Journal of Chromatography A, 1218 (2011) 3581–3587 Contents lists available at ScienceDirect Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma Carbon nanofibers extracted from soot as a sorbent for the determination of aromatic amines from wastewater effluent samples Sajini Vadukumpully, Chanbasha Basheer 1 , Cheng Suh Jeng, Suresh Valiyaveettil ∗ Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore a r t i c l e i n f o Article history: Received 21 September 2010 Received in revised form 30 March 2011 Accepted 1 April 2011 Available online 8 April 2011 Keywords: Microextraction Carbon nanofibers Electrospun nanofibers mat Aromatic amines a b s t r a c t The isolation and characterization of carbon nanofibers from soot obtained by burning natural oil is reported. The fibers were extracted from the soot with tetrahydrofuran followed by sonication. The carbon nanofibers were mixed with poly(vinyl alcohol) and electrospun to get the nanofiber mat. The extraction ability of electrospun nanofibers for the separation and preconcentration of aromatic compounds such as 3-nitroaniline, 4-chloroaniline, 4-bromoaniline and 3,4-dichloroaniline were tested and efficiently evaluated using high performance liquid chromatography. Under optimized conditions, the method showed good linearity in a range of 0.5–50 ␮g L−1 with correlation coefficient ranging from 0.989 to 0.998. High precision of the extraction with RSD values of 4.5–5.8% and low LOD value in a range of 0.009–0.081 ␮g L−1 for all aniline compounds were achieved. The proposed microextraction method offers advantages such as easy operation, high recovery, fast extraction, minimal use of organic solvent and elimination of tedious solvent evaporation and reconstitution steps. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Since the discovery of fullerenes [1] and carbon nanotubes [2], various kinds of nano-size carbon materials are studied for various applications. A few of important and best known structures are carbon nanotubes (CNTs) and carbon nanofibers (CNFs). CNTs and CNFs have been attracting a great deal of academic and industrial interest due to their novel structural features and interesting properties. CNFs are solid carbon fibers with lengths in the order of a few microns and diameters below 100 nm. Owing to their unique physical and chemical properties, these carbon nanomaterials have been used in areas as diverse as electrochemical devices [3], field emission devices [4], sensors [5], gene delivery agents [6], high strength composites [7], hydrogen and charge storage devices [8,9]. Carbon soot collected from burning of renewable sources such as plant seed oil or natural materials has been extensively studied during last decade [10–12]. The nanostructures obtained vary upon the renewable resources, combustion temperature and deposition kinetics. A variety of natural materials were explored to get good yield of desired carbon nanostructures and the flames was called ‘sooting’ flames as they spontaneously generate condensed carbon ∗ Corresponding author at: Department of Chemistry, Faculty of Science, National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore. Tel.: +65 6516 3659; fax: +65 6779 1691. E-mail address: chmsv@nus.edu.sg (S. Valiyaveettil). 1 Present address: Department of Chemistry, King Fahd University of Petroleum and Minerals, KFUPM, Box 1059, Dhahran 31261, Saudi Arabia. 0021-9673/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2011.04.003 in the form of soot agglomerates suspended in the flame gases. Instead of using hydrocarbon as fuel, recently Sarkar et al. reported the synthesis of CNTs by pyrolysing mustard oil [13]. All synthetic methods generate desired carbon nanostructures along with significant amounts of carbonaceous impurities such as amorphous carbon, fullerene, turbostratic graphite (TSG), polyhedral carbon nanoparticles (PCNs) and catalyst particles [14–18]. In order to obtain the optimum performance of the carbon nanostructures, it is important to purify the soot and remove all unwanted materials. Several techniques for isolation of pure carbon nanostructures have been applied such as chemical oxidation [17], use of polymers [19,20], surfactant assisted extraction [21], centrifugation [22], sonication [23], and filtration [24]. Carbon nanomaterials such as CNTs have been successfully used as a new sorbent material in solid phase extraction for preconcentration of analytes [25,26]. Aromatic amines, widely used in the manufacturing process of pesticides, polymers, pharmaceuticals and dyes [27,28] are suspected carcinogens and are highly toxic to aquatic life. Hence, it is necessary to determine these compounds in waste water by a rapid and sensitive analytical technique. Most of the amines have pKa around 9–11, however, aromatic amines such as nitrozoamines have pKa between 3 and 6. pKa values play a vital role on the sorbent based extraction procedure such as solid-phase extraction [29]. Previous sorbent based extraction studies suggested that non-polar sorbents in a basic medium are the best choice [30]. In this context, conventional sorbent such as C18 phases are not suitable at higher or lower pH ranges [31]. Recently, non-conventional sorbents such as C60 and C70 fullerenes, and nanotubes were reported as suitable sorbent for the amine extraction. 3582 S. Vadukumpully et al. / J. Chromatogr. A 1218 (2011) 3581–3587 Fig. 1. SEM images of (A) the raw soot, (B) extracted CNFs and (C) TEM of CNFs obtained from the soot by THF extraction. Inset in (C) shows the SAED pattern obtained from a single nanofiber. The non-polar carbon based sorbents give selectivity and precision for amine extraction [32]. Miniaturized extraction techniques such as solid-phase micro-extraction (SPME) and liquid–liquid–liquid micro-extraction (LLLME) have been employed in the extraction of aromatic amines from aqueous samples [33–35]. However, the application of CNFs in this area has not been well explored. CNFs offer large surface area and high chemical stability which are ideal for a potential sorbent material for solid-phase extraction [36–38]. Herein, we report the successful extraction, purification, characterization of new CNFs and its performance as a sorbent material for the extraction and preconcentration of aromatic amines. 2. Experimental 2.1. Reagents and materials Good quality edible oil was purchased from local market. Poly(vinyl alcohol) (PVA average molecular weight = 124 000–186 000, 87–89% hydrolyzed) and phosphoric acid were purchased from Sigma–Aldrich, USA. Glutaraldehyde for cross linking of fibers was obtained from Alfa Aesar. HPLC-grade methanol, acetonitrile, tetrahydrofuran (THF), sodium acetate and glacial acetic acid were bought from Merck (Darmstadt, Germany). Ultrapure water was obtained from Mili-Q (Milford, MA, USA) water purification system. Aromatic amine standards, 3nitroaniline, 4-chloroaniline, 4-bromoaniline and 3,4-chloroaniline were obtained from Fluka chemicals (Buchs, Switzerland) and were used without any further purification. 2.2. Sample preparation 2.3. Isolation of CNFs The oil was placed in an open container and burned using a small cotton thread at a controlled rate in ambient conditions. The emitted carbon smoke was then condensed onto a cold surface placed on top, but at a short distance from the flame. This process was continued for several days to collect enough carbon soot, which was used for subsequent experiments. Briefly, about 25 mg of soot was mixed with 50 mL THF with the aid of ultrasonication for 20 min. Only CNFs were extracted in THF. After 6 h, the supernatant solution was collected and the solvent was evaporated. The solid product was dried thoroughly in oven at 85 ◦ C for overnight and was re-suspended in THF for subsequent characterization and analysis. A 3000 -1 -1 1347 cm 1589 cm Intensity (a.u) 1048 cm -1 1455 cm-1 1386 cm -1 -1 B 2349 cm -1 2915 cm -1 2865 cm % Transmittance (a.u) The standard stock solution of 1 mg mL−1 of each analyte was prepared by weighing appropriate quantities of individual com- pound and dissolving in methanol. Since the solubility of the amines is lower in water, we used low concentrations for preparing spiked solutions. A standard solution containing all four aromatic amines (at 5 ␮g mL−1 ) was prepared by mixing an appropriate volume of 1 mg mL−1 stock solutions in ultrapure water. All solutions were filtered and stored at 4 ◦ C. All aqueous samples with total amine concentrations of 25 ␮g L−1 were freshly prepared before each extraction by spiking the standard mixture into ultrapure water. The aqueous samples for calibration were obtained by spiking ultrapure water with the standard mixture at desired concentrations. Wastewater samples were collected from a sewage treatment plant and stored in glass bottles pre-cleaned with acetone. The bottles were covered with aluminum foil, transported under cooled conditions to the laboratory, and stored in the dark at −20 ◦ C until analysis. No pretreatment of the water samples were done before analysis. -1 1623 cm 2500 2000 1500 1000 -1 Wavenumber (cm ) 500 500 1000 1500 2000 -1 Raman Shift (cm ) Fig. 2. (A) FTIR spectrum and (B) the Raman spectrum of the CNFs. 2500 S. Vadukumpully et al. / J. Chromatogr. A 1218 (2011) 3581–3587 2.4. Preparation of electrospun CNF/PVA composite nanofiber mats An aqueous solution of PVA (10 wt.%) was prepared by dissolving it in ultrapure water at 50 ◦ C with constant stirring for about 12 h. A fixed amount of CNFs were then added to the PVA solution (3 mL). After the CNFs were added into PVA solution, it was sonicated for several hours to make sure that the solution is homogeneous. The weight of CNFs in PVA varied between 0.3 and 7.0 wt.%. The CNF/PVA solution was electrospun into fiber mats using an electrospinning system. In order to reduce the swelling of fibers in water, the CNF/PVA composite fiber mats were placed in 1 vol% glutaraldehyde–acetone solution for 15 h to achieve cross linking. The resultant composite fiber mat was rinsed with ultrapure water followed by acetone and dried at ambient conditions. For the electrospinning system, assembled parts were bought from OptroBio Technologies Pte Ltd., Singapore. The electrospinning set up consists of a plastic syringe with a metal syringe needle, a syringe pump, a high voltage power supply and a grounded flat collector. The inner diameter of the syringe needle used was 0.2 mm and the distance between the syringe tip and the grounded flat collector was maintained at 10 cm. In the electrospinning, the CNF/PVA solution was placed into the plastic syringe and charged with voltage (15 kV) via connecting the syringe needle to the power supply. The flow rate of the CNF/PVA solution was optimized at 5–10 ␮L min−1 . 2.5. Materials characterization FTIR spectra were recorded at 1 cm−1 resolution with 32 scans in the wave number range of 4000–400 cm−1 using a Bio-Rad FTS 165 FTIR spectrophotometer using sample dispersed in a KBr pellet. Scanning electron micrographs (SEM) and Energy Dispersive X-ray (EDX) spectra of the CNFs and electrospun fibers were obtained using a JEOL JSM-6710F scanning electron microscope. Drop-casted samples on glass cover slips were used for SEM. Cover slips were mounted on copper stubs and coated with platinum before analysis. Transmission electron micrographs (TEM) and selected area electron diffraction (SAED) images were taken using JEOL-JEM 2010 transmission electron microscope operating at 200 KeV. The sample solution was drop-casted onto a 300 mesh copper grid and then air dried before imaging. Powder X-ray (XRD) pattern was obtained with a D5005 Siemens X-ray diffractometer with Cu K␣ (1.5 Å) radiation (40 kV, 40 mA). Raman spectroscopy was performed on Renishaw system 2000, with an excitation wavelength of 532 nm. Elemental analysis of the isolated CNFs was obtained using Elementar Vario Micro Cube CHN analyzer and 5 mg of the solid fiber sample was used for the analysis. 2.6. Micro-solid-phase-extraction (-SPE) using electrospun CNF/PVA composite nanofiber mats 10 mg of each of the nanofibers mat was placed in the samples and stirred at 900 rpm for 40 min. After extraction, the mat was removed, rinsed in ultrapure water, dried with lint free tissue and placed in a 250 ␮L auto-sampler vial. The analytes were desorbed by ultrasonication for 15 min in 75 ␮L acetonitrile, out of which 20 ␮L was used for HPLC analysis. The composite membrane can be re-used after ultrasonication in acetonitrile for 20 min to remove traces of impurities. 2.7. HPLC analysis All analyses were performed on a Shimadzu Prominence system (Shimadzu, Kyoto, Japan) consisting of a CBM-20A system 3583 controller, a LC-20AD pump, a SIL-20A auto sampler, a CTO-20A column oven, a DGU-20A5 degasser and a SPD-20A UV–vis detector. Separation of the four aromatic amines at room temperature was accomplished using a 50 mm × 3.0 mm I.D. MetaSil 5 ␮m ODS column (Varian, Palo Alto, CA, USA). The detection wavelength was set at 254 nm. A mobile phase ratio of acetate buffer (pH 3.5)–acetonitrile (85:15, v/v) at a flow rate of 0.3 mL min−1 was used. 3. Results and discussion 3.1. Characterization of the CNFs The carbon soot is obtained via burning common edible oil in air and collecting soot on a cold surface. The raw soot contained aggregated structures of carbon nanoparticles (Fig. 1A) and THF extract from the soot showed high aspect ratio nanofibers (Fig. 1B). From Fig. 1B, it can be seen that the THF extract contained nanofibers in high yield, along with small amounts of other carbon particles. The CNFs were of several micrometers in length and expected to have very high surface area [36–38]. The CNFs isolated were of 20–50 nm in diameter (Fig. S1). The EDX spectrum showed the presence of carbon and oxygen on the CNF surface (Fig. S2). The elemental analysis also confirmed the presence of carbon (86%) and oxygen (12%) in the THF extracted sample. SAED pattern and TEM image indicated amorphous character for the nanofibers (Fig. 1C). There is no interlayer correlation and therefore no significant long range of order to produce lattice diffraction (inset in Fig. 1C). XRD analysis indicated low range of graphitization in the CNFs (Fig. S3). The two predominant peaks at ∼27◦ and 43.5◦ could be assigned for (0 0 2) and (1 0 0) diffraction planes of hexagonal graphite, respectively. The fact that (0 0 2) diffraction peak was relatively low in intensity and broad in shape, suggested that the CNFs have low graphitization and crystallization which in turn supports the SAED observations. The broadening of peak also indicated the presence of some disordered structures in the products [39]. FTIR spectrum of the CNFs (Fig. 2A) showed the peaks at 2915 cm−1 and 2865 cm−1 which correspond to the aromatic –CH stretching. Besides, the sample showed a peak around ∼3400 cm−1 indicating the presence of –OH groups. Presence of oxygen shown in EDX could be from the –OH groups. Similar responses were obtained in the case of MWNTs [26,40]. Bands centered at 1623, 1455, and 1386 cm−1 could be due to (C C) stretching vibrations [41]. In the Raman spectrum (Fig. 2B) of the isolated CNFs, there were two predominant peaks at 1347 and 1589 cm−1 , respectively, corresponding to the disorder induced (D band) and E2g (G-band) mode of graphite. It is known that the D band corresponds to stretching vibrations of sp2 hybridized carbon atoms and G band is associated with disorder-induced symmetry lowering effects [42]. The band at 1347 cm−1 arises from the amorphous graphite particles and defect sites on the carbon nanofibers. 3.2. Analytical evaluation of electrospun CNF/PVA composite nanofiber mats as adsorbent for -SPE The extraction efficiency of the CNF/PVA composite nanofiber mats (Fig. S4) was evaluated based on the peak areas obtained from HPLC analysis (Fig. S5). Extractions were repeated three times to obtain the statistical mean. Fig. 3 gives a pictorial representation of the extraction and desorption steps involved in the electrospun CNF/PVA composite nanofibers mat based ␮-SPE. The performance of CNF/PVA composite nanofiber mats as a ␮-SPE device for the determination of anilines was optimized by investigating several factors which affect the extraction and desorption steps. The factors include extraction time, desorption 3584 S. Vadukumpully et al. / J. Chromatogr. A 1218 (2011) 3581–3587 Fig. 3. Schematic of the extraction and desorption steps involved in electrospun CNF/PVA composite membrane ␮-SPE. solvent, sample volume, desorption time, salting-out effect and pH effect. 3.3. Control study The CNF/PVA nanofiber mat ␮-SPE was compared with PVA fiber mat ␮-SPE to prove that the extraction of aniline is mainly due to interaction of amines with the novel CNF surface instead of the PVA support. In our preliminary investigations, amount of CNFs was varied from 0.3 to 7.0 wt.% in the electrospun mat and used for extraction of amines. However, 5% CNF/PVA combination gave more flexible mat. At higher loading of CNFs, the mat became very brittle which could not be used for routine analyses. Therefore, 5% CNF/PVA was used for further studies. As shown in Fig. 4, the extraction efficiency of the four aniline compounds is greatly enhanced with the incorporation of CNFs into PVA fibers. A comparison of the extraction efficiency between the products obtained before and after THF extraction was also carried out. The results show a better extraction performance of CNFs compared to the soot particles. The extraction efficiency depends on the interaction between anilines and the eletrospun CNF/PVA composite fibers. The adsorption of analytes on the electrospun CNF/PVA composite fibers is facilitated by the presence of electrostatic and hydrophobic interactions between anilines and graphitic sp2 hybridized carbon atoms in the CNFs. In addition, extraction efficiency is also attributed to the large surface area of electrospun nanofibers. 3.4. Extraction time mass transfer is a time-dependent process, the effect of extraction time was investigated. Constant stirring (900 rpm) of sample solution with the aid of magnetic stirring was applied in the experiment in order to facilitate the mass transfer and reduce the time required for equilibrium to be established. The extraction efficiency increase gradually with time and a maximum was achieved in 40 min. Therefore, 40 min was adopted as optimum extraction time for the subsequent experiments. 3.5. Desorption solvent In order to effectively desorb the analytes from CNF/PVA composite fibers, selection of a suitable organic solvent is critical. Factors such as solubility of analytes, solvent polarity, compatibility with HPLC, solubility and swelling of composite fibers were taken into consideration during the extraction. The electrospun CNF/PVA composite fibers are insoluble in organic solvents. Solvents such as methanol, acetonitrile, and acetate buffer were also used for desorption of aniline derivatives. From the results shown in Fig. 5, acetonitrile gave an efficient desorption. The polarity order is methanol > acetonitrile > anilines. Thus, subsequent analyses were carried using acetonitrile as desorption solvent since its polarity is closer to that of anilines. 3.6. Sample volume The effect of sample volume (5–30 mL) on the extraction efficiency was investigated. As shown in Fig. 6, the analyte enrichment and extraction efficiency increases with increase in sample volume. With the high surface area for adsorption of analytes, the electro- Membrane based ␮-SPE is an equilibrium-dependent extraction procedure operating with principle of partitioning analyte on the sorbent material. The amount of analyte extracted depends on the mass transfer from the sample solution to the sorbent. Since Fig. 4. Comparison of PVA and composite fiber mats (extraction conditions are as follows: 10 mL spiked 25 ␮g L−1 water solution, no salt added and pH adjusted, extraction time of 30 min, desorption time of 15 min in 100 ␮L of acetonitrile). Fig. 5. Comparison of different desorption solvents (extraction conditions are as follows: 10 mL spiked 25 ␮g L−1 water solution, no adjustment of salt and pH, extraction time of 40 min, desorption time of 15 min in 100 ␮L of acetonitrile). S. Vadukumpully et al. / J. Chromatogr. A 1218 (2011) 3581–3587 3585 Table 1 Quantitative data: linearity, precision (RSD), limit of detection (S/N = 3), enhancement factors and linear regression data obtained for anilines by the electrospun CNF/PVA composite membrane microextraction coupled with HPLC-UV. Analytes Enrichment factor RSD (%) (n = 3) Correlation coefficient (r) 3-Nitroaniline 4-Chloroaniline 4-Bromoaniline 3,4-Dichloroaniline 66 57 82 109 4.8 5.8 4.5 5.5 0.998 0.989 0.996 0.998 spun membrane can take up significant amount of analytes from the sample. t-Test performed on the data indicated that there is statistical difference in the amount of 3-nitroaniline, 4-bromoaniline and 3,4-dichloroaniline extracted using 10, 20 and 30 mL sample volumes (Table S1). Therefore, 30 mL of samples were used for further studies. 3.7. Desorption time Desorption of analyte was carried out via ultrasonication of the fiber mat with adsorbed analytes in acetonitrile. The desorption profile of anilines ranging from 5 to 10 min was carried out and it was found that the compounds did not show any significant change in peak area with changes in desorption time. Hence desorption time was kept as short as possible without compromising the extraction efficiency and 15 min was chosen as optimum desorption time. 3.8. Effect of ionic strength The salting out effect has been used in SPME [43] and LLE [44]. Generally, addition of sodium chloride enhances the extraction efficiency of some organic compounds, since the solubility of analytes in aqueous solutions decreases with increasing ionic strength. However, no significant variation in the extraction efficiency with increase in salt content in the sample was observed. Another effect of salt addition is the increase in viscosity of sample solution, which in turn, has impeded the mass transfer rate and reduced extraction efficiency. Therefore, no salt addition was used in further experiments. 3.9. Effect of pH The influence of pH in the extraction was evaluated since pH plays a significant role in ionizable compounds such as ani- Fig. 6. Influence of sample volume on the extraction efficiency (extraction conditions are as follows: 10 mL spiked 25 ␮g L−1 water solution no adjustment of salt content and pH, extraction time of 40 min, desorption time of 15 min in 75 ␮L of acetonitrile). Linearity range (␮g L−1 ) 0.5–50 0.5–50 0.5–50 0.5–50 LOD (␮g L−1 ) LOQ (␮g L−1 ) 0.009 0.024 0.081 0.019 0.030 0.080 0.269 0.062 lines. The pH values of water sample were adjusted using 1 M NaOH for alkaline sample and 1 M HCl for acidic sample before spiking the analytes. As shown in Fig. 7, the extraction efficiency for the four compounds increases with the increase in pH initially and level off after pH 7. pKa values for the deprotonation of protonated salts of 3-nitroaniline, 4-chloroaniline, 4bromoaniline and 3,4-dichloroaniline are 2.47, 4.15, 3.86 and 2.97, respectively. At acidic conditions, all the four aniline compounds are expected to be protonated and the resulting cations have poorer affinity over their neutral form for the surface of CNF/PVA composite fibers. Hence the basic condition of pH 7 (without any salt addition) was selected as optimum condition. 3.10. Method validation To assess the feasibility of CNF/PVA electrospun nanofiber mat as extraction device, the validation was established for the determination of several performance parameters such as linearity, precision, limit of detection (LOD) and enrichment factors under the optimized extraction conditions. Analyte enrichment factor is defined as the ratio of analyte concentration in the final extract to its initial concentration in the original sample. Table 1 summarizes the analytical data obtained using the CNF/PVA membrane extraction. Calibration curves were obtained by plotting the peak area versus the corresponding spiked concentrations in ultrapure water. The linearity of the calibration plot was evaluated over a range of 0.5–50 ␮g L−1 by least square linear regression analysis. All the four amine compounds exhibited good linearity with correlation coefficient (r) above 0.9897. This shows a directly proportional relationship between the extracted amount of analytes and the initial spiked concentration. The precision, which is expressed as relative standard deviation (RSD), were determined on triplicate analyses at different concentrations. The RSD obtained were in the range of 4.5–5.8%. The good repeatability is mainly due to simple extraction methodology and use of auto-sampler in HPLC-UV for the analysis. The LOD determined for the aniline compounds are in range of 0.009–0.081 ␮g L−1 and the limit of quantitation (LOQ) is within a range of 0.030–0.269 ␮g L−1 . Our data are consistent with the LOD Fig. 7. Effect of sample pH on the extraction efficiency (extraction conditions as follows: 30 mL spiked 25 ␮g L−1 water solution no salt addition, extraction time of 40 min, desorption time of 15 min in 75 ␮L of acetonitrile). 3586 S. Vadukumpully et al. / J. Chromatogr. A 1218 (2011) 3581–3587 Table 2 Concentrations of target aniline compounds (␮g L−1 ) in wastewater samples and the average recoveries determined at 5 ␮g L−1 . Analytes 3-Nitroaniline 4-Chloroaniline 4-Bromoaniline 3,4-Dichloroaniline Environmental water sample Relative recovery (%) A B C D E 2.3 n.d. 6.9 n.d. n.d. n.d. 0.46 n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. 0.29 0.81 n.d. n.d. 85 71 108 70 n.d. – not detected. Relative recovery values of spiked wastewater sample at 5 ␮g L−1 compared to that of spiked pure water. whatever reported for extraction of anilines in aqueous samples [30] using LLLME. for standard samples content are available free of charge via the internet. Acknowledgments Authors thank National University of Singapore (NUS) for the financial support, Department of Chemistry, NUS for technical assistance and Sajini Vadukumpully thanks NUS Nanoscience and Nanotechnology Initiative (NUSNNI) for the graduate scholarship. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.chroma.2011.04.003. References 3.11. Application of method for water analysis The current method was developed as pre-treatment for analysis of aniline compounds in environmental water samples. Water samples were analyzed to assess the contamination of the four aniline compounds. 3 out of 5 samples were found to contain some of the aniline compounds and results are shown in Table 2. However, the complex matrix of environmental sample affects the performance of the composite fiber mat in the extraction process. A standard addition method was used to study the matrix effect and relative recovery of the aniline compounds. From the results shown in Table 2, it could be seen that relative recoveries for 4-chloroaniline and 3,4-dichloroanilines are comparatively lower, which is about 70%. The low recovery could be due to the clogging of nanofibers surface by the particulate matters present in the sample. There might be strong competition between analytes of interest and other organic compounds in the waste water. Hence it can be concluded that CNF/PVA composite nanofibers are suitable for pretreated water analysis to assess the performance of treatment plant in removing amine compounds. The proposed extraction method using the CNF/PVA composite nanofibers is more cost-effective since it is robust, durable and reusable. Furthermore, it is fast and simple involving only a few steps. 4. Conclusions Carbon nanofibers were isolated from carbon soot by a simple and effective methodology in good yield. The isolated CNFs were characterized in detail by spectroscopic and microscopic techniques. The CNFs were several 100 nm in length and 20–50 nm in diameters. The high surface area electrospun CNF/PVA composite membrane has been successfully developed as a novel sorbent material for micro-extraction of aniline compounds. The presence of CNFs in the composite membrane ensures and enhances the extraction efficiency of method for aniline compounds under optimum condition. The composite membrane sorbent yielded satisfactory parameters for micro-extraction. Finally, the applicability of the method was further validated to determine aniline compounds in environmental samples. The proposed microextraction method offers advantages such as easy operation, high recovery, fast extraction, minimal use of organic solvent and elimination of tedious solvent evaporation and reconstitution steps. Supporting information EDX spectra of the carbon nanofibers, XRD of the soot particles and the CNFs, SEM image of the CNF/PVA fiber mats, t-test results on the effect of sample volume and the HPLC chromatogram obtained [1] H.W. Kroto, J.R. Heath, S.C. O’Brien, R.F. Curl, R.E. Smalley, Nature 318 (1985) 162. [2] S. Iijima, Nature 354 (1991) 56. [3] R.H. Baughman, C.X. Cui, A.A. Zakhidov, Z. Iqbal, J.N. Barisci, G.M. Spinks, G.G. Wallace, A. Mazzoldi, D.D. Rossi, A.G. Rinzler, O. Jaschinski, S. Roth, M. Kertesz, Science 284 (1999) 1340. [4] S.S. Fan, M.G. Chapline, N.R. Franklin, T.W. Tombler, A.M. Cassell, H.J. Dai, Science 283 (1999) 512. [5] J. Kong, N.R. Franklin, C.W. Zhou, M.G. Chapline, S. Peng, K.J. Cho, H.J. Dai, Science 287 (2000) 622. [6] R.H. Baughman, A.A. Zakhidov, W.A. de Heer, Science 297 (2002) 787. [7] E.T. Thostenson, Z.F. Ren, T.W. Chou, Compos. Sci. Technol. 61 (2001) 1899. [8] C. Liu, Y.Y. Fan, M. Liu, H.T. Cong, H.M. Cheng, M.S. Dresselhaus, Science 286 (1999) 1127. [9] J. Lee, H. Kim, S.J. Kahng, G. Kim, Y.W. Son, J. Ihm, H. Kato, Z.W. Wang, T. Okazaki, H. Shinohara, Y. Kuk, Nature 415 (2002) 1005. [10] J.B. Howard, C.K. Das, J.B. VanderSande, Nature 370 (1994) 603. [11] C.K. Das, J.B. Howard, J.B. VanderSande, J. Mater. Res. 11 (1996) 341. [12] H. Richter, K. Hernadi, R. Caudano, A. Fonseca, H.N. Migeon, J.B. Nagy, S. Schneider, J. Vandooren, P.J. VanTiggelen, Carbon 34 (1996) 427. [13] S. Sarkar, P. Dubey, D.P. Mukhopadhyay, Pramana-J. Phys. 65 (2005) 681. [14] A. Thess, R. Lee, P. Nikolaev, H.J. Dai, P. Petit, J. Robert, C.H. Xu, Y.H. Lee, S.G. Kim, A.G. Rinzler, D.T. Colbert, G.E. Scuseria, D. Tomanek, J.E. Fischer, R.E. Smalley, Science 273 (1996) 483. [15] Z.J. Shi, Y.F. Lian, X.H. Zhou, Z.N. Gu, Y.G. Zhang, S. Iijima, H.D. Li, K.T. Yue, S.L. Zhang, J. Phys. Chem. B 103 (1999) 8698. [16] M.J. Bronikowski, P.A. Willis, D.T. Colbert, K.A. Smith, R.E. Smalley, J. Vac. Sci. Technol. A 19 (2001) 1800. [17] E. Borowiak-Palen, T. Pichler, X. Liu, M. Knupfer, A. Graff, O. Jost, W. Pompe, R.J. Kalenczuk, Chem. Phys. Lett. 363 (2002) 567. [18] D.S. Bethune, C.H. Kiang, M.S. de Vries, G. Gorman, R. Savoy, J. Vazquez, R. Beyers, Nature 363 (1993) 605. [19] H. Kajiura, S. Tsutsui, H.J. Huang, Y. Murakami, Chem. Phys. Lett. 364 (2002) 586. [20] J.N. Coleman, A.B. Dalton, S. Curran, A. Rubio, A.P. Davey, A. Drury, B. McCarthy, B. Lahr, P.M. Ajayan, S. Roth, R.C. Barklie, W.J. Blau, Adv. Mater. 12 (2000) 213. [21] A.G. Rinzler, J. Liu, H. Dai, P. Nikolaev, C.B. Huffman, F.J. Rodriguez-Macias, P.J. Boul, A.H. Lu, D. Heymann, D.T. Colbert, R.S. Lee, J.E. Fischer, A.M. Rao, P.C. Eklund, R.E. Smalley, Appl. Phys. A 67 (1998) 29. [22] H.B. Jia, Y.F. Lian, M.O. Ishitsuka, T. Nakahodo, Y. Maeda, T. Tsuchiya, T. Wakahara, T. Akasaka, Sci. Technol. Adv. Mater. 6 (2005) 571. [23] K.B. Shelimov, R.O. Esenaliev, A.G. Rinzler, C.B. Huffman, R.E. Smalley, Chem. Phys. Lett. 282 (1998) 429. [24] S. Bandow, A.M. Rao, K.A. Williams, A. Thess, R.E. Smalley, P.C. Eklund, J. Phys. Chem. B 101 (1997) 8839. [25] P. Liang, Y. Liu, L. Guo, J. Zeng, H. Lu, J. Anal. At. Spectrom. 19 (2004) 1489. [26] C. Basheer, A.A. Anass, B.S.M. Rao, H.K. Lee, S. Valiyaveettil, Anal. Chem. 78 (2006) 2853. [27] G.P. Tang, M.S. Pfeifer, M.F. Denissenko, Z. Feng, W. Hu, A. Pao, Y. Zheng, J.B. Zheng, H. Li, J.X. Chen, Int. J. Hyg. Environ. Health 205 (2002) 103. [28] J.F. Jen, C.T. Chang, T.C. Yang, J. Chromatogr. A 930 (2001) 119. [29] A. Pielesz, Chromatographic Analysis of the Environment, CRC Press–Taylor & Francis, Boca Raton, FL, 2006, p. 377. [30] S.M. Lloret, C.M. Legua, P.C. Falcó, J. Chromatogr. A 978 (2002) 59. [31] B. Jurado-Sánchez, E. Ballesteros, M. Gallego, J. Chromatogr. A 1154 (2007) 66. [32] B. Jurado-Sánchez, E. Ballesteros, M. Gallego, Talanta 79 (2009) 613. [33] W.Y. Chang, Y.H. Sung, S.D. Huang, Anal. Chim. Acta 495 (2003) 109. [34] H. van Doorn, C.B. Grabanski, D.J. Miller, S.B. Hawthorne, J. Chromatogr. A 829 (1998) 223. [35] L. Zhu, C.B. Tay, H.K. Lee, J. Chromatogr. A 985 (2003) 167. [36] V. Vamvakaki, K. Tsagaraki, N. Chaniotakis, Anal. Chem. 78 (2006) 5538. [37] J. Li, M.J. Vergne, E.D. Mowles, W.H. Zhong, D.M. Hercules, C.M. Lukehart, Carbon 43 (2005) 2883. [38] R. Zheng, Y. Zhao, H. Liu, C. Liang, G. Cheng, Carbon 44 (2006) 742. S. Vadukumpully et al. / J. Chromatogr. A 1218 (2011) 3581–3587 [39] D. Takayuki, F. Akihito, Y. Iriyama, A. Takeshi, Z. Ogumi, N. Kiyoharu, A. Toshihiro, Electrochem. Commun. 7 (2005) 10. [40] X.Y. Liu, Y.S. Ji, H.X. Zhang, M. Cang, J. Chromatogr. A 1212 (2008) 10. [41] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Co-ordination Compounds, 4th ed., John Wiley & Sons, New York, 1986. 3587 [42] K. Tanaka, T. Yamabe, K. Fukui, The Science and Technology of Carbon Nanotubes, 2nd ed., Elsevier, Amsterdam, Lausanne, New York, Oxford, Shannon, Singapore, Tokyo, 1999. [43] C. Grote, E. Belau, K. Levsen, G. Wünsch, Acta Hydrochim. Hydrobiol. 27 (1999) 193. [44] M.T. Kelly, D. McGuirk, F.J. Bloomfield, J. Chromatogr. B 668 (1995) 117.