Journal of Chromatography A, 1348 (2014) 80–86
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Development of a polymeric ionic liquid coating for direct-immersion solid-phase microextraction using polyhedral oligomeric silsesquioxane as cross-linker Chunyan Chen, Xiaotong Liang, Jianping Wang, Ying Zou, Huiping Hu, Qingyun Cai ∗ , Shouzhuo Yao ∗ State Key Laboratory of Chemo/Biosensing & Chemometrics, College of Chemistry & Chemical Engineering, Hunan University, Changsha 410082, China
a r t i c l e
i n f o
Article history: Received 19 January 2014 Received in revised form 21 April 2014 Accepted 29 April 2014 Available online 9 May 2014 Keywords: Polymeric ionic liquid Solid phase microextraction Polyhedral oligomeric silsesquioxane Perfluorinated compounds
a b s t r a c t A novel solid-phase microextraction (SPME) fiber was developed by chemical binding of a crosslinked polymeric ionic liquid (PIL) on the surface of an anodized Ti wire, and was applied in direct-immersion mode for the extraction of perfluorinated compounds (PFCs) from water samples coupled with high performance liquid chromatography–tandem mass spectrometry analysis. The PIL coatings were synthesized by using 1-vinyl-3-hexylimidazolium hexafluorophosphate as monomer and methylacryloyl-substituted polyhedral oligomeric silsesquioxane (POSS) as cross-linker via free radical reaction. The proposed fiber coating exhibited high mechanical stability due to the chemical bonding between the coating and the Ti wire surface. The integration of POSS reagent enhanced the organic solvent resistance of the coating. The parameters affecting the extraction performance of the fiber coating including extraction time, pH of solution, ionic strength and desorption conditions were optimized. The developed PIL-POSS fiber showed good linearity (R < 0.998) between 0.1 and 50 ng mL−1 with method detection limits ranging from 0.005 to 0.08 ng mL−1 depending on the analyte, and with relative standard deviation for single-fiber repeatability and fiber-to-fiber reproducibility less than 8.6% and 9.5%, respectively. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Solid-phase microextraction (SPME), as a solvent free sample preparation technique, was introduced by Pawliszyn and co-workers in the early 1990s [1]. Compared to conventional sample preparation techniques, SPME possesses many advantages such as rapidity, simplicity and ease of operation, and has been widely applied in food, pharmaceutical, environmental and biological analysis. SPME is composed of a fiber support which is coated with sorbent materials and is based on the distribution of analyte between a sample matrix and a sorbent phase. Remarkably, SPME integrates sampling and sample preparation into one step and can be easily coupled with analytical techniques such as gas chromatography (GC) and high performance liquid chromatography (HPLC). Analytes can be extracted by three different modes including headspace SPME, direct-immersion SPME and membrane-protected SPME. Subsequent to extraction, desorption of analytes from fiber sorbent coating is achieved by thermal or solvent desorption.
∗ Corresponding authors. Tel.: +86 731 88821968. E-mail addresses: [email protected] (Q. Cai), [email protected] (S. Yao). http://dx.doi.org/10.1016/j.chroma.2014.04.098 0021-9673/© 2014 Elsevier B.V. All rights reserved.
Sorbent coating is considered as the heart of SPME technique. The composition and immobilization methods of sorbent coatings play the most important role in providing unique selectivity, high sensitivity and strong robustness for SPME technique. Intensive researches focused on the development of new SPME sorbent coatings have been reported in recent years [2–5]. Ionic liquids (ILs) and polymeric ionic liquids (PILs) based SPME sorbent coatings have attracted increased attentions due to their unique physical and chemical properties. ILs are nonvolatile salts with melting points below 100 ◦ C and possess many advantages such as low vapor pressure, high thermal stability, tunable viscosity and solvation capability. The remarkable feature of ILs is that altering the combination of cations and anions can modulate the physicochemical properties of the resulting ILs, which makes ILs extremely interesting sorbent coating materials for SPME techniques. Liu et al. firstly reported a disposable IL-based coating for headspace SPME via dipcoated procedure [6]. Hsieh et al. utilized a Nafion membrane to modify the support surface to obtain more loading of ILs on the support and more stable ILs coating [7]. However, both of the two methods were tedious and the prepared fibers had to be recoated after each extraction and desorption step. Anderson and co-works firstly introduced and continuously developed PILs as SPME sorbent coatings [8–19]. Compared to IL-based SPME fiber, the PIL-based
C. Chen et al. / J. Chromatogr. A 1348 (2014) 80–86
SPME fiber can be reused and possessed higher thermal stability, good reproducibility, improved durability, and long lifetimes. The PIL-based SPME fibers have been successfully applied in the extraction of various classes of target analytes such as polycyclic aromatic hydrocarbons (PAHs), benzene derivatives, alcohol and amines, estrogens. However, only a few ILs and PILs coated fiber was applied in aqueous matrix using direct immersion extraction mode, which extremely limits the expansion of IL and PIL-based SPME techniques to those compounds which have low volatility or possess high affinity to the aqueous matrix [10,19–21]. Liu et al. fabricated a SPME fiber by polymerization of an IL onto a surface-modified stainless steel wire through covalent bonds [20]. In order to enhance the attachment of the PILs sorbent coating to the support, stainless steel wire was coated with a gold film via replacement reaction between Au and Fe, modified with a self-assembled monolayer through Au S bonds, coated with a silica layer via hydrolysis and polycondensation. Finally, a vinyl-functionalized IL with alkoxy groups was chemically bonded to the silica layer of stainless steel wire to react with a functionalized IL for the preparation of a PILs sorbent coating via polymerization. The support-bonded fiber possessed comparative lifetime of about 40 extractions compared with commercial PDMS fiber in direct immersion extraction mode. Anderson et al. [19] developed highly stable cross-linked PIL-based sorbent coatings by ultraviolet photoinitiated polymerization. The exceptional stability of the prepared fiber benefited from the crosslinking of the PILs copolymer and the chemical bonding of PIL-based sorbent coatings to the etched and vinyl-functionalized fused silica fiber, making the fiber suitable for direct immersion SPME and possess high durability with a lifetime of approximately 90 extractions. Thus, the formation of IL polymer network and the attachment of the PIL-based coating to the substrate via chemical bonds can significantly improve the fiber stability, which makes the application of the fiber to aqueous matrix in direct immersion mode possible. Polyhedral oligomeric silsesquioxane (POSS) is a threedimensional, cage-type silica nanostructure, which is represented by a formula of Rn (SiO1.5 )n (n = 6, 8, 10, . . .), where R can be H or various organic functional groups. Due to the rigid and highly stable inorganic silica core and facile functionalized substituents, POSS has been widely used to fabricate various functional materials serving as additives, catalyst, monomer, cross linker, and so on. It is well known that incorporation of inorganic component to the polymer network generally increases the thermal and mechanical stability of the resulting material. Thus, employing POSS as a cross-linker to prepare PILs-based copolymer hybrid coating can be a promising method to improve the overall stability of IL-based sorbent coating. In this study, a new sorbent coating composed of crosslinked PIL-POSS copolymer was synthetized for the first time. The mixture solution containing 1-vinyl-3-hexylimidazolium hexafluorophosphate monomer and a methylacryloyl-substituted POSS cross-linker was loaded onto the modified fiber surface and then thermally polymerized using 2,2 -azobis-(2-methylpropionitrile) (AIBN) as an initiator. An anodized Ti wire with TiO2 nanostructure on its surface replaced commonly used fragile fused silica fiber to improve the mechanical strength of the fiber and was functionalized to facilitate the copolymerization with the PIL-POSS coating. The incorporation of inorganic component to the polymer network of coating via crosslinking polymerization provides high thermal, chemical and mechanical stability to the resulting coating, which makes the fiber applicable for direct immersion extraction. This study is the first report with respect to the utilization of PIL-based fiber for direct immersion SPME of aqueous samples coupled with HPLC method, which expands the application of PIL-based fiber to wider range of analytes.
81
Table 1 MRM parameters for PFCs by HPLC-ESI-MS/MS in negative ion mode. Analyte
Parent (m/z)
Daughter (m/z)
Dwell (ms)
Collision (V)
PFHA PFOA PFOS PFDA PFDoA PFTA
362.9 413.0 499.0 512.9 613.0 712.9
318.9 368.9 98.9 468.9 569.0 669.0
40 40 40 40 40 40
−10 −10 −40 −10 −20 −20
2. Experimental 2.1. Reagents and materials POSS–methacryl substituted (cage mixture, N = 8, 10, 12, POSS–MA) was purchased from Hybrid Plastics (CA, USA). 1-Vinylimidazole, hexyl bromide, ␥-methacryloxypropyltrimethoxysilane (␥-MAPS), 2,2 -azobis-(2-methylpropionitrile) (AIBN) were purchased from Aladdin (Shanghai, China). Ethanol, ethyl acetate, 2-propanol, acetone, tetrahydrofuran, chloroform and dichloromethane were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China). Ultrapure water was obtained from a Milli-Q water purification system (Millipore, Bedford, MA, USA). HPLC-grade methanol from J&K Tech. Ltd. (Beijing, China) was used for the preparation of mobile phase. Ti wire (˚ 0.20 mm, 99.9% in purity) was purchased from Lihua Non-ferrous Metals Co. Ltd. (Baoji, China). Perfluoroheptanoic acid (PFHA), perfluorooctane sulfonic acid potassium salt (PFOS-K) and perfluorooctanoic acid (PFOA) were obtained from Fluka (Milwaukee, MI, USA). Perfluorodecanoic acid (PFDA), perfluorododecanoic acid (PFDoA) and perfluorotetradecanoic acid (PFTA) were purchased from J&K Tech. Ltd. (Beijing, China). Each compound was dissolved in methanol to make a stock solution at a concentration of 1 mg mL−1 . The stock solutions were stored at 4 ◦ C and diluted to the required concentrations with ultrapure water prior to use. 2.2. HPLC-ESI-MS/MS analysis Analysis of PFCs compounds was performed on an Agilent 1290 HPLC system coupled to an Agilent 6460 Triple Quadrupole (QQQ) mass spectrometer (Agilent Technologies, Santa Clara, CA) with an electrospray ionization (ESI) source operating in negative mode. The separation of PFCs was performed using an Unitary C18 column (150 mm × 2.1 mm, particle size 2.8 m; Acchrom Technologies Co. Ltd., Beijing, CN). Acetonitrile and 10 mM ammonium acetate buffer solution were employed as mobile phase. The gradient elution with a flow rate of 0.3 mL min−1 was applied for separation as follows: 0–1 min, maintained in 30% acetonitrile; 1–4 min, increased from 30% acetonitrile to 90% acetonitrile; 4–7 min, maintained in 90% acetonitrile; 7.1 min, decreased from 90% acetonitrile to 30% acetonitrile and maintained until 13 min for equilibrium chromatographic column. For MS/MS analysis, negative ion mode was performed under the following conditions: nebulizer gas, N2 (12 L min−1 ) (35 psi); drying gas, N2 (12 L min−1 , 350 ◦ C); fragmenter voltage, 130 V; capillary voltage, 4000 V. MS detection was performed in multiple reaction monitoring (MRM) mode using the [M−H]− ion as precursor. The parameters of MRM for quantitation of PFCs are listed in Table 1. 2.3. Synthesis of ionic liquid monomer The ionic liquid 1-vinyl-3-hexylimidazolium bromide ([VHIM][Br]) was synthesized according to the report [8]. Briefly, 0.06 mol of 1-vinylimidazole and 0.06 mol of hexyl bromide were mixed in 20 mL of 2-propanol and then reacted at 60 ◦ C for 16 h under constant stirring. The resulting mixture was allowed to cool
82
C. Chen et al. / J. Chromatogr. A 1348 (2014) 80–86
to room temperature and 2-propanol was removed under vacuum. The product was dissolved in 20 mL of water and then washed five times with ethyl acetate. Finally, ethyl acetate was evaporated under vacuum and the IL product was dried under vacuum at 60 ◦ C for two days. The preparation of ionic liquid monomer 1-vinyl-3-hexylimidazolium hexafluorophosphate ([VHIM][PF6 ]) was carried out by slowly adding ammonium hexafluorophosphate to a solution of [VHIM][Br] in water. After stirring overnight, the bromide anion was exchanged with hexafluorophosphate anion by metathesis anion exchange and then phase separation occurred. The lower liquid layer was isolated and washed five times with water to remove residual bromide anion, and finally dried overnight under vacuum to yield [VHIM][PF6 ]. The structure of the purified [VHIM][PF6 ] was identified by 1 H NMR and the results were shown as follows: 1 H NMR (600 MHz, CDCl3 ): 0.873 (s, 3H), 1.303–1.324 (q, 7H), 1.952 (m, 2H), 4.192–4.230 (t, 3H), 5.420–5.432 (m, 1H), 5.750–5.797 (q, 1H), 7.031–7.092 (q, 1H), 7.374 (s, 1H), 7.567 (s, 1H), 8.817 (s, 1H). 2.4. Preparation of PIL-POSS coating Due to the unbreakable property and ease to functionalize, derived Ti wire was used as the support of SPME fiber. The derived anodized Ti wire was prepared according to our previous reports [22,23]. The preparation of PIL-POSS coating was conducted by introduction of prepolymerization solution containing IL monomer, initiator, polymerization solvent and varying amounts of cross linker into glass capillary, where the derived anodized Ti wire was placed, then the glass capillary was sealed, and then thermally initialized free radical polymerization reaction was carried out at 60 ◦ C for 12 h. After reaction, the PIL-POSS coated Ti wire was withdrawn from glass capillary and washed with ethanol to remove residual components. 2.5. SPME procedure SPME experiments were performed in a direct immersion extraction mode. All extractions were carried out using a 25 mL glass vial containing 25 mL of working solution of analytes. SPME fibers were directly exposed to the working solution and extracted analytes under constant stirring rate. The off-line solvent desorption of analytes from fiber coating was accomplished by dipping the fiber in 150 L of methanol for 5 min. Subsequently, solvent desorption process was performed three times to remove any carry over effects of analytes. 10 L of desorption solution was injected for HPLC-ESI-MS/MS analysis. 2.6. Environmental water samples The river water samples were collected from Xiangjiang River (Changsha). The collected water samples were filtered through a 0.45 m cellulose membrane immediately after sampling and stored in amber glass bottles at 4 ◦ C. The collected water samples were analyzed within two weeks. 3. Results and discussion 3.1. Preparation and characterization of the PIL-POSS coating The fabrication of the PIL-POSS fiber was carried out by introducing polymerized solution into a glass capillary via siphoning according to previous report [23]. The prepared fiber can be mechanically withdrawn from the glass capillary by controlling the polymerization conditions. The polymerization
Fig. 1. The extraction performance of the PIL-POSS coating with various POSS content.
parameters including solvent, initiator and polymerization temperature were investigated. In consideration of solubility of POSS cross linker, IL monomer and initiator, varied volumes (25–125 L) of dichloromethane, tetrahydrofuran, and ethanol were used as polymerization solvents to prepare PIL-POSS coating. It was found that the PIL-POSS fiber coating with good mechanical strength can be easily withdrawn from the glass capillary using these three organic solvents when their volume is small. While use of large volume of polymerization solvent, only the PIL-POSS fiber coating using ethanol as polymerization solvent can be easily withdrawn from the glass capillary. The reason may be due to the volume contraction of PIL-POSS copolymer coating during polymerization process. The amount of initiator and polymerization temperature did not show obvious influence on the preparation of PIL-POSS coating. Therefore, 25 L of ethanol was chosen as the polymerization solvent. 3.2. Application of PIL-POSS coating for direct immersion SPME The PIL-based SPME coating that is capable to be applied in direct immersion mode in water sample solution usually need to meet two conditions: one is that the PIL-based coating possesses hydrophobic property and cannot be dissolved during extraction of analytes in aqueous solution; another is that the PIL-based coating must be mechanically stable and strongly bonded to the substrate to avoid swelling or exfliation of coating due to the solution stirring. In this study, a derived anodized Ti wire with strong mechanical strength was used as the fiber substrate. The vinyl groups derived on the anodized Ti wire surface can be covalently bonded with the PIL-POSS coating via free radical polymerization reaction, which leads to more mechanically stable PIL-based coating and can be used in aqueous solution in direct immersion mode. According to our previous report [23], introduction of POSS cross linker to the polymerization system can increase the crosslinking degree of the resulting polymer, then decrease solubility of the resulting polymer in water or organic solvent and therefore reduce swelling of polymer. The effect of POSS content on the structure and extraction performance of PIL-POSS coating was investigated. A series of PIL-POSS coatings were prepared with varied POSS content ranging from 15 mg to 150 mg, and applied to extract PFCs from aqueous sample solutions in direct immersion mode. The results are shown in Fig. 1. It is obvious that the extraction efficiency increased with increasing the POSS content, then reached the maximum when POSS content reached 100 mg, and finally decreased at 150 mg. It can be expected that the crosslinking degree increased as the POSS content increased, leading to a three-dimensional network structure and enhanced extraction efficiency. However, the proportion
C. Chen et al. / J. Chromatogr. A 1348 (2014) 80–86
of IL monomer decreased as the POSS content increased, which could reduce amount of functional groups on the coating surface and then decrease extraction efficiency. So, 100 mg of POSS was used to prepare PIL-POSS coating. Organic solvent resistance was an important property for the PIL-POSS coating because organic solvent desorption step was used to desorb analytes from the PIL-POSS coating. In order to investigate the stability of PIL-POSS coating in organic solvent, a series of PIL-POSS coating with different POSS content were immersed into acetonitrile for 30 min. It was found that volume contraction occurred for the PIL-POSS coating with samll POSS content (15 mg and 30 mg) after immersion, accompanied with coating cracking and damage. When POSS content increased, the PIL-POSS coating remained integrated and no cracking phenomenon took place. Fig. 2 shows the SEM images of the surface microstructure of PIL-POSS coating (with 15 mg and 100 mg of POSS, respectively) before and after immersion in acetonitrile. The PIL-POSS coating with 15 mg of POSS cracked after immersion of acetonitrile, while the PIL-POSS coating with 100 mg of POSS remained homogeneous and integrated. This phenomenon indicated that increase of crosslinking degree can increase the rigidity in polymer coating and the organic solvent resistance. The chemical composition of the PIL-POSS coating was characterized by FTIR spectrum. As shown in Fig. 3, the absorption peak at 3167 cm−1 assigned to the C H stretching vibration minished, which indicated the reduction of C C bonds due to the formation of PIL-POSS copolymer. The absorption peaks at 2969 and 2874 cm−1 were ascribed to C H stretching vibration of methyl and methylene. The absorption peaks at 1735 and 1121 cm−1 were assigned to Si O Si stretching vibration, which suggested successful incorporation of POSS into PIL-based coating. 3.3. Optimization of conditions for SPME 3.3.1. Desorption conditions Desorption conditions including desorption solvent and desorption time were usually optimized to obtain a maximum desorption of the analytes from the fiber coating, and then to improve the sensitivity of SPME method. Methanol and acetonitrile, which were commonly used to compose mobile phase in reversed phase chromatography, were chosen as desorption solvents and their desorption efficiencies were compared. The result is shown in Fig. 4A. The desorption efficiency using methanol was obviously better than that using acetonitrile. The reason may be related to the molecular structures of PFCs. PFCs have polar functional groups, which may lead to better solubility in methanol than in acetonitrile because of the formation of hydrogen bonds between PFCs and methanol. So, methanol was chosen as desorption solvent. Fig. 4B shows the influence of desorption time on the desorption efficiency. As can be seen, there is no obvious change for desorption efficiency in the studied desorption time. Therefore, to accelerate the analysis speed, 5 min was chosen as desorption time. 3.3.2. Extraction time SPME is an equilibrium-based technique, the extracted amounts of target analytes are related to the extraction time. Extraction time profiles for the PIL-POSS coating in water were obtained by extracting the sample solution containing the studied analytes for various time intervals in the direct immersion mode. As can be seen from Fig. 5, the extraction efficiencies (expressed as peak area) gradually increase with an increase of the extraction time. For most of PFCs, reaching equilibration required extraction time longer than 60 min. In SPME, quantification of extracted analytes can be performed in non-equilibrium state as long as the extraction conditions remain constant. Taking the extraction efficiency and sensitivity of the
83
SPME method into consideration, 60 min was chosen as the extraction time for the subsequent experiments. 3.3.3. pH of sample solution The pH of sample solution usually has great influence on the extraction of polar compound, because it can alter the dissociation state of polar compound, making compound remaining in ionic form or neutral form, and then impact the extraction efficiency of sorbent coating. The effect of sample solution pH on the extraction efficiency was investigated in the pH range of 2–8, and the result is shown in Fig. 6. The extraction efficiency is pH-dependent, which decreases with the increase of sample solution pH. Due to the small pKa of the investigated PFCs, they are mainly present in anionic form in water. The PIL-POSS coating, which is mainly composed of alkylimidazolium cation, exhibits hydrophobic property and electropositivity, and can adsorb analytes through hydrophobic interaction and electrostatic interaction. The charge of PIL-POSS coating increases in a more acidic solution, then the electrostatic interaction between coating and analyte improves, and finally the extracted amount of analytes increases. Therefore, pH 2.0 was chosen as sample solution pH in the following experiments. 3.3.4. Ionic strength of sample solution In general, addition of salts to the solution can affect the extraction efficiency of sorbent coating by two ways: one is to affect the interface property between sorbent coating and sample solution, and then affect the partition coefficient between analytes and sorbent coating; and another is to decrease the solubility of organic compound in water via the salting-out effect. With regard to PFCs, a salting-out effect was reported in environmental waters, in which the partitioning of PFCs between water and particle increased with the increasing in water salinity [24]. For the direct immersion SPME, decreasing the solubility of target analytes in solution via saltingout effect tends to promote the partitioning of target analytes to the fiber coating. In order to investigate the effect of ionic strength on the extraction efficiency using the PIL-POSS coating, a series of experiments were carried out with a salt content ranging from 0% to 30% (w/v). As shown in Fig. 7, increasing the ionic strength of solution enhances the extraction of PFCs by the PIL-POSS coating. There may be two ways that affect the adsorption behavior of anionic PFCs on the positive charge PIL-POSS coating. One is the electrostatic interaction between negative charge PFCs and positive charge PIL-POSS coating. The double electric layer formed around the PIL-POSS coating could be compressed when ionic strength of solution increased, then weaken the electrostatic interaction between negative charge PFCs and positive charge PIL-POSS coating due to decrease of electric potential, and decrease the extraction efficiency of PIL-POSS coating. Another is electrostatic repulsion between anionic PFCs around the PIL-POSS coating surface, which could be restrained when ionic strength of solution increased, so the extraction efficiency of PIL-POSS coating enhanced. In neutral solution, the electrostatic repulsion between negative charge PFCs may play a more important role than electrostatic attraction interaction between negative charge PFCs and positive charge PIL-POSS coating, so the extraction efficiency increased with an increase of ionic strength of solution. Thus, 25% of salt was added to solution in subsequent experiments. 3.4. Analytical performance Calibration curves were constructed for the PIL-POSS coating by direct immersion SPME of PFCs under the optimized conditions. Table 2 lists the figures of merit for the proposed fiber coating. The linearity of the calibration curves is good in the range of 0.1–50 ng mL−1 for most of PFCs with correlation coefficients (R) varying from 0.994 to 0.998. Limits of detection (LODs)
84
C. Chen et al. / J. Chromatogr. A 1348 (2014) 80–86
Fig. 2. SEM images of the surface structure of PIL-POSS coating with different POSS content before (a, c) and after (b, d) immersion in acetonitrile.
Fig. 3. FTIR spectra of PIL-POSS coating (a), POSS (b) and ionic liquid monomer [VHIM][PF6 ].
Fig. 4. The effects of desorption solvent (A) and desorption time (B) on the desorption efficiency of PIL-POSS coating.
C. Chen et al. / J. Chromatogr. A 1348 (2014) 80–86
85
Table 2 Analytical figures of merit for PFCs by SPME-HPLC-ESI-MS/MS. Analytes
PFHA PFOA PFOS PFDA PFDoA PFTA
Linear ranges (ng mL−1 )
0.40–50 0.20–50 0.40–50 0.10–50 0.10–50 0.20–50
Correlation coefficient (R)
0.998 0.995 0.994 0.996 0.998 0.997
Single fiber repeatability (n = 5, %)
4.5 7.6 5.3 5.7 7.0 8.6
LODs (ng mL−1 )
Fiber-to-fiber reproducibility (n = 5, %)
5.6 9.2 5.9 6.7 8.4 9.5
Proposed method
Ion-pair SPME (GC–MS) [25]
Ion-pair DLLME (GC–MS/MS) [26]
0.04 0.08 0.08 0.005 0.005 0.008
0.75 0.1 – 0.02 – –
0.042 0.051 – 0.037 – –
Fig. 5. The effect of extraction time on the extraction efficiency of PIL-POSS coating.
Fig. 7. The effect of ionic strength on the extraction efficiency of PIL-POSS coating.
were calculated based on three times the baseline noise varying from 0.005 to 0.08 ng mL−1 . Some of them are relatively lower than other reported methods shown in Table 2. The precision was obtained by performing five consecutive extractions to the sample solution with a concentration of 10 ng mL−1 for all analytes. The RSDs for single fiber repeatability and fiber to fiber reproducibility ranged from 4.5% to 8.6% and 5.6–9.5%, respectively. In addition, the lifetime of the PIL-POSS fiber was approximately 80 times for extraction/desorption.
Table 3 Analytical results for determination of PFCs in real water samples. Analytes
Found (ng mL−1 )
RSD (%)
Spiked (ng mL−1 )
Recovery (%)
RSD (%)
PFHA
nda
–
PFOA
b
nq
8.9
PFOS
nqb
6.5
PFDA
nd
–
PFDoA
nd
–
PFTA
nd
–
0.5 5 0.5 5 0.5 5 0.5 5 0.5 5 0.5 5
88 93 95 98 103 99 90 92 89 95 86 91
7.3 6.2 5.4 6.9 4.8 7.2 8.5 9.6 5.5 6.8 5.7 4.8
3.5. Application to real sample The applicability of the proposed PIL-POSS coating was evaluated by exposing it into real water samples for the extraction of
a b
nd: not detected. nq: detected but not quantified.
PFCs in direct immersion mode. Recovery experiments were performed to investigate the effect of sample matrix on the extraction efficiency and precision of the PIL-POSS coating and the results are shown in Table 3. The recoveries obtained for PFCs were in the range of 86–103% with relative standard deviation (RSD) ranging from 4.8% to 9.6%. These results demonstrated that the accuracy of the proposed method for the analysis of PFCs in real water samples with the PIL-POSS coating were quite satisfactory. All of PFCs were found under LODs in real water samples. 4. Conclusions
Fig. 6. The effect of pH on the extraction efficiency of PIL-POSS coating.
A PIL-POSS copolymer coating was fabricated on the vinyl group functionalized anodized Ti wire surface using ionic liquid [VHIM][PF6 ] as monomer and POSS reagent as cross linker. The
86
C. Chen et al. / J. Chromatogr. A 1348 (2014) 80–86
proposed PIL-POSS coating exhibited good mechanical stability and organic solvent stability due to the covalent bonds between PIL-POSS coating and Ti wire substrate as well as the rigidity originated from the increase of crosslinking degree. The PIL-POSS coating was used to extract PFCs in aqueous solution in direct immersion mode, and successfully applied to the analysis of PFCs in environmental samples. Acknowledgements The authors are thankful to the National Basic Research Program of China (2009CB421601) and the Foundation for Innovative Research Groups of NSFC (21221003) for providing financial support. References [1] [2] [3] [4] [5] [6]
C.L. Arthur, J. Pawliszyn, Anal. Chem. 62 (1990) 2145. S. Zhang, Z. Du, G. Li, Anal. Chem. 83 (2011) 7531. X. Cui, Z. Gu, D. Jiang, Y. Li, H. Wang, X. Yan, Anal. Chem. 81 (2009) 9771. Y. Luo, B. Yuan, Q. Yu, Y. Feng, J. Chromatogr. A 1268 (2012) 9. J.-X. Wang, D.-Q. Jiang, Z.-Y. Gu, X.-P. Yan, J. Chromatogr. A 1137 (2006) 8. J. Liu, N. Li, G. Jiang, J. Liu, J.Å. Jönsson, M. Wen, J. Chromatogr. A 1066 (2005) 27.
[7] Y. Hsieh, P. Huang, I.W. Sun, T. Whang, C. Hsu, H. Huang, C. Kuei, Anal. Chim. Acta 557 (2006) 321. [8] F. Zhao, Y. Meng, J.L. Anderson, J. Chromatogr. A 1208 (2008) 1. [9] Q. Zhao, J.C. Wajert, J.L. Anderson, Anal. Chem. 82 (2009) 707. [10] J. López-Darias, V. Pino, J.L. Anderson, C.M. Graham, A.M. Afonso, J. Chromatogr. A 1217 (2010) 1236. [11] J. López-Darias, V. Pino, Y. Meng, J.L. Anderson, A.M. Afonso, J. Chromatogr. A 1217 (2010) 7189. [12] Y. Meng, J.L. Anderson, J. Chromatogr. A 1217 (2010) 6143. [13] Q. Zhao, J.L. Anderson, J. Chromatogr. A 1217 (2010) 4517. [14] Q. Zhao, J.C. Wajert, J.L. Anderson, Anal. Chem. 82 (2010) 707. [15] C.M. Graham, Y. Meng, T. Ho, J.L. Anderson, J. Sep. Sci. 34 (2011) 340. [16] J. López-Darias, J. Anderson, V. Pino, A. Afonso, Anal. Bioanal. Chem. 401 (2011) 2965. [17] Y. Meng, V. Pino, J.L. Anderson, Anal. Chim. Acta 687 (2011) 141. [18] T.D. Ho, M.D. Joshi, M.A. Silver, J.L. Anderson, J. Chromatogr. A 1240 (2012) 29. [19] T.D. Ho, H. Yu, W.T.S. Cole, J.L. Anderson, Anal. Chem. 84 (2012) 9520. [20] L. Pang, J. Liu, J. Chromatogr. A 1230 (2012) 8. [21] Y. Zhang, X. Wang, C. Lin, G. Fang, S. Wang, Chromatographia 75 (2012) 789. [22] C. Chen, S. Yang, D. Pan, Y. Long, Z. Yan, Q. Cai, S. Yao, Analyst 138 (2013) 569. [23] C. Chen, X. Liang, J. Wang, S. Yang, Z. Yan, Q. Cai, S. Yao, Anal. Chim. Acta 792 (2013) 45. [24] J. Jeon, K. Kannan, H.K. Lim, H.B. Moon, J.S. Ra, S.D. Kim, Environ. Sci. Technol. 44 (2010) 2695. [25] R. Alzaga, J.M.A. Bayona, J. Chromatogr. A 1042 (2004) 155. [26] W.-L. Liu, Y.-C. Ko, B.-H. Hwang, Z.-G. Li, T.C.-C. Yang, M.-R. Lee, Anal. Chim. Acta 726 (2012) 28.
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Development of a polymeric ionic liquid coating for direct-immersion solid-phase microextraction using polyhedral oligomeric silsesquioxane as cross-linker Chunyan Chen, Xiaotong Liang, Jianping Wang, Ying Zou, Huiping Hu, Qingyun Cai ∗ , Shouzhuo Yao ∗ State Key Laboratory of Chemo/Biosensing & Chemometrics, College of Chemistry & Chemical Engineering, Hunan University, Changsha 410082, China
a r t i c l e
i n f o
Article history: Received 19 January 2014 Received in revised form 21 April 2014 Accepted 29 April 2014 Available online 9 May 2014 Keywords: Polymeric ionic liquid Solid phase microextraction Polyhedral oligomeric silsesquioxane Perfluorinated compounds
a b s t r a c t A novel solid-phase microextraction (SPME) fiber was developed by chemical binding of a crosslinked polymeric ionic liquid (PIL) on the surface of an anodized Ti wire, and was applied in direct-immersion mode for the extraction of perfluorinated compounds (PFCs) from water samples coupled with high performance liquid chromatography–tandem mass spectrometry analysis. The PIL coatings were synthesized by using 1-vinyl-3-hexylimidazolium hexafluorophosphate as monomer and methylacryloyl-substituted polyhedral oligomeric silsesquioxane (POSS) as cross-linker via free radical reaction. The proposed fiber coating exhibited high mechanical stability due to the chemical bonding between the coating and the Ti wire surface. The integration of POSS reagent enhanced the organic solvent resistance of the coating. The parameters affecting the extraction performance of the fiber coating including extraction time, pH of solution, ionic strength and desorption conditions were optimized. The developed PIL-POSS fiber showed good linearity (R < 0.998) between 0.1 and 50 ng mL−1 with method detection limits ranging from 0.005 to 0.08 ng mL−1 depending on the analyte, and with relative standard deviation for single-fiber repeatability and fiber-to-fiber reproducibility less than 8.6% and 9.5%, respectively. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Solid-phase microextraction (SPME), as a solvent free sample preparation technique, was introduced by Pawliszyn and co-workers in the early 1990s [1]. Compared to conventional sample preparation techniques, SPME possesses many advantages such as rapidity, simplicity and ease of operation, and has been widely applied in food, pharmaceutical, environmental and biological analysis. SPME is composed of a fiber support which is coated with sorbent materials and is based on the distribution of analyte between a sample matrix and a sorbent phase. Remarkably, SPME integrates sampling and sample preparation into one step and can be easily coupled with analytical techniques such as gas chromatography (GC) and high performance liquid chromatography (HPLC). Analytes can be extracted by three different modes including headspace SPME, direct-immersion SPME and membrane-protected SPME. Subsequent to extraction, desorption of analytes from fiber sorbent coating is achieved by thermal or solvent desorption.
∗ Corresponding authors. Tel.: +86 731 88821968. E-mail addresses: [email protected] (Q. Cai), [email protected] (S. Yao). http://dx.doi.org/10.1016/j.chroma.2014.04.098 0021-9673/© 2014 Elsevier B.V. All rights reserved.
Sorbent coating is considered as the heart of SPME technique. The composition and immobilization methods of sorbent coatings play the most important role in providing unique selectivity, high sensitivity and strong robustness for SPME technique. Intensive researches focused on the development of new SPME sorbent coatings have been reported in recent years [2–5]. Ionic liquids (ILs) and polymeric ionic liquids (PILs) based SPME sorbent coatings have attracted increased attentions due to their unique physical and chemical properties. ILs are nonvolatile salts with melting points below 100 ◦ C and possess many advantages such as low vapor pressure, high thermal stability, tunable viscosity and solvation capability. The remarkable feature of ILs is that altering the combination of cations and anions can modulate the physicochemical properties of the resulting ILs, which makes ILs extremely interesting sorbent coating materials for SPME techniques. Liu et al. firstly reported a disposable IL-based coating for headspace SPME via dipcoated procedure [6]. Hsieh et al. utilized a Nafion membrane to modify the support surface to obtain more loading of ILs on the support and more stable ILs coating [7]. However, both of the two methods were tedious and the prepared fibers had to be recoated after each extraction and desorption step. Anderson and co-works firstly introduced and continuously developed PILs as SPME sorbent coatings [8–19]. Compared to IL-based SPME fiber, the PIL-based
C. Chen et al. / J. Chromatogr. A 1348 (2014) 80–86
SPME fiber can be reused and possessed higher thermal stability, good reproducibility, improved durability, and long lifetimes. The PIL-based SPME fibers have been successfully applied in the extraction of various classes of target analytes such as polycyclic aromatic hydrocarbons (PAHs), benzene derivatives, alcohol and amines, estrogens. However, only a few ILs and PILs coated fiber was applied in aqueous matrix using direct immersion extraction mode, which extremely limits the expansion of IL and PIL-based SPME techniques to those compounds which have low volatility or possess high affinity to the aqueous matrix [10,19–21]. Liu et al. fabricated a SPME fiber by polymerization of an IL onto a surface-modified stainless steel wire through covalent bonds [20]. In order to enhance the attachment of the PILs sorbent coating to the support, stainless steel wire was coated with a gold film via replacement reaction between Au and Fe, modified with a self-assembled monolayer through Au S bonds, coated with a silica layer via hydrolysis and polycondensation. Finally, a vinyl-functionalized IL with alkoxy groups was chemically bonded to the silica layer of stainless steel wire to react with a functionalized IL for the preparation of a PILs sorbent coating via polymerization. The support-bonded fiber possessed comparative lifetime of about 40 extractions compared with commercial PDMS fiber in direct immersion extraction mode. Anderson et al. [19] developed highly stable cross-linked PIL-based sorbent coatings by ultraviolet photoinitiated polymerization. The exceptional stability of the prepared fiber benefited from the crosslinking of the PILs copolymer and the chemical bonding of PIL-based sorbent coatings to the etched and vinyl-functionalized fused silica fiber, making the fiber suitable for direct immersion SPME and possess high durability with a lifetime of approximately 90 extractions. Thus, the formation of IL polymer network and the attachment of the PIL-based coating to the substrate via chemical bonds can significantly improve the fiber stability, which makes the application of the fiber to aqueous matrix in direct immersion mode possible. Polyhedral oligomeric silsesquioxane (POSS) is a threedimensional, cage-type silica nanostructure, which is represented by a formula of Rn (SiO1.5 )n (n = 6, 8, 10, . . .), where R can be H or various organic functional groups. Due to the rigid and highly stable inorganic silica core and facile functionalized substituents, POSS has been widely used to fabricate various functional materials serving as additives, catalyst, monomer, cross linker, and so on. It is well known that incorporation of inorganic component to the polymer network generally increases the thermal and mechanical stability of the resulting material. Thus, employing POSS as a cross-linker to prepare PILs-based copolymer hybrid coating can be a promising method to improve the overall stability of IL-based sorbent coating. In this study, a new sorbent coating composed of crosslinked PIL-POSS copolymer was synthetized for the first time. The mixture solution containing 1-vinyl-3-hexylimidazolium hexafluorophosphate monomer and a methylacryloyl-substituted POSS cross-linker was loaded onto the modified fiber surface and then thermally polymerized using 2,2 -azobis-(2-methylpropionitrile) (AIBN) as an initiator. An anodized Ti wire with TiO2 nanostructure on its surface replaced commonly used fragile fused silica fiber to improve the mechanical strength of the fiber and was functionalized to facilitate the copolymerization with the PIL-POSS coating. The incorporation of inorganic component to the polymer network of coating via crosslinking polymerization provides high thermal, chemical and mechanical stability to the resulting coating, which makes the fiber applicable for direct immersion extraction. This study is the first report with respect to the utilization of PIL-based fiber for direct immersion SPME of aqueous samples coupled with HPLC method, which expands the application of PIL-based fiber to wider range of analytes.
81
Table 1 MRM parameters for PFCs by HPLC-ESI-MS/MS in negative ion mode. Analyte
Parent (m/z)
Daughter (m/z)
Dwell (ms)
Collision (V)
PFHA PFOA PFOS PFDA PFDoA PFTA
362.9 413.0 499.0 512.9 613.0 712.9
318.9 368.9 98.9 468.9 569.0 669.0
40 40 40 40 40 40
−10 −10 −40 −10 −20 −20
2. Experimental 2.1. Reagents and materials POSS–methacryl substituted (cage mixture, N = 8, 10, 12, POSS–MA) was purchased from Hybrid Plastics (CA, USA). 1-Vinylimidazole, hexyl bromide, ␥-methacryloxypropyltrimethoxysilane (␥-MAPS), 2,2 -azobis-(2-methylpropionitrile) (AIBN) were purchased from Aladdin (Shanghai, China). Ethanol, ethyl acetate, 2-propanol, acetone, tetrahydrofuran, chloroform and dichloromethane were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China). Ultrapure water was obtained from a Milli-Q water purification system (Millipore, Bedford, MA, USA). HPLC-grade methanol from J&K Tech. Ltd. (Beijing, China) was used for the preparation of mobile phase. Ti wire (˚ 0.20 mm, 99.9% in purity) was purchased from Lihua Non-ferrous Metals Co. Ltd. (Baoji, China). Perfluoroheptanoic acid (PFHA), perfluorooctane sulfonic acid potassium salt (PFOS-K) and perfluorooctanoic acid (PFOA) were obtained from Fluka (Milwaukee, MI, USA). Perfluorodecanoic acid (PFDA), perfluorododecanoic acid (PFDoA) and perfluorotetradecanoic acid (PFTA) were purchased from J&K Tech. Ltd. (Beijing, China). Each compound was dissolved in methanol to make a stock solution at a concentration of 1 mg mL−1 . The stock solutions were stored at 4 ◦ C and diluted to the required concentrations with ultrapure water prior to use. 2.2. HPLC-ESI-MS/MS analysis Analysis of PFCs compounds was performed on an Agilent 1290 HPLC system coupled to an Agilent 6460 Triple Quadrupole (QQQ) mass spectrometer (Agilent Technologies, Santa Clara, CA) with an electrospray ionization (ESI) source operating in negative mode. The separation of PFCs was performed using an Unitary C18 column (150 mm × 2.1 mm, particle size 2.8 m; Acchrom Technologies Co. Ltd., Beijing, CN). Acetonitrile and 10 mM ammonium acetate buffer solution were employed as mobile phase. The gradient elution with a flow rate of 0.3 mL min−1 was applied for separation as follows: 0–1 min, maintained in 30% acetonitrile; 1–4 min, increased from 30% acetonitrile to 90% acetonitrile; 4–7 min, maintained in 90% acetonitrile; 7.1 min, decreased from 90% acetonitrile to 30% acetonitrile and maintained until 13 min for equilibrium chromatographic column. For MS/MS analysis, negative ion mode was performed under the following conditions: nebulizer gas, N2 (12 L min−1 ) (35 psi); drying gas, N2 (12 L min−1 , 350 ◦ C); fragmenter voltage, 130 V; capillary voltage, 4000 V. MS detection was performed in multiple reaction monitoring (MRM) mode using the [M−H]− ion as precursor. The parameters of MRM for quantitation of PFCs are listed in Table 1. 2.3. Synthesis of ionic liquid monomer The ionic liquid 1-vinyl-3-hexylimidazolium bromide ([VHIM][Br]) was synthesized according to the report [8]. Briefly, 0.06 mol of 1-vinylimidazole and 0.06 mol of hexyl bromide were mixed in 20 mL of 2-propanol and then reacted at 60 ◦ C for 16 h under constant stirring. The resulting mixture was allowed to cool
82
C. Chen et al. / J. Chromatogr. A 1348 (2014) 80–86
to room temperature and 2-propanol was removed under vacuum. The product was dissolved in 20 mL of water and then washed five times with ethyl acetate. Finally, ethyl acetate was evaporated under vacuum and the IL product was dried under vacuum at 60 ◦ C for two days. The preparation of ionic liquid monomer 1-vinyl-3-hexylimidazolium hexafluorophosphate ([VHIM][PF6 ]) was carried out by slowly adding ammonium hexafluorophosphate to a solution of [VHIM][Br] in water. After stirring overnight, the bromide anion was exchanged with hexafluorophosphate anion by metathesis anion exchange and then phase separation occurred. The lower liquid layer was isolated and washed five times with water to remove residual bromide anion, and finally dried overnight under vacuum to yield [VHIM][PF6 ]. The structure of the purified [VHIM][PF6 ] was identified by 1 H NMR and the results were shown as follows: 1 H NMR (600 MHz, CDCl3 ): 0.873 (s, 3H), 1.303–1.324 (q, 7H), 1.952 (m, 2H), 4.192–4.230 (t, 3H), 5.420–5.432 (m, 1H), 5.750–5.797 (q, 1H), 7.031–7.092 (q, 1H), 7.374 (s, 1H), 7.567 (s, 1H), 8.817 (s, 1H). 2.4. Preparation of PIL-POSS coating Due to the unbreakable property and ease to functionalize, derived Ti wire was used as the support of SPME fiber. The derived anodized Ti wire was prepared according to our previous reports [22,23]. The preparation of PIL-POSS coating was conducted by introduction of prepolymerization solution containing IL monomer, initiator, polymerization solvent and varying amounts of cross linker into glass capillary, where the derived anodized Ti wire was placed, then the glass capillary was sealed, and then thermally initialized free radical polymerization reaction was carried out at 60 ◦ C for 12 h. After reaction, the PIL-POSS coated Ti wire was withdrawn from glass capillary and washed with ethanol to remove residual components. 2.5. SPME procedure SPME experiments were performed in a direct immersion extraction mode. All extractions were carried out using a 25 mL glass vial containing 25 mL of working solution of analytes. SPME fibers were directly exposed to the working solution and extracted analytes under constant stirring rate. The off-line solvent desorption of analytes from fiber coating was accomplished by dipping the fiber in 150 L of methanol for 5 min. Subsequently, solvent desorption process was performed three times to remove any carry over effects of analytes. 10 L of desorption solution was injected for HPLC-ESI-MS/MS analysis. 2.6. Environmental water samples The river water samples were collected from Xiangjiang River (Changsha). The collected water samples were filtered through a 0.45 m cellulose membrane immediately after sampling and stored in amber glass bottles at 4 ◦ C. The collected water samples were analyzed within two weeks. 3. Results and discussion 3.1. Preparation and characterization of the PIL-POSS coating The fabrication of the PIL-POSS fiber was carried out by introducing polymerized solution into a glass capillary via siphoning according to previous report [23]. The prepared fiber can be mechanically withdrawn from the glass capillary by controlling the polymerization conditions. The polymerization
Fig. 1. The extraction performance of the PIL-POSS coating with various POSS content.
parameters including solvent, initiator and polymerization temperature were investigated. In consideration of solubility of POSS cross linker, IL monomer and initiator, varied volumes (25–125 L) of dichloromethane, tetrahydrofuran, and ethanol were used as polymerization solvents to prepare PIL-POSS coating. It was found that the PIL-POSS fiber coating with good mechanical strength can be easily withdrawn from the glass capillary using these three organic solvents when their volume is small. While use of large volume of polymerization solvent, only the PIL-POSS fiber coating using ethanol as polymerization solvent can be easily withdrawn from the glass capillary. The reason may be due to the volume contraction of PIL-POSS copolymer coating during polymerization process. The amount of initiator and polymerization temperature did not show obvious influence on the preparation of PIL-POSS coating. Therefore, 25 L of ethanol was chosen as the polymerization solvent. 3.2. Application of PIL-POSS coating for direct immersion SPME The PIL-based SPME coating that is capable to be applied in direct immersion mode in water sample solution usually need to meet two conditions: one is that the PIL-based coating possesses hydrophobic property and cannot be dissolved during extraction of analytes in aqueous solution; another is that the PIL-based coating must be mechanically stable and strongly bonded to the substrate to avoid swelling or exfliation of coating due to the solution stirring. In this study, a derived anodized Ti wire with strong mechanical strength was used as the fiber substrate. The vinyl groups derived on the anodized Ti wire surface can be covalently bonded with the PIL-POSS coating via free radical polymerization reaction, which leads to more mechanically stable PIL-based coating and can be used in aqueous solution in direct immersion mode. According to our previous report [23], introduction of POSS cross linker to the polymerization system can increase the crosslinking degree of the resulting polymer, then decrease solubility of the resulting polymer in water or organic solvent and therefore reduce swelling of polymer. The effect of POSS content on the structure and extraction performance of PIL-POSS coating was investigated. A series of PIL-POSS coatings were prepared with varied POSS content ranging from 15 mg to 150 mg, and applied to extract PFCs from aqueous sample solutions in direct immersion mode. The results are shown in Fig. 1. It is obvious that the extraction efficiency increased with increasing the POSS content, then reached the maximum when POSS content reached 100 mg, and finally decreased at 150 mg. It can be expected that the crosslinking degree increased as the POSS content increased, leading to a three-dimensional network structure and enhanced extraction efficiency. However, the proportion
C. Chen et al. / J. Chromatogr. A 1348 (2014) 80–86
of IL monomer decreased as the POSS content increased, which could reduce amount of functional groups on the coating surface and then decrease extraction efficiency. So, 100 mg of POSS was used to prepare PIL-POSS coating. Organic solvent resistance was an important property for the PIL-POSS coating because organic solvent desorption step was used to desorb analytes from the PIL-POSS coating. In order to investigate the stability of PIL-POSS coating in organic solvent, a series of PIL-POSS coating with different POSS content were immersed into acetonitrile for 30 min. It was found that volume contraction occurred for the PIL-POSS coating with samll POSS content (15 mg and 30 mg) after immersion, accompanied with coating cracking and damage. When POSS content increased, the PIL-POSS coating remained integrated and no cracking phenomenon took place. Fig. 2 shows the SEM images of the surface microstructure of PIL-POSS coating (with 15 mg and 100 mg of POSS, respectively) before and after immersion in acetonitrile. The PIL-POSS coating with 15 mg of POSS cracked after immersion of acetonitrile, while the PIL-POSS coating with 100 mg of POSS remained homogeneous and integrated. This phenomenon indicated that increase of crosslinking degree can increase the rigidity in polymer coating and the organic solvent resistance. The chemical composition of the PIL-POSS coating was characterized by FTIR spectrum. As shown in Fig. 3, the absorption peak at 3167 cm−1 assigned to the C H stretching vibration minished, which indicated the reduction of C C bonds due to the formation of PIL-POSS copolymer. The absorption peaks at 2969 and 2874 cm−1 were ascribed to C H stretching vibration of methyl and methylene. The absorption peaks at 1735 and 1121 cm−1 were assigned to Si O Si stretching vibration, which suggested successful incorporation of POSS into PIL-based coating. 3.3. Optimization of conditions for SPME 3.3.1. Desorption conditions Desorption conditions including desorption solvent and desorption time were usually optimized to obtain a maximum desorption of the analytes from the fiber coating, and then to improve the sensitivity of SPME method. Methanol and acetonitrile, which were commonly used to compose mobile phase in reversed phase chromatography, were chosen as desorption solvents and their desorption efficiencies were compared. The result is shown in Fig. 4A. The desorption efficiency using methanol was obviously better than that using acetonitrile. The reason may be related to the molecular structures of PFCs. PFCs have polar functional groups, which may lead to better solubility in methanol than in acetonitrile because of the formation of hydrogen bonds between PFCs and methanol. So, methanol was chosen as desorption solvent. Fig. 4B shows the influence of desorption time on the desorption efficiency. As can be seen, there is no obvious change for desorption efficiency in the studied desorption time. Therefore, to accelerate the analysis speed, 5 min was chosen as desorption time. 3.3.2. Extraction time SPME is an equilibrium-based technique, the extracted amounts of target analytes are related to the extraction time. Extraction time profiles for the PIL-POSS coating in water were obtained by extracting the sample solution containing the studied analytes for various time intervals in the direct immersion mode. As can be seen from Fig. 5, the extraction efficiencies (expressed as peak area) gradually increase with an increase of the extraction time. For most of PFCs, reaching equilibration required extraction time longer than 60 min. In SPME, quantification of extracted analytes can be performed in non-equilibrium state as long as the extraction conditions remain constant. Taking the extraction efficiency and sensitivity of the
83
SPME method into consideration, 60 min was chosen as the extraction time for the subsequent experiments. 3.3.3. pH of sample solution The pH of sample solution usually has great influence on the extraction of polar compound, because it can alter the dissociation state of polar compound, making compound remaining in ionic form or neutral form, and then impact the extraction efficiency of sorbent coating. The effect of sample solution pH on the extraction efficiency was investigated in the pH range of 2–8, and the result is shown in Fig. 6. The extraction efficiency is pH-dependent, which decreases with the increase of sample solution pH. Due to the small pKa of the investigated PFCs, they are mainly present in anionic form in water. The PIL-POSS coating, which is mainly composed of alkylimidazolium cation, exhibits hydrophobic property and electropositivity, and can adsorb analytes through hydrophobic interaction and electrostatic interaction. The charge of PIL-POSS coating increases in a more acidic solution, then the electrostatic interaction between coating and analyte improves, and finally the extracted amount of analytes increases. Therefore, pH 2.0 was chosen as sample solution pH in the following experiments. 3.3.4. Ionic strength of sample solution In general, addition of salts to the solution can affect the extraction efficiency of sorbent coating by two ways: one is to affect the interface property between sorbent coating and sample solution, and then affect the partition coefficient between analytes and sorbent coating; and another is to decrease the solubility of organic compound in water via the salting-out effect. With regard to PFCs, a salting-out effect was reported in environmental waters, in which the partitioning of PFCs between water and particle increased with the increasing in water salinity [24]. For the direct immersion SPME, decreasing the solubility of target analytes in solution via saltingout effect tends to promote the partitioning of target analytes to the fiber coating. In order to investigate the effect of ionic strength on the extraction efficiency using the PIL-POSS coating, a series of experiments were carried out with a salt content ranging from 0% to 30% (w/v). As shown in Fig. 7, increasing the ionic strength of solution enhances the extraction of PFCs by the PIL-POSS coating. There may be two ways that affect the adsorption behavior of anionic PFCs on the positive charge PIL-POSS coating. One is the electrostatic interaction between negative charge PFCs and positive charge PIL-POSS coating. The double electric layer formed around the PIL-POSS coating could be compressed when ionic strength of solution increased, then weaken the electrostatic interaction between negative charge PFCs and positive charge PIL-POSS coating due to decrease of electric potential, and decrease the extraction efficiency of PIL-POSS coating. Another is electrostatic repulsion between anionic PFCs around the PIL-POSS coating surface, which could be restrained when ionic strength of solution increased, so the extraction efficiency of PIL-POSS coating enhanced. In neutral solution, the electrostatic repulsion between negative charge PFCs may play a more important role than electrostatic attraction interaction between negative charge PFCs and positive charge PIL-POSS coating, so the extraction efficiency increased with an increase of ionic strength of solution. Thus, 25% of salt was added to solution in subsequent experiments. 3.4. Analytical performance Calibration curves were constructed for the PIL-POSS coating by direct immersion SPME of PFCs under the optimized conditions. Table 2 lists the figures of merit for the proposed fiber coating. The linearity of the calibration curves is good in the range of 0.1–50 ng mL−1 for most of PFCs with correlation coefficients (R) varying from 0.994 to 0.998. Limits of detection (LODs)
84
C. Chen et al. / J. Chromatogr. A 1348 (2014) 80–86
Fig. 2. SEM images of the surface structure of PIL-POSS coating with different POSS content before (a, c) and after (b, d) immersion in acetonitrile.
Fig. 3. FTIR spectra of PIL-POSS coating (a), POSS (b) and ionic liquid monomer [VHIM][PF6 ].
Fig. 4. The effects of desorption solvent (A) and desorption time (B) on the desorption efficiency of PIL-POSS coating.
C. Chen et al. / J. Chromatogr. A 1348 (2014) 80–86
85
Table 2 Analytical figures of merit for PFCs by SPME-HPLC-ESI-MS/MS. Analytes
PFHA PFOA PFOS PFDA PFDoA PFTA
Linear ranges (ng mL−1 )
0.40–50 0.20–50 0.40–50 0.10–50 0.10–50 0.20–50
Correlation coefficient (R)
0.998 0.995 0.994 0.996 0.998 0.997
Single fiber repeatability (n = 5, %)
4.5 7.6 5.3 5.7 7.0 8.6
LODs (ng mL−1 )
Fiber-to-fiber reproducibility (n = 5, %)
5.6 9.2 5.9 6.7 8.4 9.5
Proposed method
Ion-pair SPME (GC–MS) [25]
Ion-pair DLLME (GC–MS/MS) [26]
0.04 0.08 0.08 0.005 0.005 0.008
0.75 0.1 – 0.02 – –
0.042 0.051 – 0.037 – –
Fig. 5. The effect of extraction time on the extraction efficiency of PIL-POSS coating.
Fig. 7. The effect of ionic strength on the extraction efficiency of PIL-POSS coating.
were calculated based on three times the baseline noise varying from 0.005 to 0.08 ng mL−1 . Some of them are relatively lower than other reported methods shown in Table 2. The precision was obtained by performing five consecutive extractions to the sample solution with a concentration of 10 ng mL−1 for all analytes. The RSDs for single fiber repeatability and fiber to fiber reproducibility ranged from 4.5% to 8.6% and 5.6–9.5%, respectively. In addition, the lifetime of the PIL-POSS fiber was approximately 80 times for extraction/desorption.
Table 3 Analytical results for determination of PFCs in real water samples. Analytes
Found (ng mL−1 )
RSD (%)
Spiked (ng mL−1 )
Recovery (%)
RSD (%)
PFHA
nda
–
PFOA
b
nq
8.9
PFOS
nqb
6.5
PFDA
nd
–
PFDoA
nd
–
PFTA
nd
–
0.5 5 0.5 5 0.5 5 0.5 5 0.5 5 0.5 5
88 93 95 98 103 99 90 92 89 95 86 91
7.3 6.2 5.4 6.9 4.8 7.2 8.5 9.6 5.5 6.8 5.7 4.8
3.5. Application to real sample The applicability of the proposed PIL-POSS coating was evaluated by exposing it into real water samples for the extraction of
a b
nd: not detected. nq: detected but not quantified.
PFCs in direct immersion mode. Recovery experiments were performed to investigate the effect of sample matrix on the extraction efficiency and precision of the PIL-POSS coating and the results are shown in Table 3. The recoveries obtained for PFCs were in the range of 86–103% with relative standard deviation (RSD) ranging from 4.8% to 9.6%. These results demonstrated that the accuracy of the proposed method for the analysis of PFCs in real water samples with the PIL-POSS coating were quite satisfactory. All of PFCs were found under LODs in real water samples. 4. Conclusions
Fig. 6. The effect of pH on the extraction efficiency of PIL-POSS coating.
A PIL-POSS copolymer coating was fabricated on the vinyl group functionalized anodized Ti wire surface using ionic liquid [VHIM][PF6 ] as monomer and POSS reagent as cross linker. The
86
C. Chen et al. / J. Chromatogr. A 1348 (2014) 80–86
proposed PIL-POSS coating exhibited good mechanical stability and organic solvent stability due to the covalent bonds between PIL-POSS coating and Ti wire substrate as well as the rigidity originated from the increase of crosslinking degree. The PIL-POSS coating was used to extract PFCs in aqueous solution in direct immersion mode, and successfully applied to the analysis of PFCs in environmental samples. Acknowledgements The authors are thankful to the National Basic Research Program of China (2009CB421601) and the Foundation for Innovative Research Groups of NSFC (21221003) for providing financial support. References [1] [2] [3] [4] [5] [6]
C.L. Arthur, J. Pawliszyn, Anal. Chem. 62 (1990) 2145. S. Zhang, Z. Du, G. Li, Anal. Chem. 83 (2011) 7531. X. Cui, Z. Gu, D. Jiang, Y. Li, H. Wang, X. Yan, Anal. Chem. 81 (2009) 9771. Y. Luo, B. Yuan, Q. Yu, Y. Feng, J. Chromatogr. A 1268 (2012) 9. J.-X. Wang, D.-Q. Jiang, Z.-Y. Gu, X.-P. Yan, J. Chromatogr. A 1137 (2006) 8. J. Liu, N. Li, G. Jiang, J. Liu, J.Å. Jönsson, M. Wen, J. Chromatogr. A 1066 (2005) 27.
[7] Y. Hsieh, P. Huang, I.W. Sun, T. Whang, C. Hsu, H. Huang, C. Kuei, Anal. Chim. Acta 557 (2006) 321. [8] F. Zhao, Y. Meng, J.L. Anderson, J. Chromatogr. A 1208 (2008) 1. [9] Q. Zhao, J.C. Wajert, J.L. Anderson, Anal. Chem. 82 (2009) 707. [10] J. López-Darias, V. Pino, J.L. Anderson, C.M. Graham, A.M. Afonso, J. Chromatogr. A 1217 (2010) 1236. [11] J. López-Darias, V. Pino, Y. Meng, J.L. Anderson, A.M. Afonso, J. Chromatogr. A 1217 (2010) 7189. [12] Y. Meng, J.L. Anderson, J. Chromatogr. A 1217 (2010) 6143. [13] Q. Zhao, J.L. Anderson, J. Chromatogr. A 1217 (2010) 4517. [14] Q. Zhao, J.C. Wajert, J.L. Anderson, Anal. Chem. 82 (2010) 707. [15] C.M. Graham, Y. Meng, T. Ho, J.L. Anderson, J. Sep. Sci. 34 (2011) 340. [16] J. López-Darias, J. Anderson, V. Pino, A. Afonso, Anal. Bioanal. Chem. 401 (2011) 2965. [17] Y. Meng, V. Pino, J.L. Anderson, Anal. Chim. Acta 687 (2011) 141. [18] T.D. Ho, M.D. Joshi, M.A. Silver, J.L. Anderson, J. Chromatogr. A 1240 (2012) 29. [19] T.D. Ho, H. Yu, W.T.S. Cole, J.L. Anderson, Anal. Chem. 84 (2012) 9520. [20] L. Pang, J. Liu, J. Chromatogr. A 1230 (2012) 8. [21] Y. Zhang, X. Wang, C. Lin, G. Fang, S. Wang, Chromatographia 75 (2012) 789. [22] C. Chen, S. Yang, D. Pan, Y. Long, Z. Yan, Q. Cai, S. Yao, Analyst 138 (2013) 569. [23] C. Chen, X. Liang, J. Wang, S. Yang, Z. Yan, Q. Cai, S. Yao, Anal. Chim. Acta 792 (2013) 45. [24] J. Jeon, K. Kannan, H.K. Lim, H.B. Moon, J.S. Ra, S.D. Kim, Environ. Sci. Technol. 44 (2010) 2695. [25] R. Alzaga, J.M.A. Bayona, J. Chromatogr. A 1042 (2004) 155. [26] W.-L. Liu, Y.-C. Ko, B.-H. Hwang, Z.-G. Li, T.C.-C. Yang, M.-R. Lee, Anal. Chim. Acta 726 (2012) 28.
