
Textile I
Textile II
Test no.
1 2 3 1 2 3
PFOA concentration (μmol L 1) Added
Found
0 10.0 30.0 0 10.0 30.0
n.d. 11.3 7 0.5 28.5 7 0.3 n.d. 10.9 7 0.2 29.2 7 0.5
Recovery (%)
– 113.0 7 4.4 95.0 7 1.1 – 109.0 7 1.8 97.3 7 1.7
3.5. Detection of PFOA in real samples 3.4. Effects of possible interfering compounds The effect of potential interfering compounds, including carboxylic acids and alcohols, was also studied. Compared with PFOA at the same concentration of 40 μmol L 1, the tested carboxylic acids and alcohols, including formic acid, propionic acid, octanoic acid, methanol, ethanol, isopropanol, butanol and hexanol, exhibited no significant quenching effect on the fluorescence intensity of MPA-CdS QDs. Moreover, the effect of these compounds on the fluorescence of MPACdS QDs was studied by mixing PFOA and interfering carboxylic acids or alcohols at twice the amount of PFOA. As shown in Fig. 9a, the relative errors due to the added interferents on the estimated PFOA concentration ranged from 4% to 16%. These results suggest that the present method exhibits good selectivity toward PFOA. Furthermore, taking into account that other PFCAs might also be capable of inducing nanoparticle aggregation, several typical C3-C7 PFCAs were also examined as potential interferents. It was found that C3-C7 PFCAs could also quench the fluorescence of MPA-CdS QDs, but the quenching effect was 71–88% that for PFOA at the same molar concentration. In addition, it should be kept in mind that PFOA accounts for over 80% of the PFCA group. When the C3–C7 PFCAs were individually added to the solution containing PFOA at their relatively normal concentration levels in the environment (i.e., no more than one-fourth of the PFOA concentration), the relative error induced was not greater than 11% (Fig. 9b). This result indicates that the MPA-CdS QDs-based fluorometric system proposed herein could be applied to detect PFOA in real samples.
To validate the reliability of the proposed method in determining the concentration of PFOA, two types of textile samples obtained from a local supermarket were used as real sample models. Because no PFOA was detected by the proposed method, known quantities of PFOA were added to the textile suspensions, and then the PFOAspiked textile samples were analyzed according to the procedures described in Section 2.5. As shown in Table 1, the resulting fortification recoveries ranged from 95% to 113%, and the relative standard deviations (RSD) obtained for every three measurements were less than 5%. These results indicate that the proposed technique provides a suitable strategy for determining PFOA in real samples.
4. Conclusions A rapid, simple and sensitive fluorescence sensor was proposed for determining PFOA by the utilization of MPA-CdS QDs. By using TEM, FTIR, zeta potential and fluorescence techniques, we demonstrated that PFOA could be bonded to the surface of MPA-CdS QDs, which resulted in significant interparticle aggregation of MPA-CdS QDs and consequent quenching of their fluorescence emission. Little or no interference from many carboxylic acids and alcohols suggested that both a carboxyl group and fluorocarbon group are required for fluorescence quenching. This method could detect PFOA within approximately 10 min, and it provided a working range of 0.5 to 40 μmol L 1 with a detection limit of 0.3 μmol L 1.
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
The analysis of textile samples yielded an acceptable recovery of 95% to 113%. This work also provides an alternative methodology for designing nanoparticle-based fluorescent assays for the determination of PFOA. Acknowledgements This work was financially supported by the National Natural Science Foundation of China (nos. 21107027 and 21177044) and Innovation Projects of Huazhong University of Science and Technology (no. 2014QN109). References [1] L.M. Yim, S. Taniyasu, L.W.Y. Yeung, G. Lu, L. Jin, Y. Yang, P.K.S. Lam, K. Kannan, N. Yamashita, Perfluorinated compounds in tap water from China and several other countries, Environ. Sci. Technol. 43 (2009) 4824. [2] R. Ma, K. Shih, Perfluorochemicals in wastewater treatment plants and sediments in Hong Kong, Environ. Pollut. 158 (2010) 1354. [3] H.W. Sun, A.C. Gerecke, W. Giger, A.C. Alder, Long-chain perfluorinated chemicals in digested sewage sludges in Switzerland, Environ. Pollut. 159 (2011) 654. [4] M.K. So, S. Taniyasu, N. Yamashita, J.P. Giesy, J. Zheng, Z. Fang, S.H. Im, Paul K.S. Lam, Perfluorinated compounds in coastal waters of Hong Kong, South China, and Korea, Environ. Sci. Technol. 38 (2004) 4056. [5] K. Yanase, M. Eda, H. Kamiya, K. Omori, K. Kamiya, Method of Recovering Fluorochemical Emulsifying Agent, PCT International Patent Application. WO 03/066533 A1, 2003. [6] K. Saito, E. Uemura, A. Ishizaki, H. Kataoka, Determination of perfluorooctanoic acid and perfluorooctane sulfonate by automated in-tube solid-phase microextraction coupled with liquid chromatography-mass spectrometry, Anal. Chim. Acta 658 (2010) 141. [7] W.-L. Liu, Y.-C. Ko, B.-H. Hwang, Z.-G. Li, T.C.-C. Yang, M.-R. Lee, Determination of perfluorocarboxylic acids in water by ion-pair dispersive liquid-liquid microextraction and gas chromatography-tandem mass spectrometry with injection port derivatization, Anal. Chim. Acta 726 (2012) 28. [8] W.M. Henderson, E.J. Weber, S.E. Duirk, J.W. Washington, M.A. Smith, Quantification of fluorotelomer-based chemicals in mammalian matrices by monitoring perfluoroalkyl chain fragments with GC/MS, J. Chromatogr. B 846 (2007) 155. [9] V. Dufkováa, R. Èabalaa, D. Maradováa, M. Štíchab, A fast derivatization procedure for gas chromatographic analysis of perfluorinated organic acids, J. Chromatogr. A 1216 (2009) 8659.
381
[10] B.F. Scott, C.A. Moody, C. Spencer, J.M. Small, D.C.G. Muir, S.A. Mabury, Analysis for perfluorocarboxylic acids/anions in surface waters and precipitation using GC–MS and analysis of PFOA from large-volume samples, Environ. Sci. Technol. 40 (2006) 6405. [11] M. Trojanowicz, M. Koc, Recent developments in methods for analysis of perfluorinated persistent pollutants, Microchim. Acta 180 (2013) 957. [12] T. Ohya, N. Kudo, E. Suzuki, Y. Kawashima, Determination of perfluorinated carboxylic acids in biological samples by high-performance liquid chromatography, J. Chromatogr. B 720 (1998) 1. [13] E. Poboży, E. Król, L. Wójcik, M. Wachowicz, M. Trojanowicz, HPLC determination of perfluorinated carboxylic acids with fluorescence detection, Microchim. Acta 172 (2011) 409. [14] S. Valsecchi, M. Rusconi, S. Polesello, Determination of perfluorinated compounds in aquatic organisms: a review, Anal. Bioanal. Chem. 405 (2013) 143. [15] M. Takayose, K. Akamatsu, H. Nawafune, T. Murashima, J. Matsui, Colorimetric detection of perfluorooctanoic acid (PFOA) utilizing polystyrene-modified gold nanoparticles, Anal. Lett. 45 (2012) 2856. [16] A.K. Keshari, P.K. Singh, V. Parashar, A.C. Pandey, Temperature induced size selective synthesis of hybrid CdS/pepsin nanocrystals and their photoluminescence investigation, J. Lumin. 130 (2010) 315. [17] H. Zhang, Z. Zhou, B. Yang, M.Y. Gao, The influence of carboxyl groups on the photoluminescence of mercaptocarboxylic acid-stabilized CdTe nanoparticles, J. Phys. Chem. B 107 (2003) 8. [18] M.S. Hosseini, H. Jahanbani, Synthesis of CdS nanoparticles quantum dots capped by 2,20 -dithiodibenzoic acid and study of its interaction with some transition metal ions, J. Lumin. 140 (2013) 65. [19] L.L. Zi, Y. Huang, Z.Y. Yan, S.H. Liao, Thioglycolic acid-capped CuInS2/ZnS quantum dots as fl uorescent probe for cobalt ion detection, J. Lumin. 148 (2014) 359. [20] S. Wageh, A.A. Higazy, A.S. Hassouna, Optical properties and thermal degradation of CdSe capped with 3-mercaptopropionic acid, J. Mater. Sci.—Mater. Electron 24 (2013) 3049. [21] M. Koneswaran, R. Narayanaswamy, Mercaptoacetic acid capped CdS quantum dots as fluorescence single shot probe for mercury(II), Sens. Actuators, B 139 (2009) 91. [22] R. Tang, H. Lee, S. Achilefu, Induction of pH sensitivity on the fluorescence lifetime of quantum dots by NIR fluorescent dyes, J. Am. Chem. Soc. 134 (2012) 4545. [23] W.R. Ware, Oxygen quenching of fluorescence in solution, an experimental study of the diffusion process, J. Phys. Chem. 66 (1962) 455. [24] B.Y. Ren, Z. Tong, X.X. Liu, C.Y. Wang, F. Zeng, Mesomorphous structure and properties of non-equimolar complexes of poly(ethylenimine) and perfluorooctanoic acid, Langmuir 20 (2004) 10737. [25] S.A. Sukhishvili, S. Granick, Layered, erasable polymer multilayers formed by hydrogen-bonded sequential self-assembly, Macromolecules 35 (2002) 301.
Contents lists available at ScienceDirect
Journal of Luminescence journal homepage: www.elsevier.com/locate/jlumin
Full Length Article
Rapid fluorometric determination of perfluorooctanoic acid by its quenching effect on the fluorescence of quantum dots Qi Liu, Aizhen Huang, Nan Wang n, Guan Zheng, Lihua Zhu College of Chemistry and Chemical Engineering, Huazhong University of Science and Technology, Wuhan 430074, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 27 August 2014 Received in revised form 26 December 2014 Accepted 17 January 2015 Available online 28 January 2015
Analysis of perfluorooctanoic acid (PFOA) usually requires a combination of high-performance liquid chromatography and mass spectrometry, which is expensive and time-consuming. In the present work, water-soluble CdS quantum dots (QDs) were employed to develop a simple and rapid fluorometric method for the determination of PFOA. Strongly fluorescent CdS QDs were prepared by using 3-mercaptopropionic acid (MPA) as a stabilizer. It was observed that PFOA strongly quenched the fluorescence emission of the MPA-CdS QDs because PFOA promotes the aggregation of MPA-CdS QDs through a fluorine–fluorine affinity interaction. Under optimum conditions, the fluorescence intensity of MPA-CdS QDs was observed to decrease linearly with an increase in the concentration of PFOA from 0.5 to 40 μmol L 1, with a limit of detection of 0.3 μmol L 1. This new method was successfully implemented for the analysis of PFOA-spiked textile samples, with recoveries ranging from 95% to 113%. & 2015 Elsevier B.V. All rights reserved.
Keywords: Perfluorooctanoic acid CdS quantum dots Fluorescence Quenching
1. Introduction Perfluorocarboxylic acids (PFCAs) have been widely used as emulsifying agents in fluoropolymer manufacturing and as surfactants in paints, lubricants, photolithography, polishers, food packaging and fire-fighting foams. Within the PFCA group, perfluorooctanoic acid (C7F15COOH, PFOA) is the most commonly detected compound for two reasons: PFOA finds use in numerous applications and PFOA is a stable degradation product of precursor perfluorinated chemicals. PFOA has been widely found in sediment, municipal wastewater, coastal water, and even tap water [1–3]. It was reported that PFOA was detectable in the coastal seawaters of the Pearl River Delta, including the South China Sea, and in Korea, with concentration ranges of 0.24 to 16 and 0.24 to 320 pg L 1, respectively [4]. Much higher concentrations were found in the wastewaters of factories involved in the manufacture or use of PFOA, e.g., concentrations of up to 3.35 mmol L 1 were detected in the untreated wastewater from a fluoropolymer manufacturing plant [5]. PFOA exhibits bioaccumulation in wildlife and humans, and it is potentially carcinogenic. Therefore, it is important to develop analytical methods for the determination of PFOA. The main analytical methods used for PFOA determination are gas chromatography (GC) and high-performance liquid chromatography (HPLC). Due to the high polarity of PFOA, the direct injection of PFOA into a GC system will result in rather severe tailing of the peaks. Therefore, GC determination of PFOA requires appropriate
n
Corresponding author. Tel.: þ 86 27 87543032; fax: þ 86 27 87543632. E-mail address: [email protected] (N. Wang).
http://dx.doi.org/10.1016/j.jlumin.2015.01.045 0022-2313/& 2015 Elsevier B.V. All rights reserved.
derivatizations [6], which can be carried out with tetrabutylammonium hydrogen sulfate [7], diazomethane [8], isobutyl chloroformate [9] or 2,4-difluoroaniline [10]. Due to a lack of chromophores, PFOA is not easily amenable to traditional HPLC methodologies. Among the reported HPLC methods for PFOA determination, mass spectrometry (MS) detection is employed most frequently. However, HPLC–MS requires expensive instrumentation, a professional operator, complicated sample pretreatment and considerable analysis time [11]. To develop new HPLC methods for simpler detection, Ohya et al. derivatized PFCAs by using laboratory-synthesized 3-bromoacetyl7-methoxycoumarin for HPLC analysis with fluorescence detection [12]. Based on the use of the commercially available fluorophore 3-bromoacetyl coumarin as a derivatization reagent, Poboży developed another HPLC method with fluorimetric detection [13]. However, the derivatization procedures are time-consuming and the formed derivatives exhibit limited stability, producing a substantial source of uncertainty in PFOA analysis [14]. Recently, Takayose et al. developed a colorimetric method for the detection of PFOA utilizing polystyrene-modified gold nanoparticles (Au NPs) [15]. PFOA with carboxylate groups can be bound to the surface of Au NPs, whereas fluorine–fluorine interactions between PFOA-modified Au NPs encourage interparticle aggregation, causing the color of Au NP suspensions to change from red to blue-purple. However, this colorimetric sensor has very limited sensitivity, and it cannot clearly detect PFOA at concentrations below 250 μmol L 1. It is known that fluorescent sensors usually provide a considerably higher sensitivity than colorimetric methods. The purpose of the present work was to develop a rapid fluorometric method for the
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
determination of PFOA. To this end, we considered another group of NPs, quantum dots (QDs). It is worth noting that QDs have sizedependent and interparticle-distance-dependent optical and electronic properties. By varying the size of QDs, the emission wavelength and fluorescence quantum yield can be tuned. For example, differentsized CdS QDs emit blue to near-UV light [16], whereas the fluorescence quantum yield of CdSe QDs can be changed by varying the particle size (i.e., 18% for 2.8 nm, 38% for 3.3 nm, and 2% for 4.0 nm) [17]. In addition, the luminescence intensity of QDs is also sensitive to the interparticle distance; therefore, it is expected that fluorine–fluorine affinity-induced QDs aggregation may cause a change in fluorescence intensity, which makes QDs-based fluorescence analysis of PFOA feasible. To the best of our knowledge, thus far, no reports have employed the photoluminescence of QDs to detect PFOA. Thus, in the present work, the direct involvement of PFOA in the control of aggregation-dispersion of MPA-CdS QDs was studied and a simple and sensitive MPA-CdS QDs system for the fluorescence detection of PFOA was developed.
375
having been washed 5 times, the MPA-CdS QD deposits were dried at 60 1C by a vacuum drier. 2.3. Apparatus Transmission electron microscopy (TEM) images of MPA-CdS QDs were acquired on an FEI TECNAI G2 20U-TWIN at an accelerating voltage of 200 kV. The colloidal solution of the NPs in water was dropped onto a 0.1-nm-thick carbon-coated copper grid, with the excess solution immediately removed. FTIR spectra were recorded on an Equinox 55 Fourier Transform Spectrometer (Bruker Germany). UV–vis absorption spectra were recorded on a Cary 60 UV–vis spectrophotometer (Agilent Technologies). Fluorescence measurements were performed on an FP6200 fluorescence spectrophotometer (Jasco, Japan). pH measurements were conducted using a Sartorius PB-10 pH meter. All optical measurements were performed at room temperature and under ambient conditions. Zeta potentials and hydrodynamic diameter measurements were performed on a Malvern ZEN 3690.
2. Experimental section
2.4. Measurements
2.1. Reagents
Unless specified otherwise, the basic reaction conditions for the determination of PFOA were as follows: 1.0 mL of MPA-CdS QDs stock dispersion (200 mg L 1) and a given volume of PFOA standard stock solution were rapidly mixed in a 10-mL test tube, followed by the addition of 1.0 mL of Na2HPO4–NaOH buffer to maintain the pH at 10. Then, the mixture solution was diluted to 10 mL with water. After 10 min, fluorescence emission spectra were obtained at an excitation wavelength of 365 nm. Both the excitation and emission slit widths were set to 10 nm, and the photomultiplier tube voltage was 400 V.
3-Mercaptopropionic acid (MPA, HS–CH2–CH2–COOH) was obtained from Aladdin Chemical Reagent Company, Cd(ClO4)2 6H2O was obtained from Alfa Aesar Company, and sodium sulfide nonahydrate was obtained (Na2S 9H2O) from Shanghai LingFeng Chemical Reagent Company. PFOA (96%), heptafluorobutyric acid (PFBA, C3F7COOH, 99%) and pentafluoropropionic acid (PFPrA, C2F5COOH, 97%) were purchased from Acros (New Jersey, USA), perfluoroheptanoic acid (PFHpA, C6F13COOH, 98%) from Alfa Aesar (Lancs, UK), undecafluorohexanoic acid (PFHeA, C5F11COOH, 98%) and perfluoropentanoic acid (PFPeA, C4F9COOH, 98%) from Tokyo Kasei Kogyo Co., Ltd. (Tokyo, Japan). Other chemicals were purchased from Shanghai Chemical Reagent Company (Shanghai, China). All reagents were of analytical reagent grade or the highest purity available and directly used without further purification. All aqueous solutions were prepared with double-distilled deionized water. A PFOA stock standard solution (0.5 mmol L 1) was prepared and then stored at 4 1C. A series of pH buffer solutions was prepared by mixing 0.1 mol L 1 NaH2PO4 and 0.1 mol L 1 H3PO4 to maintain a pH range of 3.0 to 4.0, 0.1 mol L 1 NaH2PO4 and 0.1 mol L 1 Na2HPO4 for a pH range of 5.0 to 8.0, and 0.1 mol L 1 Na2HPO4 and 0.1 mol L 1 NaOH for a pH range of 8.0 to 11.0.
Untreated textiles were purchased from a local supermarket. Samples of fabric ( 5 cm 5 cm) were placed into a 50-mL beaker, followed by the addition of 25 mL of water. The samples were then fortified by adding PFOA stock solution, if needed. For example, a recovery study was carried out on samples spiked with 10–30 μmol L 1 PFOA to evaluate the developed method. After the mixture had been left at 40 1C under ultrasound irradiation for 40 min, the sample was filtered under vacuum and washed with water. The collected water sample was then diluted to 100 mL for analysis.
2.2. Preparation of CdS QDs
3. Results and discussion
Water-soluble CdS QDs were prepared with MPA, a capping agent, through the procedures described by Hosseini [18], with some modifications. Cd(ClO4)2 6H2O (0.84 g, 2 mmol) was dissolved in 100 mL of water to obtain a 0.02 mol L 1 Cd2 þ solution. MPA (1.6 mL, 20 mmol) was added to the Cd2 þ solution, resulting in a Cd2 þ /MPA molar ratio of 10:1. After the solution pH had been adjusted to 10 by adding a 2.0 mol L NaOH solution, the obtained mixture was transferred to a 500-mL three-neck, round-bottom flask. Under purging with pure nitrogen gas, the mixture was heated to the boiling point and maintained at that temperature for 30 min. Next, 100 mL of 0.02 mol L 1 freshly prepared Na2S solution was added to the solution to reach a Cd2 þ /S2 molar ratio of 1:1. The bright-yellow colloid was sealed for 2 h of incubation at its boiling point. All steps were carried out under continuous magnetic stirring and purging with pure nitrogen gas. After having been cooled to room temperature, the colloid solution was washed with an equal volume of alcohol and centrifuged to remove excess precursors and contaminants. After
3.1. Characterization of MPA-CdS QDs
2.5. Analysis of textile samples
Hydrophilic MPA molecules were adopted as capping agents to prepare water-soluble CdS QDs. Fig. 1a showed the FTIR spectra of the bare CdS QDs, MPA and MPA-CdS QDs. The characteristic IR absorption bands occurred at 3430 cm 1 (O–H stretching vibration, vOH (adsorbed H2O)), 1623 cm 1 (O–H bending vibration, δOH(H2O)), 1117 cm 1(vS–S), 1013 cm 1(vS ¼ O), and 626 cm 1 (vCd–S) for the bare CdS QDs (curve 1) and at 3595 (vOH of free COOH), 3400– 3099 cm 1 (vOH of COOH with intermolecular hydrogen bonding), 2947 cm 1 (vCH(CH2)), 2661 cm 1 (vS–CH), 2569 cm 1 (vS–H), 1710 (vC ¼ O), 1426 cm 1 (δOH of C–OH) and 1250 cm 1 (vC–O) for MPA (curve 2). In the FTIR spectrum of the MPA-CdS QDs, a broad absorption band at approximately 3376 cm 1 was a characteristic peak of vOH(H2O), the strong absorption peaks at 1565 cm 1 and 1399 cm 1 could be assigned to the asymmetric stretching vibration (vas(COO )) and symmetric stretching vibration (vs(COO )) of
376
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
800
Absorbance (a.u.)
Transmittance (%)
1
2
3
600
0.8
400 0.4
1
2
200
1:CdS 2:MPA 3:MPA-CdS QDs
4000
3000
1800
1200
600
Fluorescence intensity (a.u.)
1.2
0
0.0 200
300
-1
Wavenumber O cm P
400 500 Wavelength (nm)
600
Fig. 1. (a) FTIR spectra of naked CdS QDs, MPA and MPA-CdS QDs. (b) The UV–vis absorption (1) and fluorescence emission (2) spectra of MPA-capped CdS solution (20 mg L 1) at pH 10.0. The fluorescence excitation wavelength was 365 nm and the slit widths were 10 nm for both excitation and emission. 40
(c)
Number (%)
30
20
10
0 0.1
1
10 Size (nm)
100
1000
Fig. 2. TEM images of (a) bare and (b) PFOA-adsorbed MPA-CdS QDs. The PFOA-adsorbed MPA-CdS QDs samples were prepared by pre-immersing MPA-CdS QDs in PFOA solution for 2 h. (c) The hydrodynamic diameter and size distribution of MPA-CdS QDs.
carboxylate anion of MPA, respectively [19–20]. The prominent appearance of the vas(COO ) and vs(COO ) bands and the disappearance of the vC ¼ O band indicated that the carboxyl group of MPA dissociated into carboxylate species and that MPA was successfully coated as carboxylate on the CdS QDs surface [20]. However, the peaks corresponding to vS–H were absent in the spectra of the MPACdS QDs, indicating covalent bonding between the thiol group and the Cd atoms on the QDs surface [21]. The optical properties of MPA-CdS QDs were investigated by both UV–vis absorption spectrometry and fluorescence spectroscopy. As shown in Fig. 1b, the aqueous solution of MPA-CdS QDs exhibited a characteristic absorption profile below 440 nm (curve 1) and a strong fluorescence emission peak at 560 nm when excited by radiation at 365 nm with 10-nm slit widths for both excitation and emission (curve 2). Fig. 2a showed a TEM image of the MPA-CdS QDs. As shown in Fig. 2a, the MPA-CdS QDs were spherical with a primary diameter of 6 nm. After the MPA-CdS QDs had been dispersed in water at pH 10, their hydrodynamic diameter and size distribution were studied by a laser scattering particle size distribution analyzer. The MPA-CdS QDs had a number-weighted hydrodynamic diameter of 8.071.9 nm (Fig. 2c), similar to the particle sizes observed by TEM. This result suggested that the MPA-CdS QDs were highly dispersible in water. 3.2. PFOA-induced quenching of fluorescence of MPA-CdS QDs Fig. 3a illustrated the emission spectra of MPA-CdS QDs in the absence and presence of PFOA. When mixed with PFOA, the fluorescence intensity of MPA-CdS QDs decreased dramatically, and the quenching became stronger with increasing PFOA concentration. For
example, the relative fluorescence intensity (F/F0) of MPA-CdS QDs was 0.971 for 2 μmol L 1 PFOA and 0.200 for 40 μmol L 1 PFOA, where F0 and F are the fluorescence intensity of the MPA-CdS QDs in the absence and presence of PFOA, respectively. This result indicates that MPA-CdS QDs may act as fluorescence probes for sensing PFOA. The fluorescence quenching data were analyzed by both the Stern–Volmer equation and the Lineweaver–Burk equation, which are usually employed to describe the dynamic and static quenching behavior, respectively. These two equations are expressed as follows: F0 ¼ 1 þ K SV ½PFOA F
ð1Þ
ðF 0 F Þ 1 ¼ ðF 0 Þ 1 þ K LB 1 ðF 0 Þ 1 ½PFOA
ð2Þ
where [PFOA] is the concentration of PFOA, KSV is the Stern–Volmer quenching constant, and KLB denotes the binding constant of PFOA and MPA-CdS QDs. The plot of F0/F versus [PFOA] showed a positive deviation (R2 ¼0.8722), whereas the plot of (F0 F) 1 versus [PFOA] showed good linear behavior (R2 ¼0.9953). This finding suggests that the quenching was not initiated by dynamic collision but mainly by the formation of a complex, namely static quenching. From the slope and intercept of the Lineweaver–Burk curve, KLB was calculated to be 1.7 104 L mol 1. The quenching rate constant, Kq, could be calculated according to the following equation: Kq ¼
K LB
τ0
ð3Þ
where τ0 represents the emission lifetime of QDs. Generally, the lifetimes of various QDs fall between one and a few tens of
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
377
0.16
0.12
4
9
400
(F0-F)-1
600
F 0 /F
Fluorescence intensity (a.u.)
5
1
800
3
0.08
2 0.04
200 1
0 450
0.00
0 500
550
600
650
0
10
20
30
0.0
40
c PFOA (µmol L ) -1
Wavelength (nm)
0.5
1.0
c PFOA-1(L
1.5
2.0 -1
µmol )
1
Fig. 3. (a) The fluorescence emission spectra of MPA-CdS solution (20 mg L ) in the absence (1) and presence (2–9) of PFOA at pH 10.0. From 2 to 9: 0.5, 2, 5, 15, 25, 30, 35 and 40 μmol L 1 PFOA. (b) The Stern–Volmer and (c) Lineweaver–Burk curves for analyzing the fluorescence quenching data of MPA-CdS QDs by PFOA at pH 10.0.
1800
1400
1000
600
1
-30
Transmittance (%)
0.4
Zeta potential (mV)
1
2
3 1: MPA-CdS QDs 2: MPA-CdS QDs+PFOA 3: PFOA
1800
1400
1000 -1 Wavenumber O cm P
600
-34
2 -38 PFOA octoic acid
-42 0
20
40
60
c (µmol L ) -1
Fig. 4. (a) FTIR spectra of MPA-CdS QDs before (1) and after (2) the adsorption of PFOA. The FTIR spectrum of PFOA was also given as a reference in (a) (curve 3). (b) Zeta potentials of MPA-CdS QDs solution in the presence of (1) PFOA and (2) octanoic acid at different concentrations.
nanoseconds (approximately 30 ns) [22]. Therefore, Kq ranges from approximately 5.7 1011 to 1.7 1013 L mol 1 s 1, values that are much greater than those (2 1010 L mol 1 s 1), through the maximum scatter collision mechanism [23]. This finding further supports the notion that the fluorescence quenching mainly arises from the formation of a ground-state complex between MPA-CdS QDs and PFOA. To better understand the formation of the complex, we compared the FTIR spectra of the MPA-CdS QDs before and after the addition of PFOA. The FTIR absorption bands of the CF2 and CF3 groups appeared at the same positions for both PFOA and PFOA-adsorbed MPA-CdS QDs, i.e., at 1203 cm 1 and 1147 cm 1 (curves 2 and 3 in Fig. 4a), which supports the binding of PFOA on the MPA-QD surfaces. The complex interaction between PFOA and MPA-CdS QDs could be recognized from the stretching band shift of the carbonyl group of PFOA, as shown in Fig. 4a. The C¼ O stretching bands of pure PFOA powder appeared at 1770 and 1727 cm 1, corresponding to the free carbonyl group and the carbonyl group with intermolecular hydrogen bonding, respectively [24]. In contrast, two vibrational peaks at 1694 and 1659 cm 1 were observed in the spectrum of PFOA-adsorbed
CdS QDs, presumably due to the associated or dimerized carboxylic groups of PFOA [25]. This shift in the C¼O band to lower wavenumbers also suggests that the associated carboxylic groups persisted in the hydrogen-bonded form with MPA. Moreover, the vas(COO ) and vs(COO ) bands in the case of the PFOA-adsorbed MPA-CdS QDs were shifted to lower wavenumbers of 1555 cm 1 and 1415 cm 1, respectively, compared with those of the MPA-CdS QDs (vas(COO ), 1565 cm 1; vs(COO ), 1399 cm 1), further suggesting a strong interaction between the anionic carboxylate group of MPA-CdS QDs and PFOA molecules. The interaction between PFOA and the MPA-CdS QDs was also indicated by the change in the surface charge of the MPA-CdS QDs before and after the addition of PFOA. Fig. 4b shows that the zeta potential of the MPA-CdS QDs depended on the concentration of PFOA. As the concentration of PFOA increased from 0 to 40 μmol L 1, the zeta potential of the MPA-capped CdS QDs at pH 10.0 varied from 41.3 to 30.7 mV. Prior to the addition of PFOA, the MPA-CdS QDs showed a negatively charged surface at pH 10.0 because the MPA-CdS QDs were protected by the carboxylate anion, as indicated by FTIR analysis (Fig. 1a). In the presence of PFOA, PFOA
378
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
can be adsorbed on the MPA-CdS QDs surface through the interaction between the proton of PFOA and the anionic carboxylate group, which thereby offsets the contribution of the carboxylate anions to the net surface charge of the MPA-CdS QDs, resulting in a shift in the zeta potential to less-negative values. However, it should be noted that octanoic acid, which possesses a similar carboxylic structure, only induced a slight change in the zeta potential of MPA-CdS QDs from 41.3 to 37.3 mV. This disparity suggests that there is another essential factor that caused the adsorption-induced variations in the zeta potential of PFOA-adsorbed MPA-CdS QDs. In particular, the fluorine–fluorine interaction between PFOAadsorbed MPA-CdS QDs encourages MPA-CdS QDs to aggregate with each other, causing the negatively charged carboxylate anions of MPA-CdS QDs to become embedded within aggregates. Consequently, the net charge was further reduced. The TEM images in Fig. 2a and b clearly showed that, in the absence of PFOA, welldispersed MPA-CdS QDs tended to aggregate after the addition of PFOA. Takayose also observed a similar phenomenon: the addition of PFOA caused the detachment of the polystyrene layer from the surface of gold nanoparticles, which resulted in interparticle aggregation [15]. Once the aggregation occurred, the collision between MPA-CdS QDs themselves would quench the fluorescence emission. Thus, the principle of detecting PFOA was based on the aggregationinduced fluorescence quenching of MPA-CdS QDs, as shown in Fig. 5. PFOA was bound on the MPA-CdS QDs through the carboxylate group, and consequently, the fluorine–fluorine affinity induced MPA-CdS QDs aggregation, which decreased the fluorescence emission via collisional self-quenching.
3.3. Optimization of assay conditions 3.3.1. Effect of pH The effect of pH was studied over a pH range between 3 and 11. The fluorescence intensity of the MPA-CdS QDs themselves increased with pH and then became nearly constant when the solution pH was greater than 10 (curve 1 in Fig. 6a). An increase in pH promoted the dissociation of the carboxyl groups of the MPACdS QDs to negatively charged carboxylate species, which increased the absolute zeta potential value of the MPA-CdS QDs (Fig. 6b), as shown in Fig. 6b. This increase in absolute potential resulted in an increase in the electrostatic repulsion between particles, promoting better water dispersibility and stronger fluorescence emission of the MPA-CdS QDs. In the presence of PFOA, except for the effect on fluorescence measurements, the solution pH also affected the adsorption of PFOA on the MPA-CdS QDs, which is a primary requirement for fluorescence quenching. In this study, the pH was adjusted in two steps: the pH of PFOA was adjusted first and then the pH values of the mixtures of PFOA and the MPA-CdS QDs were adjusted prior to the fluorescence measurements. One set of samples was prepared by varying the pH of the first step (pH1st) from 3 to 10 and keeping the pH of the second step (pH2nd) fixed at 10.0. It was observed that the value of F/F0 reached an approximately constant value at pH 3.0–4.5 (0.173–0.200) and then increased to 0.352 at pH 7, to 0.635 at pH 8.5, and to 0.994 at pH 10. This difference was observed because the carboxyl group of PFOA provides a proton to an anionic carboxylate-modifying group of CdS QDs, as indicated by FTIR analysis (see Section 3.2). In turn,
Fig. 5. Schematic illustration of the fluorescent analysis of PFOA using MPA-CdS QDs. MPA-CdS QDs are stably dispersed and show strong fluorescence, while the adsorption of PFOA induces MPA-CdS QDs aggregations via fluorous affinity, leading to a significant fluorescence quenching.
1000
1.0
2
0.6 500 0.4 CdS CdS+40µM PFOA 250
1
Zeta potential (mV)
750
F/F0
Fluorsecencec intensity (a.u.)
0
0.8 CdS CdS+40µM PFOA
-15
-30 2
0.2 1
-45
0
0.0 3
5
7 pH
9
11
3
5
7 pH
9
11
Fig. 6. Effects of pH on (a) fluorescence intensity and (b) zeta potential of MPA-capped CdS (20 mg L 1) in the absence (1) and presence (2) of 40 μmol L 1 PFOA.
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
this donation of a proton is due to the dissociation of carboxylic acid groups in both PFOA and MPA-CdS QDs at high pH, which would have been harmful to the interaction between PFOA anions and negatively charged CdS QDs. Another set of samples was prepared by changing pH2nd at a given pH1st of 4.0. As shown in Fig. 6a, PFOAinduced fluorescence quenching was strongest at approximately pH 9 11 (curve 2). Accordingly, the PFOA-induced change in the zeta potential of the MPA-CdS QDs increased roughly with increasing solution pH and then decreased after reaching a maximum at pH 10 (curve 2 in Fig. 6b). Therefore, 4.0 and 10.0 were chosen as the optimum pH levels for the formation of PFOA-adsorbed MPA-CdS QDs first and subsequent fluorescence measurements, respectively.
3.3.2. Effect of MPA-CdS QDs concentration Fig. 7 illustrated the effect of MPA-CdS QDs concentration on fluorescence intensity in the absence and presence of PFOA. With an increased loading of MPA-CdS QDs, the fluorescence intensity of the MPA-CdS QDs increased initially (Fig. 7a). Then, the intensities increased very slightly after the MPA-CdS QDs concentration reached 40 mg L 1 (Fig. 7a), possibly due to the self-quenching of fluorescent MPA-CdS QDs at higher concentrations [21]. In addition, MPA-CdS QDs clusters easily
aggregated at higher concentrations, which caused a reduction in the fluorescence intensity. After the addition of PFOA, the fluorescence intensity of the MPA-CdS QDs decreased, and the fluorescence quenching efficiency was also found to depend on the MPA-CdS QDs concentration. As shown in Fig. 7b, as the MPA-CdS QDs concentration decreased, the F/ F0 value became more sensitive to changes in the PFOA concentration. However, as the MPA-CdS QDs concentration decreased to 10 mg L 1, F/ F0 was not linearly dependent on the PFOA concentration if the PFOA concentration was greater than 20 mg L 1, indicating a decrease in the linear relationship between F/F0 and the PFOA concentration. To reach a compromise between the sensitivity and the linear range of the calibration function, 20 mg L 1 was selected as the optimum concentration of MPA-CdS QDs.
3.3.3. Effect of reaction time By monitoring the fluorescence spectra at different reaction time intervals, we observed that the fluorescence of the MPA-CdS QDs was immediately quenched immediately after the addition of PFOA. As shown in Fig. 8a, the fluorescence intensity of the MPA-CdS QDs in the presence of PFOA reached an approximately constant value within 5 min, which could be maintained for at least 30 min. Thus,
1500
1.0 0.8
1000 F / F0
Fluorescence intensity (a.u.)
379
0.6
cQDs -1
10 mg L -1 20 mg L -1 30 mg L -1 40 mg L -1 50 mg L
0.4
500
0.2 0
0.0 0
20
40 cQDs (mg L-1)
0
60
10
20
30
40
cPFOA(µmol L-1)
Fig. 7. Effect of the QDs concentration on (a) the fluorescence intensity of MPA-CdS QDs and (b) the relative fluorescence intensity (F/F0) in the presence of PFOA.
1.0
0.8
0.8
0.6
0.6
F/F0
F/F0
1.0
0.4
0.4
0.2
0.2
F/F0=-0.0209c+0.9978 R2=0.9950
0.0
0.0 0
10
20 Time (min)
30
0
10
20
30
40
50
cPFOA(µmol L-1 )
Fig. 8. (a) Effect of the reaction time on the relative fluorescence intensity (F/F0) of MPA-CdS (20 mg L 1) in the presence of 40 μmol L 1 PFOA at pH 10. (b) Calibration plot for the PFOA determination with MPA-CdS QDs.
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
100
75
75
0
PF Pr A
0
PF BA
25
PF Pe A
25
PF H xA
50
PF H pA
50
Relative error (%)
100
Fo rm ic ac Pr id op an oi ca cid O ct an oi ca cid M et ha no l Et ha no Is l op ro pa no l Bu ta no l H ex an ol
Relative error (%)
380
Fig. 9. The relative error due to added interferents on the estimated PFOA concentration through the present fluorescence method. The concentration of added PFOA and interfering compound in (a) was 20 and 40 μmol L 1, respectively. Because PFOA account for over 80% of the PFCAs group, the relative concentration ratio of PFOA to other interfering PFCAs was set as 4:1 in (b), i.e., 32 μmol L 1 of PFOA and 8 μmol L 1 of C3-C7 PFCAs.
fluorescence measurements were conducted 10 min after mixing the reactants. Under optimal conditions, the calibration curve for PFOA determination was obtained by establishing the correlation between the relative fluorescence intensity of MPA-CdS QDs (F/F0) and the PFOA concentration (c). Fig. 8b showed that F/F0 was linearly dependent on c over the range of 0.5 to 40 μmol L 1 with a regression equation of F/F0 ¼ 0.0209cþ0.9978 (R2 ¼0.9950). The detection limit was calculated to be 0.3 μmol L 1 (S/N¼3). The limit of detection for this method was far superior to that of the Au NP-based colorimetric method (250 μmol L 1) [15]. Moreover, the relative standard deviations of 11 replicate measurements for 5 and 30 μmol L 1 PFOA were 1.69% and 1.40%, respectively, indicating that the present method provided high reproducibility.
Table 1 Measurements of PFOA spiked textile samples using the proposed fluoremetric method (n ¼3). Sample
Textile I
Textile II
Test no.
1 2 3 1 2 3
PFOA concentration (μmol L 1) Added
Found
0 10.0 30.0 0 10.0 30.0
n.d. 11.3 7 0.5 28.5 7 0.3 n.d. 10.9 7 0.2 29.2 7 0.5
Recovery (%)
– 113.0 7 4.4 95.0 7 1.1 – 109.0 7 1.8 97.3 7 1.7
3.5. Detection of PFOA in real samples 3.4. Effects of possible interfering compounds The effect of potential interfering compounds, including carboxylic acids and alcohols, was also studied. Compared with PFOA at the same concentration of 40 μmol L 1, the tested carboxylic acids and alcohols, including formic acid, propionic acid, octanoic acid, methanol, ethanol, isopropanol, butanol and hexanol, exhibited no significant quenching effect on the fluorescence intensity of MPA-CdS QDs. Moreover, the effect of these compounds on the fluorescence of MPACdS QDs was studied by mixing PFOA and interfering carboxylic acids or alcohols at twice the amount of PFOA. As shown in Fig. 9a, the relative errors due to the added interferents on the estimated PFOA concentration ranged from 4% to 16%. These results suggest that the present method exhibits good selectivity toward PFOA. Furthermore, taking into account that other PFCAs might also be capable of inducing nanoparticle aggregation, several typical C3-C7 PFCAs were also examined as potential interferents. It was found that C3-C7 PFCAs could also quench the fluorescence of MPA-CdS QDs, but the quenching effect was 71–88% that for PFOA at the same molar concentration. In addition, it should be kept in mind that PFOA accounts for over 80% of the PFCA group. When the C3–C7 PFCAs were individually added to the solution containing PFOA at their relatively normal concentration levels in the environment (i.e., no more than one-fourth of the PFOA concentration), the relative error induced was not greater than 11% (Fig. 9b). This result indicates that the MPA-CdS QDs-based fluorometric system proposed herein could be applied to detect PFOA in real samples.
To validate the reliability of the proposed method in determining the concentration of PFOA, two types of textile samples obtained from a local supermarket were used as real sample models. Because no PFOA was detected by the proposed method, known quantities of PFOA were added to the textile suspensions, and then the PFOAspiked textile samples were analyzed according to the procedures described in Section 2.5. As shown in Table 1, the resulting fortification recoveries ranged from 95% to 113%, and the relative standard deviations (RSD) obtained for every three measurements were less than 5%. These results indicate that the proposed technique provides a suitable strategy for determining PFOA in real samples.
4. Conclusions A rapid, simple and sensitive fluorescence sensor was proposed for determining PFOA by the utilization of MPA-CdS QDs. By using TEM, FTIR, zeta potential and fluorescence techniques, we demonstrated that PFOA could be bonded to the surface of MPA-CdS QDs, which resulted in significant interparticle aggregation of MPA-CdS QDs and consequent quenching of their fluorescence emission. Little or no interference from many carboxylic acids and alcohols suggested that both a carboxyl group and fluorocarbon group are required for fluorescence quenching. This method could detect PFOA within approximately 10 min, and it provided a working range of 0.5 to 40 μmol L 1 with a detection limit of 0.3 μmol L 1.
Q. Liu et al. / Journal of Luminescence 161 (2015) 374–381
The analysis of textile samples yielded an acceptable recovery of 95% to 113%. This work also provides an alternative methodology for designing nanoparticle-based fluorescent assays for the determination of PFOA. Acknowledgements This work was financially supported by the National Natural Science Foundation of China (nos. 21107027 and 21177044) and Innovation Projects of Huazhong University of Science and Technology (no. 2014QN109). References [1] L.M. Yim, S. Taniyasu, L.W.Y. Yeung, G. Lu, L. Jin, Y. Yang, P.K.S. Lam, K. Kannan, N. Yamashita, Perfluorinated compounds in tap water from China and several other countries, Environ. Sci. Technol. 43 (2009) 4824. [2] R. Ma, K. Shih, Perfluorochemicals in wastewater treatment plants and sediments in Hong Kong, Environ. Pollut. 158 (2010) 1354. [3] H.W. Sun, A.C. Gerecke, W. Giger, A.C. Alder, Long-chain perfluorinated chemicals in digested sewage sludges in Switzerland, Environ. Pollut. 159 (2011) 654. [4] M.K. So, S. Taniyasu, N. Yamashita, J.P. Giesy, J. Zheng, Z. Fang, S.H. Im, Paul K.S. Lam, Perfluorinated compounds in coastal waters of Hong Kong, South China, and Korea, Environ. Sci. Technol. 38 (2004) 4056. [5] K. Yanase, M. Eda, H. Kamiya, K. Omori, K. Kamiya, Method of Recovering Fluorochemical Emulsifying Agent, PCT International Patent Application. WO 03/066533 A1, 2003. [6] K. Saito, E. Uemura, A. Ishizaki, H. Kataoka, Determination of perfluorooctanoic acid and perfluorooctane sulfonate by automated in-tube solid-phase microextraction coupled with liquid chromatography-mass spectrometry, Anal. Chim. Acta 658 (2010) 141. [7] W.-L. Liu, Y.-C. Ko, B.-H. Hwang, Z.-G. Li, T.C.-C. Yang, M.-R. Lee, Determination of perfluorocarboxylic acids in water by ion-pair dispersive liquid-liquid microextraction and gas chromatography-tandem mass spectrometry with injection port derivatization, Anal. Chim. Acta 726 (2012) 28. [8] W.M. Henderson, E.J. Weber, S.E. Duirk, J.W. Washington, M.A. Smith, Quantification of fluorotelomer-based chemicals in mammalian matrices by monitoring perfluoroalkyl chain fragments with GC/MS, J. Chromatogr. B 846 (2007) 155. [9] V. Dufkováa, R. Èabalaa, D. Maradováa, M. Štíchab, A fast derivatization procedure for gas chromatographic analysis of perfluorinated organic acids, J. Chromatogr. A 1216 (2009) 8659.
381
[10] B.F. Scott, C.A. Moody, C. Spencer, J.M. Small, D.C.G. Muir, S.A. Mabury, Analysis for perfluorocarboxylic acids/anions in surface waters and precipitation using GC–MS and analysis of PFOA from large-volume samples, Environ. Sci. Technol. 40 (2006) 6405. [11] M. Trojanowicz, M. Koc, Recent developments in methods for analysis of perfluorinated persistent pollutants, Microchim. Acta 180 (2013) 957. [12] T. Ohya, N. Kudo, E. Suzuki, Y. Kawashima, Determination of perfluorinated carboxylic acids in biological samples by high-performance liquid chromatography, J. Chromatogr. B 720 (1998) 1. [13] E. Poboży, E. Król, L. Wójcik, M. Wachowicz, M. Trojanowicz, HPLC determination of perfluorinated carboxylic acids with fluorescence detection, Microchim. Acta 172 (2011) 409. [14] S. Valsecchi, M. Rusconi, S. Polesello, Determination of perfluorinated compounds in aquatic organisms: a review, Anal. Bioanal. Chem. 405 (2013) 143. [15] M. Takayose, K. Akamatsu, H. Nawafune, T. Murashima, J. Matsui, Colorimetric detection of perfluorooctanoic acid (PFOA) utilizing polystyrene-modified gold nanoparticles, Anal. Lett. 45 (2012) 2856. [16] A.K. Keshari, P.K. Singh, V. Parashar, A.C. Pandey, Temperature induced size selective synthesis of hybrid CdS/pepsin nanocrystals and their photoluminescence investigation, J. Lumin. 130 (2010) 315. [17] H. Zhang, Z. Zhou, B. Yang, M.Y. Gao, The influence of carboxyl groups on the photoluminescence of mercaptocarboxylic acid-stabilized CdTe nanoparticles, J. Phys. Chem. B 107 (2003) 8. [18] M.S. Hosseini, H. Jahanbani, Synthesis of CdS nanoparticles quantum dots capped by 2,20 -dithiodibenzoic acid and study of its interaction with some transition metal ions, J. Lumin. 140 (2013) 65. [19] L.L. Zi, Y. Huang, Z.Y. Yan, S.H. Liao, Thioglycolic acid-capped CuInS2/ZnS quantum dots as fl uorescent probe for cobalt ion detection, J. Lumin. 148 (2014) 359. [20] S. Wageh, A.A. Higazy, A.S. Hassouna, Optical properties and thermal degradation of CdSe capped with 3-mercaptopropionic acid, J. Mater. Sci.—Mater. Electron 24 (2013) 3049. [21] M. Koneswaran, R. Narayanaswamy, Mercaptoacetic acid capped CdS quantum dots as fluorescence single shot probe for mercury(II), Sens. Actuators, B 139 (2009) 91. [22] R. Tang, H. Lee, S. Achilefu, Induction of pH sensitivity on the fluorescence lifetime of quantum dots by NIR fluorescent dyes, J. Am. Chem. Soc. 134 (2012) 4545. [23] W.R. Ware, Oxygen quenching of fluorescence in solution, an experimental study of the diffusion process, J. Phys. Chem. 66 (1962) 455. [24] B.Y. Ren, Z. Tong, X.X. Liu, C.Y. Wang, F. Zeng, Mesomorphous structure and properties of non-equimolar complexes of poly(ethylenimine) and perfluorooctanoic acid, Langmuir 20 (2004) 10737. [25] S.A. Sukhishvili, S. Granick, Layered, erasable polymer multilayers formed by hydrogen-bonded sequential self-assembly, Macromolecules 35 (2002) 301.