Journal of Chromatography A, 1324 (2014) 11–20
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Electrospun modified silica-polyamide nanocomposite as a novel fiber coating Habib Bagheri ∗ , Ali Roostaie Environmental and Bio-Analytical Laboratories, Department of Chemistry, Sharif University of Technology, P.O. Box 11365-9516, Tehran, Iran
a r t i c l e
i n f o
Article history: Received 13 July 2013 Received in revised form 30 October 2013 Accepted 9 November 2013 Available online 17 November 2013 Keywords: Modified silica nanoparticles Chlorobenzenes Electrospun Nanocomposites
a b s t r a c t In the present work, a new solid phase microextraction (SPME) fiber coating based on modified silicapolyamide (PA) nanocomposite was electrospun on a stainless steel wire. Four modified silica-PA nanocomposites together with PA were fabricated by dispersing several typical modified silica nanoparticles in PA polymer solution prior to electrospinning. The surface characteristic of PA nanofibers and modified silica-PA nanocomposites was investigated using scanning electron microscopy (SEM). The homogeneity and the porous surface structure of the modified silica-PA nanocomposites were confirmed by SEM, showing nanofibers diameters lower than 170 nm. The applicability of the new fiber coating was examined by headspace SPME of some selected chlorobenzenes (CBs), as model compounds, from aqueous samples. Subsequently, the extracted analytes were transferred into a gas chromatography (GC) by thermal desorption. Influencing parameters on the morphology of nanocomposites such as type of modified silica nanoparticles and the weight ratio of components were optimized. In addition, effects of different parameters influencing the extraction efficiency including extraction temperature, extraction time, ionic strength, desorption temperature, and desorption time were investigated and optimized. Eventually, the developed method was validated by gas chromatography–mass spectrometry (GC–MS). At the optimum conditions, the relative standard deviation values for a double distilled water spiked with the selected CBs at 100 ng L−1 were 4–12% (n = 3) and the limit of detection for the studied compounds was between 5 and 30 ng L−1 . The calibration curves of analytes were investigated in the range of 50–1000 ng L−1 and correlation coefficients (R2 ) between 0.9897 and 0.9992 were obtained. © 2013 Elsevier B.V. All rights reserved.
1. Introduction SPME is a well-established solvent-free extraction technique in which sampling, preconcentration, and sample introduction are integrated in a single process. Although this technique is predominantly performed on SPME fibers [1,2], the conventional fibers could be broken easily and their coatings are susceptible to stripping when they are exposed to high temperatures and organic solvents. The developments of new fibers are mostly focused on improving the thermal, mechanical, and chemical stabilities, imparting diverse functionalities and polarities, and enhancing the fiber capacity [3–5]. The use of nanocomposites coatings [6–8] and replacing the fragile fused silica substrates by wire metals [9] are mostly being used in recent years. The appropriate selection of the fiber coating is one of the most critical steps in SPME method development. The nature of the coating should be selected based on the chemical properties of the target analytes and affects the overall extraction
∗ Corresponding author. Tel.: +98 21 66005718; fax: +98 21 66012983. E-mail address: [email protected] (H. Bagheri). 0021-9673/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.chroma.2013.11.024
performance, including method sensitivity, selectivity, and reproducibility. The common aspects in the microextraction strategies are mostly focused on miniaturized sorbent amount and the extraction device geometry which often lead to the extraction of small fraction of the analytes from the sample into the extracting phase [10]. One important strategy to improve the sensitivity of the microextraction methods relies on the usage of nanostructured materials in which the interactions between the desired analytes and the sorbent are enhanced. The high surface-to-volume ratio of nanostructured materials provides a great deal of active sites for adsorption, making an appropriate platform for the efficient separation, extraction, and enrichment. Silica nanoparticles (SNPs), due to their chemical inertness, nontoxicity, optical transparency, and excellent thermal stability [11–13], have gained significant attentions. They can be widely used in catalysis [14], chemical process industry [15], removal of metal ions [16], and metal ion preconcentration [17–19] through polymer coatings or other functional groups. Chemical modification of the silica surface makes it possible to combine the mechanical and structural properties of the pure silica particles with the ability of specific chemical and physical interactions. The use of modified silica in stationary phases
12
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
in different chromatographic techniques, supported materials in heterogeneous catalysis and reinforcement agents in composite materials are just a few examples [20]. Large specific surface area and high thermal stability of nanoporous silica materials (e.g., SBA15) have made them to be used as an alternative for SPME coatings. The chemical functionalization of SBA-15 by incorporating organic groups and its capability as a support for the construction of conductive polymer/SBA-15 nanocomposites extends the range of materials applied as SPME coatings [21–23]. The recent studies on polypyrrole/SBA-15 [23] and polyaniline/silica nanocomposite [24] have proven that these can be incorporated as facile coatings for hydrophobic compounds. Organic/inorganic nanocomposites are generally organic polymer composites with inorganic nanoscale building blocks. They combine the advantages of the inorganic material (e.g., rigidity, thermal stability) and the organic polymer (e.g., flexibility, dielectric, ductility, and processability). It is well known that the polymer composites can be fabricated by the incorporation of inorganic reinforcements into the polymer network. The properties of the resulting polymer composites depend on the characteristics, dimensions, and shapes of the inorganic fillers, and also on the interfacial bonding strength [25,26]. It is proposed that with decreasing filler dimension, the specific area of the filler is significantly improved, and in turn it would greatly and effectively improve the transfer of the load between the fillers and the polymer matrix [27]. The inorganic nanofillers were successfully incorporated into the polymeric matrix to strengthen and improve the ductile polymer to be more stiff and resistant for abrasion [28–34]. Recently, it has been shown [35,36] that the electrospinning technique is quite efficient to synthesize a variety of nanofibers coatings on thin stainless steel wires. In methodology, by applying high voltages to a viscous polymeric solution, when the electrostatic repulsive force overcomes the surface tension of the polymeric solution, a charged polymeric jet ejects from the solution and afterwards flies toward the collector and forms the fibrous material with diameters in the scale of nano- to micrometer. The electrospun sorbents were used in SPE [37–40], micro-SPE [41], SPME [42], and membrane extraction [43] of some organic and inorganic species. In this work, for the first time, different amounts of silica nanoparticles and modified silica nanoparticles were embedded in polyamide nanofibers by the electrospinning technique. A rotating stainless steel SPME needle was used as an electrospinning collector and PA nanofibers containing silica nanoparticles were directly formed on the SPME needle. The applicability of this novel SPME coating, prepared only in one step, was assessed for SPME of selected CBs, as model pollutants, from aqueous samples using a homemade SPME unit.
2. Experimental 2.1. Reagents The CB compounds including 1,4-dichlorobenzene (14DCB), 1,2dichlorobenzene (12DCB), 1,2,4-trichlorobenzene (124TCB) and monochlorobenezne (MCB) were purchased from Merck (Darmstadt, Germany). Standard solution (1000 mg L−1 ) of CBs mixture was prepared in HPLC-grade methanol (Merck) and stored in the refrigerator. The working standard solutions were prepared weekly by diluting the standard solution with methanol, and more diluted working solutions were prepared daily by diluting this solution with double distilled water (DDW). Sodium chloride was purchased from Merck. Nylon-6 (N6) was purchased from Kolon Industries Inc. (Seoul, Korea) and formic acid was obtained from Riedel-de
Table 1 Properties of different silica nanoparticles examined in this work. Type of silica nanoparticle
Particle size diameter
Wettability
SiO2 Si-SiO2 Methyl-SiO2
10–50 10–50 10–50
Hydrophilic Neutral Hydrophobic
Table 2 Retention times and selected ions of compounds studied by GC–MS. Compound
Retention time (min)
Selected ions (m/z)
MCB 14DCB 12DCB 124TCB
4 5.5 5.7 6.7
77, 112 146, 148 146, 148 180, 182
Haën (Seelze-Hannover, Germany). All solvents used in this study were of analytical reagent grade or HPLC grade. The different SiO2 nanoparticles with diameter of about 12–50 nm were obtained from Degussa Company (Darmstadt, Germany) (Table 1). Cis-9octadecenoic acid (oleic acid, OA), ethanol and n-hexane were obtained from Merck (Darmstadt, Germany). 2.2. Apparatus A gas chromatograph model Agilent 6820, with a split/splitless injection port and a flame ionization detection system, was used for the optimization process. A homemade glass inlet liner with a total length of 7.7 cm, 6 mm o.d. and 1 mm i.d. was deactivated by trimethylchlorosilane and used. The injector and detector temperatures were set at 200 ◦ C and 290 ◦ C, respectively. For quantitative determination, a Hewlett-Packard (HP, Palo Alto, USA) HP 6890 series GC equipped with a split/splitless injector and a HP 5973 mass-selective detector system were used. The MS was operated in the EI mode (70 eV). For both GC–FID and GC–MS systems, the separation of analytes was carried out using a capillary column HP-5 MS (60 m, 0.25 mm i.d.) with 0.25 m film thickness (Hewlett-Packard, Palo Alto, CA, USA). The carrier gas was helium (99.999%) at a flow rate of 1 mL min−1 . Both instruments were operated in the splitless mode and the split valve was kept closed for 3 min. The column was held at 50 ◦ C for 3 min, increased to 110 ◦ C at a rate of 15 ◦ C min−1 and raised to 210 ◦ C at 30 ◦ C min−1 and kept at this temperature for 1 min. To obtain the highest possible sensitivity, the MS detection was operated using time-scheduled SIM based on the selection of some mass peaks of the highest intensity for each compound. The retention times and selected masses for each compound studied by GC–MS are listed in Table 2. The SPME syringe was made in our laboratories consisting of two spinal needles with gauge numbers of 22 and 26. The schematic diagram of homemade SPME syringe is shown in Fig. 1a. Four sections of the homemade SPME syringe includes, (A) the SPME plunger, acting as the internal needle (gauge 26), (B) the SPME barrel which has the role of external needle (gauge 22), (C) the silicon glue and (D) the end tip of the plunger. Fig. 1b shows the schematic diagram of the apparatus used in the electrospinning process. In addition, a photo of the developed SPME fiber (Fig. 1c) along with a commercial SPME fiber (Fig. 1d) is shown. The scanning electron microscopy (SEM) images were obtained by a TESCAN VEGA II XMU (Brno, Czech Republic) and Fourier transform infrared spectroscopy (FTIR) spectra were recorded by an ABB Bomem MB100 (Quebec, Canada). 2.3. Preparation of modified silica nanoparticles In this study, a rather inexpensive modifier, cis-9-octadecenoic acid (OA), which contains a long alkyl chain, was used to
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
13
bar. The extraction temperature was controlled using a water bath. Thermal desorption of retained compounds on the fiber coating was carried out at 200 ◦ C while the split valve of the GC injector was kept closed for 3 min. In all experiments, 4 mL of double distilled water or real water samples were spiked with the selected CBs in 7 mL vial containing 0.4 g of sodium chloride and placed on a magnetic stirrer with maximum stirring rate (1000 rpm). Afterward, the vials were sealed with a PTFE-faced septum and an aluminum cap. The extraction was performed by exposing the fiber coating to the headspace over the aqueous sample for 10 min at 40 ◦ C. After reaching the extraction time, the SPME probe containing analytes from the sample was withdrawn from the vial and inserted into the GC injection port for thermal desorption. 3. Results and discussion
Fig. 1. The schematic diagram of homemade SPME syringe (a), electrospinning setup (b), a photo of newly developed SPME fiber (c) along with a commercial SPME fiber (d).
modify the surface of the nano-scaled silica particles. To modify the SiO2 nanoparticles, first 100 mL hexane and 1.0 g of SiO2 nanoparticles were mixed in a flask equipped with a stirrer, and then 0.46 g amount of OA was added into the solution. The mixture was heated under vigorous stirring at 60 ◦ C for 4 h. It was cooled at −5 ◦ C and then filtered with filter paper to obtain the white precipitate. Finally, the precipitate was rinsed thoroughly with the mixture of ethanol and deionized water (3:7 in volume). The white powder obtained was identified as the OA surface-modified silica nanoparticles [44]. 2.4. Electrospinning First, 0.1 g of PA (10%, w/v) was dissolved in 1 mL of formic acid and stirred for 60 min to obtain a homogenous solution. After complete dissolving of polymer and obtaining a homogeneous solution, various amounts of different silica nanoparticles were added into polymer solution and it was vigorously stirred for 60 min (1000 rpm) to obtain homogenous solution. Then 0.5 mL of this solution was withdrawn into a 2.5 mL syringe which was eventually located in a syringe pump. A homemade SPME device was attached to a small electrical motor in a way that the plunger wire could be rotated at 20 cycles per min while electrospinning was performed. Under this condition, the external needle and the polymer containing syringe needle were connected to the high voltage power supply terminals. A length of 1.5 cm from the end part of plunger wire was used for collecting the nanofibers. The other part of the SPME assembly was protected from the nanofibers flying toward the collector by a packed paper insulator. The distance between the needle and the collector was set at 10 cm. The SPME assembly was in perpendicular position in respect to the syringe. A voltage of 16 kV was applied for the nanofibers production while a flow rate of 3 mL h−1 was set for the polymer solution delivery using a syringe pump. All the electrospinning processes were performed under the ventilation. All fibers were electrospun for 10 min. Finally, the SPME fibers were inserted into the GC injector port for conditioning at 200 ◦ C for 4 h. 2.5. The SPME process The SPME process was performed using the synthesized modified silica-PA nanocomposite fiber coating (∼60 m film thickness), mounted in a home-made device. During the headspace extraction, the aqueous sample was continuously stirred by a magnetic stir
To evaluate the capability of new fiber coating, the feasibility of HS-SPME of some important organic pollutants from water samples was considered. The applicability of the new fiber coating was examined by headspace SPME of some selected CBs, as model compounds, from aqueous samples. Influencing parameters on the morphology of nanocomposite such as typical of modified silica nanoparticle and the weight ratio of components were optimized. In addition, effect of different parameters influencing the extraction efficiency including extraction temperature, extraction time, ionic strength, desorption temperature, and desorption time were investigated and optimized. 3.1. Synthesis of modified silica-PA nanocomposite coating It has been recently shown that PA nanofibers coating can be used as a sorbent for extraction of organic analytes from aqueous samples while some degree of flexibility and mechanical stability. The enhancement of surface area and porosity of polymer structure, being synthesized as sorbent, surely improves the extraction efficiency. The use of inorganic nanoparticles in polymer structure is an efficient way to fulfill this purpose. Silica nanoparticles are inorganic materials which can increase the surface area of PA nanofibers with no or little interactions toward organic compounds. Therefore, the modification of silica nanoparticles with an organic modifier was considered to increase the interactions with the desired analytes. Also, it is well known that the dispersion of the nanofillers in the polymer network can be improved with the aids of surface modification by chemical reaction or non-reactive modifiers [32–34]. After successful preliminary experiments, it was necessary to optimize the synthesizing conditions in order to achieve the highest possible extraction efficiency. To do so, the effects of different types of silica nanoparticles were investigated at a fixed concentration of 0.005 g in 1 mL of PA solution. Fig. 2 shows that PA-silica nanocomposites, due to synergistic effect of silica nanoparticles and PA nanofibers, in compare with PA nanofibers exhibit higher extraction efficiencies for the selected CBs. Our results revealed that the modified silica-PA nanocomposites (C) containing silica nanoparticles modified by OA could lead to higher extraction efficiencies in comparison with the nanocomposites containing pristine silica nanoparticle, silica nanoparticles modified by methyl group (A), and Si (B). These results might be due to the effective dispersion of OA-silica nanoparticle in network polymer or chemical structure of OA can enhance contributes to hydrophobic and – interaction between analytes and the modified nanocomposite compared to the others silica nanoparticles. 3.2. Characterization of modified silica-PA nanocomposites FTIR was employed to examine and recognition of any changes on PA and modified silica-PA nanocomposites after applying
14
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
Fig. 2. Effect of the different type of silica nanoparticles on the extraction efficiency, experimental condition as follows: extraction of 4 mL of sample with maximum agitation; NaCl (10%, w/v); concentration of each CB, 1 mg L−1 ; extraction time, 10 min; extraction temperature, 45 ◦ C; desorption time for 3 min at 200 ◦ C; amount silica nanoparticles (5%, w/v).
electrospinning (Fig. 3). The absorption peaks at ∼3000–3600 cm−1 and 2800–3000 cm−1 correspond to the N H and C H stretching vibrations of PA, respectively. However, the spectra of modified silica-PA nanocomposites exhibit two new absorption peaks appeared at 1107 and 466 cm−1 , which are attributed to the presence of SiO2 in the nanocomposite. FTIR was also employed to study the surface of different silica nanoparticles (Fig. 4). The Si O Si absorption peak is observed at 1107 cm−1 in four spectra obtained for different types of SiO2 and modified SiO2 . The modification process carried out on the surface of the silica nanoparticles and recognition of any changes on the surface before and after modification by OA was studied by FTIR. As shown in Fig. 4a, the characteristic peak of COOH, originated from OA, at 1729 cm−1 vanished [45]. The absorption bands from the stretching of the carboxylate and the bending of the long alkyl chain are observed at 1712 and 724 cm−1 , respectively. In addition, the peaks at 2928, 2856, and 1464 cm−1 could be seen in the spectra of the SiO2 -OA samples for the absorptions of the CH2 asymmetrical stretching, symmetrical stretching, and scissoring vibrations, respectively, indicating that OA has successfully modified the surface of the SiO2 nanoparticles via reacting the carboxylic acid groups with the Si OH groups. The spectrum of
methyl-modified SiO2 is shown in Fig. 4b. The Si O Si absorption peak is observed at 1107 cm−1 and the peaks appeared at 2928 and 2856 cm−1 are due to the presence of the CH2 asymmetrical and symmetrical stretching vibrations, respectively. This indicates that the surface of the SiO2 nanoparticles is covered by methyl groups. According to the spectrum of pristine SiO2 nanoparticles, shown in Fig. 4c, the Si O Si and O H absorption peaks are observed at 1107 cm−1 and 3200–3600 cm−1 , respectively. The single absorption peak observed in Si-modified silica nanoparticles appeared at 1107 cm−1 is attributed to Si O Si bond (Fig. 4d). The SEM images obtained along with the energy dispersive Xray analysis (EDX) spectra of modified silica-PA nanocomposites are exhibited in Fig. 5. The homogeneity and the porous surface structure of the modified silica-PA nanocomposites from different parts of film were confirmed by SEM and it shows that the composite nanofibers diameters are lower than 170 nm (Fig. 5b). The thickness of the nanocomposite coating after electrospun was about 60 m. The obtained EDX spectra also confirm the SEM results. The spectrum of the pristine PA nanofibers exhibits the following atomic composition (Fig. 5c): ≈76.12% C, ≈21.13% O and ≈0.25% Si. While the EDX chemical analysis of the modified silica-PA nanocomposites nanofibers gives approximately the following
Fig. 3. FTIR spectra of modified silica-PA nanocomposite (a) and polyamide polymer (b).
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
15
Fig. 4. FTIR spectra of OA-SiO2 (a), methyl-SiO2 (b), pristin-SiO2 (c) and Si-SiO2 (d).
atomic composition (Fig. 5d): ≈72.56% C, ≈25.92% O and ≈1.59% Si. The change of Si content in modified silica-PA nanocomposites can be attributed to SiO2 nanoparticles. The SEM image of the fiber’s overall appearance is shown in Fig. 5e. 3.3. Optimization of influence parameters For investigating the effect of modified silica nanoparticles doping level on PA nanocomposite extraction capability, five modified silica-PA nanocomposite by doped with 0.001, 0.005, 0.015, 0.02o and 0.025 g of OA-modified silica nanoparticles (1–25%, w/v) were prepared. As shown in Fig. 6a, extraction capabilities of the OA-modified silica nanocomposites were enhanced by increase amount of OA-silica nanoparticles. These results revealed that the silica nanoparticles doping level has an important role in the sorbent extraction ability. Apparently for most analytes, increasing the modified silica nanoparticles doping level up to 0.025 g leads to an improvement in extraction efficiency, while at higher concentration of silica nanoparticles, the electrospinning process becomes rather difficult to perform. According to these results a doping level of 0.025 g of the OA-modified silica nanoparticles was chosen as the optimum value. The influence of salt addition on the extraction efficiency was also investigated. Usually, the presence of salt increases the ionic strength of aqueous solution and would affect the solubility of organic solutes. This effect leads to varying the partition coefficient of analytes between sample headspace and solution (Khs ) hence the extraction efficiency may be changed. In this experiment, the influence of sodium chloride concentration ranging from 0.0 to 30% (w/v) was evaluated while the extraction temperature and time were kept at 40 ◦ C and 10 min, respectively (Fig. 6b). When amount of salt was raised up to 20% the distribution of CBs compound between sample and headspace was increased, leading to the increased extraction efficiency. The same results obtained for the rest of analytes when the salt concentration was increased up to 20%, due to the enhancement in solution viscosity, a decrease in extraction efficiency of analytes was observed when the amount of salt exceeded 20%. Consequently, an amount of 20% NaCl was added to the extraction solution in further experiments. Temperature has an important role in the extraction process by affecting the diffusion rate of analytes into coating. The effect of temperature was evaluated by varying the temperature from 20 to 60 ◦ C by exposing the SPME fiber to the headspace of sample
solution for 10 min (Fig. 6c). Temperature can influence the extraction efficiency due to three different effects. From one side, higher temperature causes an increase in diffusion coefficient in water leading to shorter extraction time. On other hand at higher temperatures the distribution constant at equilibrium decreases, because sorption is an exothermic process, so the extraction efficiency decreases. Furthermore higher temperature leads to higher vapor pressure of the analytes and hence their concentration in headspace increases. As it is shown, the extraction efficiency was at maximum level for all of CBs up to 26 ◦ C and a decrease in extraction efficiency was observed when temperature was further increased. Considering these results a temperature of 26 ◦ C was selected as optimum extraction temperature. The highest possible desorption temperature, without damaging the fiber coating should be applied to the fiber in order to carry over of the analyte during the extraction process be avoided. Thermal desorption was investigated by varying the temperature of injection chamber in the range of 130–250 ◦ C (Fig. 7a). The maximum signal was obtained at the temperature 200 ◦ C. Also, desorption time was evaluated at the time range of 1–4 min. After each desorption process, carry over effect was evaluated. Finally, the temperature of 200 ◦ C and the desorption time of 3 min were chosen (Fig. 7b). Since the SPME technique is an equilibrium process, between the fiber coating and solution, it is important to determine the acceptable extraction time. The SPME equilibrium time was investigated by exposing the modified nanocomposite coating to the headspace of sample containing the target analytes for 5–55 min. The equilibrium time of most analytes could be reached after 20 min (Fig. 7c). In the routine analysis it is not necessary to reach the equilibrium, thus an extraction time of 15 min was chosen as optimum value to shorten the analysis time. An important aspect in the development of any SPME fiber coating relies on the verification of desorption efficiency. This is performed by evaluation of carryover effect, which is defined as the percentage of the remaining analyte in the coating after the completion of desorption step to the amount of analyte desorbed during the desorption step. Maximum percentage of the carry over for 3 min duration at 200 ◦ C using the splitless mode was evaluated by exposing the prepared fiber to the headspace of samples spiked with relatively high concentrations of CBs (1 mg L−1 ). The carry-over percentage for each analyte was lower than 0.02%, indicating the insignificancy of the memory effect for the subsequent
16
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
Fig. 5. SEM of polyamide polymer (a) and OA-modified silica nanocomposites (b) in the same scale; EDX of polyamide polymer (c), OA-modified silica nanocomposites (d) and SEM of fiber’s overall appearance (e).
run. Although these values were negligible for the 4 min duration time, but for enhancing the fiber life time a 3 min time duration was selected.
3.4. The physical and chemical properties of the developed fiber coating Almost all commercially available SPME fibers are prepared based on the fused-silica substrates which are fragile and should be handled with care throughout the sampling and desorption process. To overcome this drawback, a modified silica-PA nanocomposite was coated over a home-made unbreakable metal wire by the use of electrospinning. The prepared SPME fiber is physically stable and unbreakable during the whole practical stages. Also, the obtained results showed that the PA nanocomposite is rather stable in organic solvents including methanol, ethanol, acetone, chloroform, dichloromethane, carbontetrachloride, diethyl ether, ethyl acetate, hexane, and toluene. Furthermore, the prepared
nanocomposite fiber was found to be stable in acidic and basic solutions as well. While C8- and C18-bonded silica, inorganic/organic mesoporous silica, cyclodextrin-bonded silica and restricted access materials are incorporated onto the SPME fibers by epoxy glue [46–50], in this study, electrospinning allowed us to prepare the modified nanosilica-PA composite and employ it as a fiber coating with no need to any adhesive. The surface of the silica nanoparticles was modified by various compounds to improve the compatibility between the nanoparticles and the polymer network. The presence of alkyl and silicon on the surface of the nano-sized silica could reduce the interactions among the silica nanoparticles, and also lowers the size of agglomerates as the nanoparticles content is increased. The coating life is rather important as far as the practical applications are concerned. The coating damage mainly occurs at high temperatures and/or in acidic or alkaline environment. The headspace mode is usually preferred and considered as a very gentle process with almost no durability problems. Our experiments
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
17
Fig. 6. Effect of nanoparticle amount on the extraction efficiency (a), effect of salt concentration on the extraction efficiency (b) and effect of extraction temperature on the extraction efficiency (c).
Fig. 7. Effect of desorption temperature on the extraction efficiency (a), effect of desorption time on the extraction efficiency (b) and effect of extraction time on the extraction efficiency (c).
18
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
Table 3 Some analytical data obtained from HS-SPME of selected CBs using the modified silica-PA nanocomposite fiber and GC–MS. Compound
MCB 14DCB 12DCB 124TCB
LDR (ng mL−1 )
LOD (ng L−1 )
R2
75–1000 50–1000 50–1000 50–1000
0.9897 0.9967 0.9987 0.9992
30 10 5 5
Inter-day RSD% (n = 3)
Intra-day RSD% (n = 3)
12 7 6 4
14 12 10 9
Fiber-to-fiber RSD% (n = 3)
13 8 9 5
Intra-day relative recovery%
Kalan dam water
Tap water
94 98 103 97
98 102 99 96
Table 4 Comparison of the current method with other methods. Methods
Compounds 14DCB a
−1
Current method
LOD (ng L RSD%b
HS-SPMEc
LOD (ng L−1 ) RSD%
HS-SPMEd
LOD (ng L−1 ) RSD%
10 4
10 4
DI-SPMEe
LOD (ng L−1 ) RSD%
42 1.9
21 5.6
a b c d e
)
10 7
12DCB
6 5.97
124TCB
Sample volume (mL)
The linearity range (ng mL−1)
R2
Refs.
5 6
5 4
4
0.05–1
>0.9967
CM
6 5.17
4 3.97
5
0.02–20
>0.991
[51]
10
0.05–0.9
>0.9627
[35]
>0.9957
[52]
10 7 265 0.6
5
1–100
Limits of detection (S/N = 3). Relative standard deviation (n = 3). HS-SPME with PDMS (100 m) fiber and GC/MS. HS-SPME with polyurethane (25 m) fiber and GC/MS. Direct immersion solid phase microextraction (DI-SPME) with PDMS (100 m) fiber and GC/FID.
on the developed fiber coating including its exposure to different organic solvents for lengthy duration and its subsequent analysis revealed that similar data for extraction of CBs could be obtained. Durability of a single fiber was evaluated by consecutive analysis of distilled water samples spiked with standard solutions of CBs at 1 mg L−1 . The GC–FID responses of all analytes were decreased by almost 1% after 64 analyses while they reduced to 2% after 120 runs except MCB, showing a reduction of 4%.
3.5. Analytical results The optimized headspace adsorptive microextraction method using the fabricated modified silica-PA nanocomposite fiber was evaluated by quantitative analysis of the spiked DDW samples. The major quantitative data obtained for the selected CBs are tabulated in Table 3. The limits of detection (S/N = 3) were 5–30 ng L−1 and the limits of quantification (LOQ) were 50–75 ng L−1 (the lowest points of the calibration curves). At the optimum conditions, the relative standard deviation values for a double distilled water spiked with the selected CBs at 100 ng L−1 were 4–12% (n = 3). The calibration curves of analytes were investigated in the range of 50–1000 ng L−1 and R2 between 0.9897 and 0.9992 were obtained. In addition, the reproducibility of the fiber production, expressed as fiber-to-fiber RSD%, was evaluated on the base of peak areas found out after the headspace extraction of a 100 ng L−1 mixture of CBs. The fiber-tofiber reproducibility for three different fibers, synthesized under the same condition described in Section 2.4, was also investigated. The RSD% for the three fibers under optimum extraction/desorption condition is listed in Table 3. The use of modified silica nanoparticles, acting as nanofillers, is quite important for homogeneous dispersion of nanoparticles in PA network, leading to an acceptable reproducibility. To evaluate the applicability of the developed method, spiked (0.1 ng mL−1 ) and non-spiked water samples from Calan Dam (Malayer-Iran) and Tehran drinking water were extracted and
Fig. 8. Mass chromatogram obtained after HS-SPME of a real water sample spiked with CBs (100 ng L−1 ) using SIM mode under optimized condition.
analyzed. None of the selected CBs could be detected in these nonspiked samples. Relative recovery (RR) was measured as the peak area ratio of real sample and double distilled water sample spiked with analyte at the same level. Relative recoveries obtained for the spiked real water samples were in the range of 93–103% at the concentration level of 0.1 ng mL−1 (Table 3), indicating the absence of major matrix effects on the extraction performance of the modified silica-PA nanocomposite coating. The GC–MS chromatogram obtained after headspace SPME of a Calan Dam sample at the concentration of 0.1 ng mL−1 is shown in Fig. 8.
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
19
Table 5 Comparison of extraction efficiency* of the developed fiber coating with other commercial fibers. Compounds
MCB 12DCB 14DCB 124TCB * a b c d e f g h k m
Fiber Develop coating
Aa
Bb
Cc
Dd
Ee
Ff
Gg
Hh
Kk
Mm
21 23 24 31
70 80 75 –
21 39 41 –
10 15 15 –
3.2 31 30 –
4.2 11 11 –
– – – 4.8
– – – 7.9
6 15 15 –
2 4 5 –
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Electrospun modified silica-polyamide nanocomposite as a novel fiber coating Habib Bagheri ∗ , Ali Roostaie Environmental and Bio-Analytical Laboratories, Department of Chemistry, Sharif University of Technology, P.O. Box 11365-9516, Tehran, Iran
a r t i c l e
i n f o
Article history: Received 13 July 2013 Received in revised form 30 October 2013 Accepted 9 November 2013 Available online 17 November 2013 Keywords: Modified silica nanoparticles Chlorobenzenes Electrospun Nanocomposites
a b s t r a c t In the present work, a new solid phase microextraction (SPME) fiber coating based on modified silicapolyamide (PA) nanocomposite was electrospun on a stainless steel wire. Four modified silica-PA nanocomposites together with PA were fabricated by dispersing several typical modified silica nanoparticles in PA polymer solution prior to electrospinning. The surface characteristic of PA nanofibers and modified silica-PA nanocomposites was investigated using scanning electron microscopy (SEM). The homogeneity and the porous surface structure of the modified silica-PA nanocomposites were confirmed by SEM, showing nanofibers diameters lower than 170 nm. The applicability of the new fiber coating was examined by headspace SPME of some selected chlorobenzenes (CBs), as model compounds, from aqueous samples. Subsequently, the extracted analytes were transferred into a gas chromatography (GC) by thermal desorption. Influencing parameters on the morphology of nanocomposites such as type of modified silica nanoparticles and the weight ratio of components were optimized. In addition, effects of different parameters influencing the extraction efficiency including extraction temperature, extraction time, ionic strength, desorption temperature, and desorption time were investigated and optimized. Eventually, the developed method was validated by gas chromatography–mass spectrometry (GC–MS). At the optimum conditions, the relative standard deviation values for a double distilled water spiked with the selected CBs at 100 ng L−1 were 4–12% (n = 3) and the limit of detection for the studied compounds was between 5 and 30 ng L−1 . The calibration curves of analytes were investigated in the range of 50–1000 ng L−1 and correlation coefficients (R2 ) between 0.9897 and 0.9992 were obtained. © 2013 Elsevier B.V. All rights reserved.
1. Introduction SPME is a well-established solvent-free extraction technique in which sampling, preconcentration, and sample introduction are integrated in a single process. Although this technique is predominantly performed on SPME fibers [1,2], the conventional fibers could be broken easily and their coatings are susceptible to stripping when they are exposed to high temperatures and organic solvents. The developments of new fibers are mostly focused on improving the thermal, mechanical, and chemical stabilities, imparting diverse functionalities and polarities, and enhancing the fiber capacity [3–5]. The use of nanocomposites coatings [6–8] and replacing the fragile fused silica substrates by wire metals [9] are mostly being used in recent years. The appropriate selection of the fiber coating is one of the most critical steps in SPME method development. The nature of the coating should be selected based on the chemical properties of the target analytes and affects the overall extraction
∗ Corresponding author. Tel.: +98 21 66005718; fax: +98 21 66012983. E-mail address: [email protected] (H. Bagheri). 0021-9673/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.chroma.2013.11.024
performance, including method sensitivity, selectivity, and reproducibility. The common aspects in the microextraction strategies are mostly focused on miniaturized sorbent amount and the extraction device geometry which often lead to the extraction of small fraction of the analytes from the sample into the extracting phase [10]. One important strategy to improve the sensitivity of the microextraction methods relies on the usage of nanostructured materials in which the interactions between the desired analytes and the sorbent are enhanced. The high surface-to-volume ratio of nanostructured materials provides a great deal of active sites for adsorption, making an appropriate platform for the efficient separation, extraction, and enrichment. Silica nanoparticles (SNPs), due to their chemical inertness, nontoxicity, optical transparency, and excellent thermal stability [11–13], have gained significant attentions. They can be widely used in catalysis [14], chemical process industry [15], removal of metal ions [16], and metal ion preconcentration [17–19] through polymer coatings or other functional groups. Chemical modification of the silica surface makes it possible to combine the mechanical and structural properties of the pure silica particles with the ability of specific chemical and physical interactions. The use of modified silica in stationary phases
12
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
in different chromatographic techniques, supported materials in heterogeneous catalysis and reinforcement agents in composite materials are just a few examples [20]. Large specific surface area and high thermal stability of nanoporous silica materials (e.g., SBA15) have made them to be used as an alternative for SPME coatings. The chemical functionalization of SBA-15 by incorporating organic groups and its capability as a support for the construction of conductive polymer/SBA-15 nanocomposites extends the range of materials applied as SPME coatings [21–23]. The recent studies on polypyrrole/SBA-15 [23] and polyaniline/silica nanocomposite [24] have proven that these can be incorporated as facile coatings for hydrophobic compounds. Organic/inorganic nanocomposites are generally organic polymer composites with inorganic nanoscale building blocks. They combine the advantages of the inorganic material (e.g., rigidity, thermal stability) and the organic polymer (e.g., flexibility, dielectric, ductility, and processability). It is well known that the polymer composites can be fabricated by the incorporation of inorganic reinforcements into the polymer network. The properties of the resulting polymer composites depend on the characteristics, dimensions, and shapes of the inorganic fillers, and also on the interfacial bonding strength [25,26]. It is proposed that with decreasing filler dimension, the specific area of the filler is significantly improved, and in turn it would greatly and effectively improve the transfer of the load between the fillers and the polymer matrix [27]. The inorganic nanofillers were successfully incorporated into the polymeric matrix to strengthen and improve the ductile polymer to be more stiff and resistant for abrasion [28–34]. Recently, it has been shown [35,36] that the electrospinning technique is quite efficient to synthesize a variety of nanofibers coatings on thin stainless steel wires. In methodology, by applying high voltages to a viscous polymeric solution, when the electrostatic repulsive force overcomes the surface tension of the polymeric solution, a charged polymeric jet ejects from the solution and afterwards flies toward the collector and forms the fibrous material with diameters in the scale of nano- to micrometer. The electrospun sorbents were used in SPE [37–40], micro-SPE [41], SPME [42], and membrane extraction [43] of some organic and inorganic species. In this work, for the first time, different amounts of silica nanoparticles and modified silica nanoparticles were embedded in polyamide nanofibers by the electrospinning technique. A rotating stainless steel SPME needle was used as an electrospinning collector and PA nanofibers containing silica nanoparticles were directly formed on the SPME needle. The applicability of this novel SPME coating, prepared only in one step, was assessed for SPME of selected CBs, as model pollutants, from aqueous samples using a homemade SPME unit.
2. Experimental 2.1. Reagents The CB compounds including 1,4-dichlorobenzene (14DCB), 1,2dichlorobenzene (12DCB), 1,2,4-trichlorobenzene (124TCB) and monochlorobenezne (MCB) were purchased from Merck (Darmstadt, Germany). Standard solution (1000 mg L−1 ) of CBs mixture was prepared in HPLC-grade methanol (Merck) and stored in the refrigerator. The working standard solutions were prepared weekly by diluting the standard solution with methanol, and more diluted working solutions were prepared daily by diluting this solution with double distilled water (DDW). Sodium chloride was purchased from Merck. Nylon-6 (N6) was purchased from Kolon Industries Inc. (Seoul, Korea) and formic acid was obtained from Riedel-de
Table 1 Properties of different silica nanoparticles examined in this work. Type of silica nanoparticle
Particle size diameter
Wettability
SiO2 Si-SiO2 Methyl-SiO2
10–50 10–50 10–50
Hydrophilic Neutral Hydrophobic
Table 2 Retention times and selected ions of compounds studied by GC–MS. Compound
Retention time (min)
Selected ions (m/z)
MCB 14DCB 12DCB 124TCB
4 5.5 5.7 6.7
77, 112 146, 148 146, 148 180, 182
Haën (Seelze-Hannover, Germany). All solvents used in this study were of analytical reagent grade or HPLC grade. The different SiO2 nanoparticles with diameter of about 12–50 nm were obtained from Degussa Company (Darmstadt, Germany) (Table 1). Cis-9octadecenoic acid (oleic acid, OA), ethanol and n-hexane were obtained from Merck (Darmstadt, Germany). 2.2. Apparatus A gas chromatograph model Agilent 6820, with a split/splitless injection port and a flame ionization detection system, was used for the optimization process. A homemade glass inlet liner with a total length of 7.7 cm, 6 mm o.d. and 1 mm i.d. was deactivated by trimethylchlorosilane and used. The injector and detector temperatures were set at 200 ◦ C and 290 ◦ C, respectively. For quantitative determination, a Hewlett-Packard (HP, Palo Alto, USA) HP 6890 series GC equipped with a split/splitless injector and a HP 5973 mass-selective detector system were used. The MS was operated in the EI mode (70 eV). For both GC–FID and GC–MS systems, the separation of analytes was carried out using a capillary column HP-5 MS (60 m, 0.25 mm i.d.) with 0.25 m film thickness (Hewlett-Packard, Palo Alto, CA, USA). The carrier gas was helium (99.999%) at a flow rate of 1 mL min−1 . Both instruments were operated in the splitless mode and the split valve was kept closed for 3 min. The column was held at 50 ◦ C for 3 min, increased to 110 ◦ C at a rate of 15 ◦ C min−1 and raised to 210 ◦ C at 30 ◦ C min−1 and kept at this temperature for 1 min. To obtain the highest possible sensitivity, the MS detection was operated using time-scheduled SIM based on the selection of some mass peaks of the highest intensity for each compound. The retention times and selected masses for each compound studied by GC–MS are listed in Table 2. The SPME syringe was made in our laboratories consisting of two spinal needles with gauge numbers of 22 and 26. The schematic diagram of homemade SPME syringe is shown in Fig. 1a. Four sections of the homemade SPME syringe includes, (A) the SPME plunger, acting as the internal needle (gauge 26), (B) the SPME barrel which has the role of external needle (gauge 22), (C) the silicon glue and (D) the end tip of the plunger. Fig. 1b shows the schematic diagram of the apparatus used in the electrospinning process. In addition, a photo of the developed SPME fiber (Fig. 1c) along with a commercial SPME fiber (Fig. 1d) is shown. The scanning electron microscopy (SEM) images were obtained by a TESCAN VEGA II XMU (Brno, Czech Republic) and Fourier transform infrared spectroscopy (FTIR) spectra were recorded by an ABB Bomem MB100 (Quebec, Canada). 2.3. Preparation of modified silica nanoparticles In this study, a rather inexpensive modifier, cis-9-octadecenoic acid (OA), which contains a long alkyl chain, was used to
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
13
bar. The extraction temperature was controlled using a water bath. Thermal desorption of retained compounds on the fiber coating was carried out at 200 ◦ C while the split valve of the GC injector was kept closed for 3 min. In all experiments, 4 mL of double distilled water or real water samples were spiked with the selected CBs in 7 mL vial containing 0.4 g of sodium chloride and placed on a magnetic stirrer with maximum stirring rate (1000 rpm). Afterward, the vials were sealed with a PTFE-faced septum and an aluminum cap. The extraction was performed by exposing the fiber coating to the headspace over the aqueous sample for 10 min at 40 ◦ C. After reaching the extraction time, the SPME probe containing analytes from the sample was withdrawn from the vial and inserted into the GC injection port for thermal desorption. 3. Results and discussion
Fig. 1. The schematic diagram of homemade SPME syringe (a), electrospinning setup (b), a photo of newly developed SPME fiber (c) along with a commercial SPME fiber (d).
modify the surface of the nano-scaled silica particles. To modify the SiO2 nanoparticles, first 100 mL hexane and 1.0 g of SiO2 nanoparticles were mixed in a flask equipped with a stirrer, and then 0.46 g amount of OA was added into the solution. The mixture was heated under vigorous stirring at 60 ◦ C for 4 h. It was cooled at −5 ◦ C and then filtered with filter paper to obtain the white precipitate. Finally, the precipitate was rinsed thoroughly with the mixture of ethanol and deionized water (3:7 in volume). The white powder obtained was identified as the OA surface-modified silica nanoparticles [44]. 2.4. Electrospinning First, 0.1 g of PA (10%, w/v) was dissolved in 1 mL of formic acid and stirred for 60 min to obtain a homogenous solution. After complete dissolving of polymer and obtaining a homogeneous solution, various amounts of different silica nanoparticles were added into polymer solution and it was vigorously stirred for 60 min (1000 rpm) to obtain homogenous solution. Then 0.5 mL of this solution was withdrawn into a 2.5 mL syringe which was eventually located in a syringe pump. A homemade SPME device was attached to a small electrical motor in a way that the plunger wire could be rotated at 20 cycles per min while electrospinning was performed. Under this condition, the external needle and the polymer containing syringe needle were connected to the high voltage power supply terminals. A length of 1.5 cm from the end part of plunger wire was used for collecting the nanofibers. The other part of the SPME assembly was protected from the nanofibers flying toward the collector by a packed paper insulator. The distance between the needle and the collector was set at 10 cm. The SPME assembly was in perpendicular position in respect to the syringe. A voltage of 16 kV was applied for the nanofibers production while a flow rate of 3 mL h−1 was set for the polymer solution delivery using a syringe pump. All the electrospinning processes were performed under the ventilation. All fibers were electrospun for 10 min. Finally, the SPME fibers were inserted into the GC injector port for conditioning at 200 ◦ C for 4 h. 2.5. The SPME process The SPME process was performed using the synthesized modified silica-PA nanocomposite fiber coating (∼60 m film thickness), mounted in a home-made device. During the headspace extraction, the aqueous sample was continuously stirred by a magnetic stir
To evaluate the capability of new fiber coating, the feasibility of HS-SPME of some important organic pollutants from water samples was considered. The applicability of the new fiber coating was examined by headspace SPME of some selected CBs, as model compounds, from aqueous samples. Influencing parameters on the morphology of nanocomposite such as typical of modified silica nanoparticle and the weight ratio of components were optimized. In addition, effect of different parameters influencing the extraction efficiency including extraction temperature, extraction time, ionic strength, desorption temperature, and desorption time were investigated and optimized. 3.1. Synthesis of modified silica-PA nanocomposite coating It has been recently shown that PA nanofibers coating can be used as a sorbent for extraction of organic analytes from aqueous samples while some degree of flexibility and mechanical stability. The enhancement of surface area and porosity of polymer structure, being synthesized as sorbent, surely improves the extraction efficiency. The use of inorganic nanoparticles in polymer structure is an efficient way to fulfill this purpose. Silica nanoparticles are inorganic materials which can increase the surface area of PA nanofibers with no or little interactions toward organic compounds. Therefore, the modification of silica nanoparticles with an organic modifier was considered to increase the interactions with the desired analytes. Also, it is well known that the dispersion of the nanofillers in the polymer network can be improved with the aids of surface modification by chemical reaction or non-reactive modifiers [32–34]. After successful preliminary experiments, it was necessary to optimize the synthesizing conditions in order to achieve the highest possible extraction efficiency. To do so, the effects of different types of silica nanoparticles were investigated at a fixed concentration of 0.005 g in 1 mL of PA solution. Fig. 2 shows that PA-silica nanocomposites, due to synergistic effect of silica nanoparticles and PA nanofibers, in compare with PA nanofibers exhibit higher extraction efficiencies for the selected CBs. Our results revealed that the modified silica-PA nanocomposites (C) containing silica nanoparticles modified by OA could lead to higher extraction efficiencies in comparison with the nanocomposites containing pristine silica nanoparticle, silica nanoparticles modified by methyl group (A), and Si (B). These results might be due to the effective dispersion of OA-silica nanoparticle in network polymer or chemical structure of OA can enhance contributes to hydrophobic and – interaction between analytes and the modified nanocomposite compared to the others silica nanoparticles. 3.2. Characterization of modified silica-PA nanocomposites FTIR was employed to examine and recognition of any changes on PA and modified silica-PA nanocomposites after applying
14
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
Fig. 2. Effect of the different type of silica nanoparticles on the extraction efficiency, experimental condition as follows: extraction of 4 mL of sample with maximum agitation; NaCl (10%, w/v); concentration of each CB, 1 mg L−1 ; extraction time, 10 min; extraction temperature, 45 ◦ C; desorption time for 3 min at 200 ◦ C; amount silica nanoparticles (5%, w/v).
electrospinning (Fig. 3). The absorption peaks at ∼3000–3600 cm−1 and 2800–3000 cm−1 correspond to the N H and C H stretching vibrations of PA, respectively. However, the spectra of modified silica-PA nanocomposites exhibit two new absorption peaks appeared at 1107 and 466 cm−1 , which are attributed to the presence of SiO2 in the nanocomposite. FTIR was also employed to study the surface of different silica nanoparticles (Fig. 4). The Si O Si absorption peak is observed at 1107 cm−1 in four spectra obtained for different types of SiO2 and modified SiO2 . The modification process carried out on the surface of the silica nanoparticles and recognition of any changes on the surface before and after modification by OA was studied by FTIR. As shown in Fig. 4a, the characteristic peak of COOH, originated from OA, at 1729 cm−1 vanished [45]. The absorption bands from the stretching of the carboxylate and the bending of the long alkyl chain are observed at 1712 and 724 cm−1 , respectively. In addition, the peaks at 2928, 2856, and 1464 cm−1 could be seen in the spectra of the SiO2 -OA samples for the absorptions of the CH2 asymmetrical stretching, symmetrical stretching, and scissoring vibrations, respectively, indicating that OA has successfully modified the surface of the SiO2 nanoparticles via reacting the carboxylic acid groups with the Si OH groups. The spectrum of
methyl-modified SiO2 is shown in Fig. 4b. The Si O Si absorption peak is observed at 1107 cm−1 and the peaks appeared at 2928 and 2856 cm−1 are due to the presence of the CH2 asymmetrical and symmetrical stretching vibrations, respectively. This indicates that the surface of the SiO2 nanoparticles is covered by methyl groups. According to the spectrum of pristine SiO2 nanoparticles, shown in Fig. 4c, the Si O Si and O H absorption peaks are observed at 1107 cm−1 and 3200–3600 cm−1 , respectively. The single absorption peak observed in Si-modified silica nanoparticles appeared at 1107 cm−1 is attributed to Si O Si bond (Fig. 4d). The SEM images obtained along with the energy dispersive Xray analysis (EDX) spectra of modified silica-PA nanocomposites are exhibited in Fig. 5. The homogeneity and the porous surface structure of the modified silica-PA nanocomposites from different parts of film were confirmed by SEM and it shows that the composite nanofibers diameters are lower than 170 nm (Fig. 5b). The thickness of the nanocomposite coating after electrospun was about 60 m. The obtained EDX spectra also confirm the SEM results. The spectrum of the pristine PA nanofibers exhibits the following atomic composition (Fig. 5c): ≈76.12% C, ≈21.13% O and ≈0.25% Si. While the EDX chemical analysis of the modified silica-PA nanocomposites nanofibers gives approximately the following
Fig. 3. FTIR spectra of modified silica-PA nanocomposite (a) and polyamide polymer (b).
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
15
Fig. 4. FTIR spectra of OA-SiO2 (a), methyl-SiO2 (b), pristin-SiO2 (c) and Si-SiO2 (d).
atomic composition (Fig. 5d): ≈72.56% C, ≈25.92% O and ≈1.59% Si. The change of Si content in modified silica-PA nanocomposites can be attributed to SiO2 nanoparticles. The SEM image of the fiber’s overall appearance is shown in Fig. 5e. 3.3. Optimization of influence parameters For investigating the effect of modified silica nanoparticles doping level on PA nanocomposite extraction capability, five modified silica-PA nanocomposite by doped with 0.001, 0.005, 0.015, 0.02o and 0.025 g of OA-modified silica nanoparticles (1–25%, w/v) were prepared. As shown in Fig. 6a, extraction capabilities of the OA-modified silica nanocomposites were enhanced by increase amount of OA-silica nanoparticles. These results revealed that the silica nanoparticles doping level has an important role in the sorbent extraction ability. Apparently for most analytes, increasing the modified silica nanoparticles doping level up to 0.025 g leads to an improvement in extraction efficiency, while at higher concentration of silica nanoparticles, the electrospinning process becomes rather difficult to perform. According to these results a doping level of 0.025 g of the OA-modified silica nanoparticles was chosen as the optimum value. The influence of salt addition on the extraction efficiency was also investigated. Usually, the presence of salt increases the ionic strength of aqueous solution and would affect the solubility of organic solutes. This effect leads to varying the partition coefficient of analytes between sample headspace and solution (Khs ) hence the extraction efficiency may be changed. In this experiment, the influence of sodium chloride concentration ranging from 0.0 to 30% (w/v) was evaluated while the extraction temperature and time were kept at 40 ◦ C and 10 min, respectively (Fig. 6b). When amount of salt was raised up to 20% the distribution of CBs compound between sample and headspace was increased, leading to the increased extraction efficiency. The same results obtained for the rest of analytes when the salt concentration was increased up to 20%, due to the enhancement in solution viscosity, a decrease in extraction efficiency of analytes was observed when the amount of salt exceeded 20%. Consequently, an amount of 20% NaCl was added to the extraction solution in further experiments. Temperature has an important role in the extraction process by affecting the diffusion rate of analytes into coating. The effect of temperature was evaluated by varying the temperature from 20 to 60 ◦ C by exposing the SPME fiber to the headspace of sample
solution for 10 min (Fig. 6c). Temperature can influence the extraction efficiency due to three different effects. From one side, higher temperature causes an increase in diffusion coefficient in water leading to shorter extraction time. On other hand at higher temperatures the distribution constant at equilibrium decreases, because sorption is an exothermic process, so the extraction efficiency decreases. Furthermore higher temperature leads to higher vapor pressure of the analytes and hence their concentration in headspace increases. As it is shown, the extraction efficiency was at maximum level for all of CBs up to 26 ◦ C and a decrease in extraction efficiency was observed when temperature was further increased. Considering these results a temperature of 26 ◦ C was selected as optimum extraction temperature. The highest possible desorption temperature, without damaging the fiber coating should be applied to the fiber in order to carry over of the analyte during the extraction process be avoided. Thermal desorption was investigated by varying the temperature of injection chamber in the range of 130–250 ◦ C (Fig. 7a). The maximum signal was obtained at the temperature 200 ◦ C. Also, desorption time was evaluated at the time range of 1–4 min. After each desorption process, carry over effect was evaluated. Finally, the temperature of 200 ◦ C and the desorption time of 3 min were chosen (Fig. 7b). Since the SPME technique is an equilibrium process, between the fiber coating and solution, it is important to determine the acceptable extraction time. The SPME equilibrium time was investigated by exposing the modified nanocomposite coating to the headspace of sample containing the target analytes for 5–55 min. The equilibrium time of most analytes could be reached after 20 min (Fig. 7c). In the routine analysis it is not necessary to reach the equilibrium, thus an extraction time of 15 min was chosen as optimum value to shorten the analysis time. An important aspect in the development of any SPME fiber coating relies on the verification of desorption efficiency. This is performed by evaluation of carryover effect, which is defined as the percentage of the remaining analyte in the coating after the completion of desorption step to the amount of analyte desorbed during the desorption step. Maximum percentage of the carry over for 3 min duration at 200 ◦ C using the splitless mode was evaluated by exposing the prepared fiber to the headspace of samples spiked with relatively high concentrations of CBs (1 mg L−1 ). The carry-over percentage for each analyte was lower than 0.02%, indicating the insignificancy of the memory effect for the subsequent
16
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
Fig. 5. SEM of polyamide polymer (a) and OA-modified silica nanocomposites (b) in the same scale; EDX of polyamide polymer (c), OA-modified silica nanocomposites (d) and SEM of fiber’s overall appearance (e).
run. Although these values were negligible for the 4 min duration time, but for enhancing the fiber life time a 3 min time duration was selected.
3.4. The physical and chemical properties of the developed fiber coating Almost all commercially available SPME fibers are prepared based on the fused-silica substrates which are fragile and should be handled with care throughout the sampling and desorption process. To overcome this drawback, a modified silica-PA nanocomposite was coated over a home-made unbreakable metal wire by the use of electrospinning. The prepared SPME fiber is physically stable and unbreakable during the whole practical stages. Also, the obtained results showed that the PA nanocomposite is rather stable in organic solvents including methanol, ethanol, acetone, chloroform, dichloromethane, carbontetrachloride, diethyl ether, ethyl acetate, hexane, and toluene. Furthermore, the prepared
nanocomposite fiber was found to be stable in acidic and basic solutions as well. While C8- and C18-bonded silica, inorganic/organic mesoporous silica, cyclodextrin-bonded silica and restricted access materials are incorporated onto the SPME fibers by epoxy glue [46–50], in this study, electrospinning allowed us to prepare the modified nanosilica-PA composite and employ it as a fiber coating with no need to any adhesive. The surface of the silica nanoparticles was modified by various compounds to improve the compatibility between the nanoparticles and the polymer network. The presence of alkyl and silicon on the surface of the nano-sized silica could reduce the interactions among the silica nanoparticles, and also lowers the size of agglomerates as the nanoparticles content is increased. The coating life is rather important as far as the practical applications are concerned. The coating damage mainly occurs at high temperatures and/or in acidic or alkaline environment. The headspace mode is usually preferred and considered as a very gentle process with almost no durability problems. Our experiments
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
17
Fig. 6. Effect of nanoparticle amount on the extraction efficiency (a), effect of salt concentration on the extraction efficiency (b) and effect of extraction temperature on the extraction efficiency (c).
Fig. 7. Effect of desorption temperature on the extraction efficiency (a), effect of desorption time on the extraction efficiency (b) and effect of extraction time on the extraction efficiency (c).
18
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
Table 3 Some analytical data obtained from HS-SPME of selected CBs using the modified silica-PA nanocomposite fiber and GC–MS. Compound
MCB 14DCB 12DCB 124TCB
LDR (ng mL−1 )
LOD (ng L−1 )
R2
75–1000 50–1000 50–1000 50–1000
0.9897 0.9967 0.9987 0.9992
30 10 5 5
Inter-day RSD% (n = 3)
Intra-day RSD% (n = 3)
12 7 6 4
14 12 10 9
Fiber-to-fiber RSD% (n = 3)
13 8 9 5
Intra-day relative recovery%
Kalan dam water
Tap water
94 98 103 97
98 102 99 96
Table 4 Comparison of the current method with other methods. Methods
Compounds 14DCB a
−1
Current method
LOD (ng L RSD%b
HS-SPMEc
LOD (ng L−1 ) RSD%
HS-SPMEd
LOD (ng L−1 ) RSD%
10 4
10 4
DI-SPMEe
LOD (ng L−1 ) RSD%
42 1.9
21 5.6
a b c d e
)
10 7
12DCB
6 5.97
124TCB
Sample volume (mL)
The linearity range (ng mL−1)
R2
Refs.
5 6
5 4
4
0.05–1
>0.9967
CM
6 5.17
4 3.97
5
0.02–20
>0.991
[51]
10
0.05–0.9
>0.9627
[35]
>0.9957
[52]
10 7 265 0.6
5
1–100
Limits of detection (S/N = 3). Relative standard deviation (n = 3). HS-SPME with PDMS (100 m) fiber and GC/MS. HS-SPME with polyurethane (25 m) fiber and GC/MS. Direct immersion solid phase microextraction (DI-SPME) with PDMS (100 m) fiber and GC/FID.
on the developed fiber coating including its exposure to different organic solvents for lengthy duration and its subsequent analysis revealed that similar data for extraction of CBs could be obtained. Durability of a single fiber was evaluated by consecutive analysis of distilled water samples spiked with standard solutions of CBs at 1 mg L−1 . The GC–FID responses of all analytes were decreased by almost 1% after 64 analyses while they reduced to 2% after 120 runs except MCB, showing a reduction of 4%.
3.5. Analytical results The optimized headspace adsorptive microextraction method using the fabricated modified silica-PA nanocomposite fiber was evaluated by quantitative analysis of the spiked DDW samples. The major quantitative data obtained for the selected CBs are tabulated in Table 3. The limits of detection (S/N = 3) were 5–30 ng L−1 and the limits of quantification (LOQ) were 50–75 ng L−1 (the lowest points of the calibration curves). At the optimum conditions, the relative standard deviation values for a double distilled water spiked with the selected CBs at 100 ng L−1 were 4–12% (n = 3). The calibration curves of analytes were investigated in the range of 50–1000 ng L−1 and R2 between 0.9897 and 0.9992 were obtained. In addition, the reproducibility of the fiber production, expressed as fiber-to-fiber RSD%, was evaluated on the base of peak areas found out after the headspace extraction of a 100 ng L−1 mixture of CBs. The fiber-tofiber reproducibility for three different fibers, synthesized under the same condition described in Section 2.4, was also investigated. The RSD% for the three fibers under optimum extraction/desorption condition is listed in Table 3. The use of modified silica nanoparticles, acting as nanofillers, is quite important for homogeneous dispersion of nanoparticles in PA network, leading to an acceptable reproducibility. To evaluate the applicability of the developed method, spiked (0.1 ng mL−1 ) and non-spiked water samples from Calan Dam (Malayer-Iran) and Tehran drinking water were extracted and
Fig. 8. Mass chromatogram obtained after HS-SPME of a real water sample spiked with CBs (100 ng L−1 ) using SIM mode under optimized condition.
analyzed. None of the selected CBs could be detected in these nonspiked samples. Relative recovery (RR) was measured as the peak area ratio of real sample and double distilled water sample spiked with analyte at the same level. Relative recoveries obtained for the spiked real water samples were in the range of 93–103% at the concentration level of 0.1 ng mL−1 (Table 3), indicating the absence of major matrix effects on the extraction performance of the modified silica-PA nanocomposite coating. The GC–MS chromatogram obtained after headspace SPME of a Calan Dam sample at the concentration of 0.1 ng mL−1 is shown in Fig. 8.
H. Bagheri, A. Roostaie / J. Chromatogr. A 1324 (2014) 11–20
19
Table 5 Comparison of extraction efficiency* of the developed fiber coating with other commercial fibers. Compounds
MCB 12DCB 14DCB 124TCB * a b c d e f g h k m
Fiber Develop coating
Aa
Bb
Cc
Dd
Ee
Ff
Gg
Hh
Kk
Mm
21 23 24 31
70 80 75 –
21 39 41 –
10 15 15 –
3.2 31 30 –
4.2 11 11 –
– – – 4.8
– – – 7.9
6 15 15 –
2 4 5 –
