Metal Carbonyl Vapor Generation Coupled with Dielectric Barrier

Dec 16, 2014 - Metal Carbonyl Vapor Generation Coupled with Dielectric Barrier. Discharge To Avoid Plasma Quench for Optical Emission. Spectrometry...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Metal Carbonyl Vapor Generation Coupled with Dielectric Barrier Discharge To Avoid Plasma Quench for Optical Emission Spectrometry Yi Cai,† Shao-Hua Li,§ Shuai Dou,† Yong-Liang Yu,*,†,‡ and Jian-Hua Wang*,†,∥ †

Research Center for Analytical Sciences, and ‡Department of Chemistry , College of Sciences, Northeastern University, Box 332, Shenyang 110819, China § Hebei First Environmental Protection Technology Co., LTD, Shijiazhuang 050035, China ∥ Collaborative Innovation Center of Chemical Science and Engineering, Tianjin 300071, China S Supporting Information *

ABSTRACT: The scope of dielectric barrier discharge (DBD) microplasma as a radiation source for optical emission spectrometry (OES) is extended by nickel carbonyl vapor generation. We proved that metal carbonyl completely avoids the extinguishing of plasma, and it is much more suitable for matching the DBD excitation and OES detection with respect to significant DBD quenching by concomitant hydrogen when hydride generation is used. A concentric quartz UV reactor allows sample solution to flow through the central channel wherein to efficiently receive the uniformly distributed UV irradiation in the confined cylindrical space between the concentric tubes, which facilitates effective carbonyl generation in a nickel solution. The carbonyl is transferred into the DBD excitation chamber by an argon stream for nickel excitation, and the characteristic emission of nickel at 232.0 nm is detected by a charge-coupled device (CCD) spectrometer. A 1.0 mL sample solution results in a linear range of 5−100 μg L−1 along with a detection limit of 1.3 μg L−1 and a precision of 2.4% RSD at 50 μg L−1. The present DBD−OES system is validated by nickel in certified reference materials.

H

excitation source instead of conventional ICP has received more attention, and it was first used for the determination of mercury.18,19 For further development of DBD as an excitation source for more metal species, special interest has been focused on the exploitation of a novel and efficient sample introduction approach to match DBD excitation and to extend the applications of DBD−OES systems. DBD microplasma is formed by applying high-frequency voltage between two electrodes filled with dielectric barriers.22 Direct introduction of sample solution into DBD is quite problematic, as evaporation of solvent and matrix consumes most of its energy, which makes it incapable of exciting the target species or even extinguishing DBD. Due to the limited atomization/excitation capability of DBD, it is only feasible for direct excitation of species on the surface of a very small amount of solution,23,24 which highly confines sampling volume and deteriorates the detection sensitivity. Instead, the introduction of gaseous species, e.g., ammonia25 and nitrogen,26 into DBD is preferential and avoids the above problem. In addition, electrothermal vaporization (ETV)27 and chemical vapor generation (CVG) facilitate the introduction of a few

eavy metal quantification plays an important role in the fields of environmental science and clinical treatment, especially for highly toxic metal/metalloid species.1,2 Among numerous alternative analytical methodologies, atomic spectrometry and inductively coupled plasma−mass spectrometry have been established as the most attractive options for sensitive and accurate determination of trace metals or their species. However, conventional instrumentations are in general costly and bulky/heavy due to their high temperature atomization/radiation and ionization sources, in addition to complicated optical systems. Thus, miniaturization as portable detectors or for on-site and real-time analysis is not easy. Nowadays, more attention is directed to the development of miniaturized analytical instrumentations for field analysis.3,4 Microplasmas, e.g., dielectric barrier discharge (DBD),5 glow discharge,6 microhollow-cathode discharge,7 microfabricated inductively coupled plasma,8 capacitively coupled plasma,9 and microwave microstrip plasma10 are suitable for the development of miniaturized analytical devices due to their small size and low power consumption. Especially, due to its high dissociation/excitation/ionization capability, DBD microplasma is well suited as an atomization source for AAS and AFS,4,11−13 chemical vapor generators,14,15 gas chromatographic detectors,16,17 excitation sources,18,19 and ionization sources.20,21 In particular, for OES, DBD microplasma as a nonthermal © 2014 American Chemical Society

Received: November 12, 2014 Accepted: December 16, 2014 Published: December 16, 2014 1366

DOI: 10.1021/ac5042457 Anal. Chem. 2015, 87, 1366−1372

Article

Analytical Chemistry

Figure 1. Schematic diagram of the DBD−OES system for nickel determination based on UV-induced carbonyl generation. SP: syringe pump; SV: six-port selection valve; PP: peristaltic pump; HC: holding coil; GLS: gas−liquid separator; MF: mass flowmeter. The dashed part stands for the UVinduced carbonyl generation unit, and the photograph of concentric quartz UV-induced reactor is shown.

A stock solution of 1000 mg L−1 nickel is prepared by dissolving 0.5 g of spectropure nickel powder in 10 mL of 50% (v/v) HNO3 and diluted to 500 mL. After stepwise dilution of the stock solution, the final working standards of nickel are prepared in 15% (v/v) formic acid. Formic acid is distilled at 100.8 °C (boiling point) for its purification prior to use. Deionized water is used as carrier solution. Sample Pretreatment. The validation of the DBD−OES system for nickel determination is performed by using certified reference materials (National Center for Standard Materials, China), including GBW09101b (human hair), GBW10023 (laver), and GBW08608 (water). The sample pretreatment procedure is given in the following: A 0.1 g amount of human hair and laver each is immersed separately into 2 mL of HNO3 and 2 mL of HNO3 with 1 mL of H2O2 in a PTFE tank for microwave digestion (COOLPEX microwave digestion system, Yiyao Instrument Technology Development Co., LTD, Shanghai, China), and details for microwave digestion parameters are summarized in Table S1, Supporting Information. After that, the residue of acid is heated gently to near dryness and diluted with deionized water to 5 mL. For the 5 mL GBW08608 (water) sample, it contains 0.5 mol L−1 nitric acid, which might interfere with the determination of nickel in the DBD−OES system. The nitrate is thus eliminated by heating gently to near dryness and diluted with deionized water to 5 mL. As the high content of nitrate (>25 mg L−1) might interfere with nickel determination by DBD−OES, removal of nitrate in the sample solution is required by passing through a 717 anionexchange resin column, where nitrate is exchanged with chloride ion. All the effluent is collected and diluted to 10 mL to ensure that the final solution contains 15% (v/v) formic acid. The filtrates are used for nickel determination by DBD− OES. Spiking recovery of nickel in simulated water samples is also performed to further evaluate the performance of the DBD−OES system.

other analytes into DBD. The latter is particularly useful for those species capable of forming vapor or volatile species, including mercury,18,19 thimerosal,28 iodine,29 bromine,30 and arsenic.31 It should take into account that during the vapor generation process the concomitant gaseous matrix might interfere with the detection of analyte. In particular, a large amount of concomitant hydrogen during hydride generation (HG) would deteriorate the excitation capability of DBD32,33 or even extinguish DBD. That is why, so far, only very few analytes have been detected in such a way, and suitable CVG approaches are highly desired to match the excitation behavior of DBD and extend the scope of analytical applications.34,35 UV-induced CVG has been proved to be an effective approach for sample introduction in DBD−OES,36,37 where about 20 elements could be transferred into volatile species under UV irradiation in the presence of low molecular weight carboxylic acids. Its distinct feature is the avoidance of formation of concomitant gaseous species which tend to have a significant influence on the microplasma as encountered in HG.38 In the present study, UV-induced formation of metal carbonyl as an ideal vapor generation scheme has been proved to match with DBD−OES without microplasma quenching. It is demonstrated that nickel carbonyl can be readily excited in situ in the DBD excitation source to produce its characteristic emission spectra. This feature is fully demonstrated in comparison with HG wherein plasma quenching is frequently encountered. In addition, a few important parameters governing the performance of the entire system were investigated. The reliability and practicability of the UVinduced carbonyl generation DBD−OES system are validated by determination of trace nickel in a series of real sample matrixes.

■

EXPERIMENTAL SECTION Reagents. All the reagents used in this study are at least of analytical reagent grade and purchased from Sinopharm Chemical Reagent Co. (China-SCRC) unless specified otherwise. Deionized water (18 MΩ cm) is used throughout. 1367

DOI: 10.1021/ac5042457 Anal. Chem. 2015, 87, 1366−1372

Article

Analytical Chemistry

Figure 2. Optical emission spectra of the argon DBD microplasma recorded in the absence (left) and presence (right) of hydrogen: (a) pure argon, (b) nickel carbonyl generated by 1 mL of 300 μg L−1 nickel solution, (c) argon mixed with 15% (v/v) hydrogen, (d) nickel carbonyl generated by 1 mL of 300 μg L−1 nickel solution in argon mixed with 15% (v/v) hydrogen.

Experimental Setup and Operating Procedure. The configuration of the DBD−OES system is illustrated in Figure 1, with an UV-induced carbonyl generation unit, a DBD microplasma excitation chamber, and a CCD detector. The carbonyl generation unit includes a concentric quartz UV reactor with a 40 W UV-lamp (provided by Titan Instruments Co. Ltd., Beijing, China). The central tube serves as a sample flow-through channel with a capacity of ca. 1.0 mL (1.2 mm i.d., 930 mm length), while the outer tube (18 mm i.d.) offers the irradiation region. Such a design ensures a uniform distribution of the UV irradiation in the cylindrical space between the concentric tubes. UV irradiation around the sample solution in the central channel well facilitates effective production of the metal carbonyl. The sample solution containing formic acid is directed to flow through the UV-induced carbonyl generation unit by a microsequential injection system (FIAlab Instruments, Bellevue, WA). It is furnished with a 5 mL syringe pump and a six-port selection valve. The flow channels are made of 0.8 mm i.d PTFE tubing in connection with the selection valve. For the sake of sample introduction at the milliliter level, the holding coil was designed to have a capacity of ca. 2.0 mL. The DBD excitation chamber is fabricated to fulfill the requirements for generating atmospheric pressure DBD microplasma. It is a piece of quartz tube (1.3 mm i.d., 3.0 mm o.d., 100 mm length) providing a small gas path for the introduction of an argon stream. Two copper ring electrodes (2 mm length) are used to surround the outside of the quartz tube at a distance of 30 mm. It generates DBD microplasma when applying a high-frequency and high-voltage electric field from a neon power supply (ENT-106B, Xinxing Neon Light Supply Company Ltd., Guangzhou, China), which provides a sinusoidal discharging voltage of 2−4 kV with a frequency of ca. 40 kHz. The nickel carbonyl vapor isolated from the gas−liquid separator is then introduced into the DBD microplasma excitation chamber and excited therein. The argon flow is controlled by a mass flowmeter (FL-802, Flowmethod Measure and Control Systems Co., Ltd., Shenzhen, China). An AvaSpec-ULS2048-4-USB2 charge-coupled device (CCD) spectrometer (Avantes, Netherlands) combined with a fiber-optic probe (950 μm core i.d. and 20 cm length) furnished with a collimating lens (74-UV, Ocean Optics, Dunedin, FL) is placed at 5 mm distance to the side of the DBD microplasma excitation chamber for recording the optical emission spectra from 200 to 365 nm. A 25 μm slit is set for the CCD spectrometer, incorporating an 1800 lines/mm grating

which offers a spectral resolution of ca. 0.2 nm. An integration time of 300 ms and an average of five scans are employed. For each analysis run, the syringe pump aspirates 0.01 mL of air, 1.0 mL of sample solution, and 0.01 mL of air sequentially into the holding coil through port 3, port 1, and port 3, respectively. The sample solution zone stacked by two air segments followed by 1.52 mL of carrier (deionized water) is then directed to flow through the UV-induced reactor and facilitates the production of nickel carbonyl. After a 30 s delay, the reaction mixture followed by 2 mL of carrier is dispensed at 16.8 mL min−1 to meet an argon stream at 400 mL min−1 that enters the gas−liquid separator, where volatile nickel carbonyl is separated and the rest of the reaction mixture is directed to waste by a peristaltic pump. The nickel carbonyl is further transferred into the DBD excitation source for facilitating optical emission which is recorded by a CCD spectrometer. Peak height at 232.0 nm is used for quantification.

■

RESULTS AND DISCUSSION Emission Spectral Characteristics of Nickel with Excitation in DBD. In OES analysis, the background emission spectra generally exhibit significant effects on the analysis of target species. The background emission spectra of argon DBD are relatively low and clean within the range of 200−280 nm, where the sensitive atomic emission lines of most metal species are located. Therefore, it is beneficial to acquire a sensitive and low-interference OES signal for metal species within this spectral region. Figure 2 illustrates the blank emission spectra of argon DBD and that recorded in the presence of nickel carbonyl vapor generated by 1.0 mL of sample solution containing 300 μg L−1 of nickel. It is obvious that a series of characteristic atomic emission lines of nickel, e.g., 231.096, 231.234, 232.003, 232.138, and 232.579 nm, are clearly isolated from the blank emission spectra of argon DBD. This well demonstrated the feasibility for the atomization and excitation of nickel carbonyl in the DBD microplasma excitation chamber. The most sensitive Ni 232.003 nm emission line is suitable to serve as an analytical line for OES quantification, which corresponds to the electronic transition from 3G°5 to 3F4 between the electronic states of 3d8(3F)4s4p(1P°) and 3d8(3F)4s2.39 It is obvious that these sensitive atomic emission lines are closely distributed in a very narrow region within 231.096−232.579 nm, and a resolution of 0.2 nm for the CCD spectrometer is not sufficient to distinguish the adjacent emission lines at 232.003 and 232.138 nm. Fortunately, by use 1368

DOI: 10.1021/ac5042457 Anal. Chem. 2015, 87, 1366−1372

Article

Analytical Chemistry of the present CCD spectrometer, it is feasible for the quantification of nickel within a certain concentration range by measuring its emission at 232.0 nm, which is the joint contribution by both emission lines of 232.003 and 232.138 nm. It is noteworthy that vapor generation for iron and cobalt is similar to that for nickel,37 and thus their volatile species can also be detected by DBD−OES. The characteristic emission lines for iron are 248.3, 248.8, and 249.0 nm, while those for cobalt are 240.7, 242.5, and 252.1 nm. It is thus promising to detect more metals by using a similar DBD−OES system. In practice, the fluctuation of blank argon DBD microplasma tends to cause an obvious effect on the stability of nickel emission; thus, background correction is performed by adopting a previously described method18 by simultaneously detecting the background emission intensity at 233.0 nm and that of the nickel emission at 232.0 nm. A net emission intensity of nickel at 232.0 nm is derived by deducting the background signal from the raw nickel signal. Suitability of Vapor Generation with DBD−OES. The gaseous matrix produced by vapor generation has a significant influence on the emission spectra of DBD microplasma. Especially, the high content of gaseous matrix species might interfere with the detection of target analytes by DBD−OES. The present study demonstrates that carbonyl generation is an excellent sample introduction approach for trace levels of nickel, as it is well matched with the excitation mode of DBD− OES. On the contrary, when hydride generation (HG) is employed, a large amount of concomitant hydrogen would be produced by reduction with NaBH4, which significantly deteriorates the excitation capability of DBD and even quenches the microplasma; thus, no nickel emission is obtained. Figure 2 illustrates that in comparison with the background emission spectra of argon DBD microplasma, the background emission is obviously increased when the barrier gas is a mixture of 15% (v/v) hydrogen and 85% argon. In the meantime, the characteristic emission lines for nickel are significantly suppressed or even become unidentifiable. In the presence of hydrogen, a series of reactions take place between argon and hydrogen in the DBD microplasma to form a plasma jet comprising Ar+, H+, H2+, H3+, ArH+, and electrons,32 which remarkably decreases the electron and Ar+ density. This becomes even more significant with the increase of hydrogen content in the barrier gas, and thus the excitation capability of the DBD microplasma is extremely deteriorated. The effect of hydrogen on the emission spectra of nickel is further investigated in the DBD excitation source, as illustrated in Figure 3. The emission of nickel starts to drop significantly as the content of hydrogen exceeds 1% (v/v) in the barrier gas. The emission of nickel is almost quenched completely in the presence of 15% (v/v) hydrogen. When HG is used for nickel vapor production, e.g., 0.3 mol L−1 HCl confluences with 1.0% (m/v) NaBH4 at the same flow rate of 2 mL min−1 to generate volatile nickel hydride,40 a large amount of concomitant hydrogen with a content of more than 10% (v/v) will be produced in a 400 mL min−1 argon flow stream. This will significantly suppress the emission of nickel in the DBD microplasma. In fact, it is even worse that no nickel emission spectrum is observed in our practice. This might be due to uneven distribution of hydrogen in the barrier gas, giving rise to more significant quenching of the DBD microplasma. Therefore, it is conclusive that HG is not suitable for coupling with DBD microplasma as an excitation source.

Figure 3. Dependence of optical emission intensity of nickel on the concentration of hydrogen. Full details of the experimental parameters are given in Table 1.

Excitation Capability of DBD Microplasma. Generally, two typical DBD configurations can be used as an excitation chamber, i.e., parallel-plate and cylindrical DBD as illustrated in Figure 4. Both configurations can be employed for generating

Figure 4. Schematic diagrams of (a) parallel-plate and (b) cylindrical DBD configurations with photographs of argon and helium DBD microplasma. Full details of the experimental parameters are given in Table 1.

stable DBD microplasma with argon or helium as the barrier.5 It is noteworthy that no optical emission of nickel is observed with argon parallel-plate DBD as an excitation source by setting a smaller distance between the two dielectric layers at 0.7 or 1.5 mm. However, nickel optical emission can only be obtained at a wider distance between the two dielectric layers, e.g., 2.5, 3.5, 4.5 mm. This observation clearly indicates that the gas gap 1369

DOI: 10.1021/ac5042457 Anal. Chem. 2015, 87, 1366−1372

Article

Analytical Chemistry

irradiation system well facilitate effective production of the metal carbonyl. At a fixed intensity of UV irradiation, the efficiency of nickel carbonyl generation closely depends on the irradiation time and formic acid concentration. As illustrated in Figure S2, Supporting Information, the optical emission intensity of nickel is significantly improved by increasing the UV irradiation time within 0−30 s. Thereafter, the emission intensity falls into a plateau with further irradiation due to the equilibrium of vapor generation. Similarly, stable emission is obtained by setting the formic acid concentration within 15−50% (v/v). For the ensuing studies, nickel carbonyl generation is performed in 15% (v/v) formic acid with UV irradiation for 30 s. The efficiency of UV-induced carbonyl generation for nickel is investigated by using ICP-MS, and Figure 5 shows that an efficiency of ca. 60% is derived.

between the two dielectric layers exhibits significant influence on the feature of argon DBD microplasma. An argon microplasma generated in a parallel-plate DBD is a combination of homogeneous and filamentary discharge. The filamentary discharge becomes the dominant influence on electron energy with an increase in distance between the two dielectric layers. For the filamentary mode, there are a large number of shortlived current filaments, i.e., microdischarges, in the discharge gap.41 In these microdischarge channels, the electrons possess energy of a few eV, but the electron and current densities are far higher than those obtained in the homogeneous mode. The high-energy electrons collide with the ambient gas molecules to produce various radicals and ions, which highly promote molecular dissociation and excitation. In a helium microplasma generated by a parallel-plate DBD with a gas gap of 0.7−4.5 mm at atmospheric pressure (Figure 4), no optical emission of nickel is detected. This is due to the fact that helium generates a homogeneous microplasma in the above condition,42 where the relatively low electron and current densities reduce the capability of DBD for molecular dissociation and excitation. On the other hand, however, in a cylindrical DBD excitation source, optical emission of nickel could be readily recorded, and a much higher emission is obtained in argon DBD microplasma with respect to that achieved in helium DBD. This is because filamentary discharge in this mode of DBD microplasma becomes more apparent due to a relatively wider discharge gap between the two electrodes. In comparison with parallel-plate DBD, the cylindrical configuration has a small dead volume and the emission can be detected side-on from the microplasma which effectively avoids contamination of the plasma jet on the surface of optical fiber by end-on mode detection. For further investigations, we use an argon cylindrical DBD as the excitation source. The characteristics of argon cylindrical DBD microplasma are affected by a series of parameters, e.g., discharging voltage, distance between electrodes, and argon flow rate. Our experiments have indicated that at a lower discharging voltage, there is no DBD microplasma generated, while a too high discharging voltage causes air breakdown outside the cylindrical excitation source. Therefore, in practice the discharging voltage between the two electrodes should be better set within a range of 2.9−3.5 kV. Specifically, we use an input voltage of 50 VAC corresponding to a 3.2 kV discharging voltage for the generation of argon DBD microplasma. The distance between the two electrodes and argon flow rate are investigated by simple optimization, as illustrated in Figure S1, Supporting Information. A distance of 30 mm and an argon flow rate of 400 mL min−1 are employed for further experiments. Nickel Carbonyl Generation Efficiency by UV Irradiation. In the present system, volatile nickel carbonyl Ni(CO)4 is generated by the reaction of Ni2+ with formic acid under UV irradiation. The effective utilization of UV light plays an important role in the improvement of the rate and efficiency of carbonyl generation. In this respect, the design of the UVinduced reactor is highly important. In the present study, the heart of the UV-induced carbonyl generation unit is a concentric quartz reactor (Figure 1). The central tube is used as a sampling channel, and the outer tube integrates an UV lamp and serves as UV-irradiation source. The UV irradiation from the concentric system is uniformly distributed around the sample solution in the central tube. The direct irradiation in a very short distance to the sample solution effectively avoids energy loss in the form of UV light. These features of the UV

Figure 5. UV-induced carbonyl generation efficiency (histograms) of nickel obtained under various concentration levels of nickel solution as well as the calibration curve of nickel for the DBD−OES system. Full details of the experimental parameters are given in Table 1.

Interfering Effect and Its Elimination. To identify the influence of sample matrix components, in terms of matrix effect and spectral interferences, some foreign species frequently encountered in common real biological and environmental samples are spiked at 50 μg L−1 in nickel standard solution, which is afterward processed by the present DBD−OES system followed by detection. It is found that for the processing and detection of 50 μg L−1 Ni2+, 5000-fold Ca2+, Zn2+, Cr3+, Cl−, SO42−, and CO32−, 2000-fold K+, Na+, and Mg2+, 500-fold NO3−, 50-fold Fe3+ and Co2+, and 30-fold Cu2+ have no obvious influence within an error range of ±5%. On the other hand, although some transition metal cations form carbonyl and enter the DBD excitation source, no spectral interferences are identified by adopting a CCD system with a 0.2 nm spectral resolution, by which the emission lines at 248.3, 248.8, 249.0 nm for Fe and 240.7, 242.5, 252.1 nm for Co are clearly separated from that of the emission line of nickel at 232.0 nm. In the processing of real samples, nitric acid is frequently used for sample pretreatment or digestion. In this respect, a 500-fold tolerant ratio for NO3− is not sufficient, and thus an additional procedure is required for the elimination of its 1370

DOI: 10.1021/ac5042457 Anal. Chem. 2015, 87, 1366−1372

Article

Analytical Chemistry interference. In practice, NO3− is eliminated by using anionexchange resin. Analytical Performance Evaluation and System Validation. At the experimental conditions discussed previously, a vapor generation efficiency of ca. 60% is obtained. The characteristic analytical performance data of the present DBD−OES system for nickel determination are summarized in Table 1. By adopting a sample volume of 1 mL, a linear

carbonyl. There is no concomitant hydrogen formation accompanying the use of nickel carbonyl; thus, DBD quenching by hydrogen is effectively avoided, as frequently encountered when hydride generation is used for sample processing, and the excitation capability of the DBD source is significantly improved. This demonstrates the suitability of metal carbonyl for matching the DBD excitation source, and it well extends the scope of applications for the DBD−OES system.

■

Table 1. Experimental Parameters for Nickel Carbonyl Formation and Analytical Performance Data for the DBD− OES System with Nickel Carbonyl Generation parameter

value

detection wavelength discharging voltage distance between two electrodes argon flow rate UV irradiation time formic acid concentration sample loading flow rate sample volume sampling frequency linear range regression equation LOD (3σ, n = 11) RSD (50 μg L−1, n = 9)

232.0 nm 3.2 kV 30 mm 400 mL min−1 30 s 15% (v/v) 16.8 mL min−1 1.0 mL 35 h−1 5−100 μg L−1 I = 26.2C(μg L−1) − 125 1.3 μg L−1 2.4%

The microwave digestion procedures of human hair and laver samples. The dependence of optical emission intensity of nickel on distance between the two electrodes, argon flow rate, irradiation time, and concentration of formic acid. This material is available free of charge via the Internet at http://pubs.acs.org.

■

*E-mail: [email protected]. Tel: +86-24-83688944. Fax: +86-24-83676698. *E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work is financially supported by Natural Science Foundation of China (21475018 and 21235001), National Science & Technology Pillar Program (2012BAF14B09), Program of New Century Excellent Talents in University (NCET-13-0114), Scientific Research Fund of Liaoning Education Department (L2013107), and Fundamental Research Funds for the Central Universities (N120505004, N110805001, N110705002). Special thanks are due to Titan Instruments Co. Ltd. for providing the concentric quartz UVinduced reactor.

■

a

found (μg g−1)

spiked (μg L−1)

recovery (%)

GBW09101b GBW10023 GBW08608

5.77 2.25 ± 0.18 60 ± 3a

5.8 ± 0.5 2.3 ± 0.4 59 ± 8a

− − 20

− − 95 ± 4

REFERENCES

(1) Butler, O. T.; Cairns, W. R. L.; Cook, J. M.; Davidson, C. M. J. Anal. At. Spectrom. 2014, 29, 17−50. (2) Lum, T.-S.; Tsoi, Y.-K.; Leung, K. S.-Y. J. Anal. At. Spectrom. 2014, 29, 234−241. (3) Luo, D. B.; Duan, Y. X. TrAC, Trends Anal. Chem. 2012, 39, 254− 266. (4) Zhu, Z. L.; Zhang, S. C.; Lv, Y.; Zhang, X. R. Anal. Chem. 2006, 78, 865−872. (5) Meyer, C.; Muller, S.; Gurevich, E. L.; Franzke, J. Analyst 2011, 136, 2427−2440. (6) Webb, M. R.; Andrade, F. J.; Hieftje, G. M. Anal. Chem. 2007, 79, 7899−7905. (7) Meyer, C.; Demecz, D.; Gurevich, E. L.; Marggraf, U.; Jestel, G.; Franzke, J. J. Anal. At. Spectrom. 2012, 27, 677−681. (8) Minayeva, O. B.; Hopwood, J. A. J. Anal. At. Spectrom. 2003, 18, 856−863. (9) Guchardi, R.; Hauser, P. C. Analyst 2004, 129, 347−351. (10) Zapata, I. J.; Pohl, P.; Bings, N. H.; Broekaert, J. A. C. Anal. Bioanal. Chem. 2007, 388, 1615−1623. (11) Zhu, Z. L.; Zhang, S. C.; Xue, J. H.; Zhang, X. R. Spectrochim. Acta, Part B 2006, 61, 916−921. (12) Yu, Y. L.; Du, Z.; Chen, M. L.; Wang, J. H. J. Anal. At. Spectrom. 2008, 23, 493−499. (13) Xing, Z.; Kuermaiti, B.; Wang, J.; Han, G. J.; Zhang, S. C.; Zhang, X. R. Spectrochim. Acta, Part B 2010, 65, 1056−1060. (14) Zhu, Z. L.; Wu, Q. J.; Liu, Z. F.; Liu, L.; Zheng, H. T.; Hu, S. H. Anal. Chem. 2013, 85, 4150−4156.

Table 2. Determination of Nickel in Certified Reference Materials by Employing the Present DBD−OES System (n = 3) certified (μg g−1)

AUTHOR INFORMATION

Corresponding Authors

calibration range of 5−100 μg L−1 is achieved, along with a detection limit of 1.3 μg L−1 and a RSD value of 2.4% at 50 μg L−1 nickel. In comparison with the conventional ICP−OES by nebulization sample introduction for nickel determination, the present DBD−OES system provides sensitivity comparable or better than that achieved by ETAAS (LOD 0.35 μg L−1)43 and ICP−OES with vapor generation (LOD 1.8 μg L−1).40 To validate the reliability and applicability of the developed DBD−OES system, it is applied for the quantification of nickel in a series of biological and environmental samples, including certified reference materials of GBW09101b (human hair), GBW10023 (laver), and GBW08608 (water). The analysis results are given in Table 2. It is obvious that reasonable

sample

ASSOCIATED CONTENT

S Supporting Information *

μg L−1.

agreements are obtained between the certified and the values found for human hair, laver, and water samples. Meanwhile, the spiking recovery obtained in the water sample (GBW08608) is favorable within the range of 95 ± 4%. The experimental results have demonstrated that the present system is robust and reliable for trace nickel detection in real sample matrixes.

■

CONCLUSIONS In the present study, atmospheric pressure DBD microplasma as a radiation source is first used for the excitation of nickel 1371

DOI: 10.1021/ac5042457 Anal. Chem. 2015, 87, 1366−1372

Article

Analytical Chemistry (15) Yang, M.; Xue, J.; Li, M.; Han, G. J.; Xing, Z.; Zhang, S. C.; Zhang, X. R. Talanta 2014, 126, 1−7. (16) Li, W.; Zheng, C. B.; Fan, G. Y.; Tang, L.; Xu, K. L.; Lv, Y.; Hou, X. D. Anal. Chem. 2011, 83, 5050−5055. (17) Han, B. J.; Jiang, X. M.; Hou, X. D.; Zheng, C. B. Anal. Chem. 2014, 86, 936−942. (18) Yu, Y. L.; Du, Z.; Chen, M. L.; Wang, J. H. Angew. Chem., Int. Ed. 2008, 47, 7909−7912. (19) Zhu, Z. L.; Chan, G. C. Y.; Ray, S. J.; Zhang, X. R.; Hieftje, G. M. Anal. Chem. 2008, 80, 8622−8627. (20) Hayen, H.; Michels, A.; Franzke, J. Anal. Chem. 2009, 81, 10239−10245. (21) Michels, A.; Tombrink, S.; Vautz, W.; Miclea, M.; Franzke, J. Spectrochim. Acta, Part B 2007, 62, 1208−1215. (22) Hu, J.; Li, W.; Zheng, C. B.; Hou, X. D. Appl. Spectrosc. Rev. 2011, 46, 368−387. (23) He, Q.; Zhu, Z. L.; Hu, S. H.; Zheng, H. T.; Jin, L. L. Anal. Chem. 2012, 84, 4179−4184. (24) Krahling, T.; Muller, S.; Meyer, C.; Stark, A. K.; Franzke, J. J. Anal. At. Spectrom. 2011, 26, 1974−1978. (25) Wu, Z. C.; Chen, M. L.; Li, P.; Zhu, Q. Q.; Wang, J. H. Analyst 2011, 136, 2552−2557. (26) Li, W.; Jiang, X. M.; Xu, K. L.; Hou, X. D.; Zheng, C. B. Microchem. J. 2011, 99, 114−117. (27) Jiang, X. M.; Chen, Y.; Zheng, C. B.; Hou, X. D. Anal. Chem. 2014, 86, 5220−5224. (28) He, H. Y.; Zhu, Z. L.; Zheng, H. T.; Xiao, Q.; Jin, L. L.; Hu, S. H. Microchem. J. 2012, 104, 7−11. (29) Yu, Y.-L.; Dou, S.; Chen, M.-L.; Wang, J.-H. Analyst 2013, 138, 1719−1725. (30) Yu, Y.-L.; Cai, Y.; Chen, M.-L.; Wang, J.-H. Anal. Chim. Acta 2014, 809, 30−36. (31) Zhu, Z. L.; He, H. Y.; He, D.; Zheng, H. T.; Zhang, C. X.; Hu, S. H. Talanta 2014, 122, 234−239. (32) Saikia, P.; Goswami, K. S.; Saikia, B. K. Phys. Plasmas 2014, 21, 033501. (33) Wu, S.; Lu, X.; Zou, D.; Pan, Y. J. Appl. Phys. 2013, 114, 043301. (34) Wu, P.; He, L.; Zheng, C. B.; Hou, X. D.; Sturgeon, R. E. J. Anal. At. Spectrom. 2010, 25, 1217−1246. (35) Han, B. J.; Jiang, X. M.; Hou, X. D.; Zheng, C. B. Anal. Chem. 2014, 86, 6214−6219. (36) Zeng, Y.; Zheng, C. B.; Hou, X. D.; Wang, S. L. Microchem. J. 2014, 117, 83−88. (37) Guo, X. M.; Sturgeon, R. E.; Mester, Z.; Gardner, G. J. Anal. Chem. 2004, 76, 2401−2405. (38) Liu, L. W.; Deng, H.; Wu, L.; Zheng, C. B.; Hou, X. D. Talanta 2010, 80, 1239−1244. (39) NIST Atomic Spectra Database, http://physics.nist.gov/ PhysRefData/ASD/lines_form.html. (40) Marrero, J.; Smichowski, P. Anal. Bioanal. Chem. 2002, 374, 196−202. (41) Kogelschatz, U. IEEE Trans. Plasma Sci. 2002, 30, 1400−1408. (42) Kloc, P.; Wagner, H. E.; Trunec, D.; Navratil, Z.; Fedoseev, G. J. Phys. D: Appl. Phys. 2010, 43, 345205. (43) Hsiang, M. C.; Sung, Y. H.; Huang, S. D. Talanta 2004, 62, 791−799.

1372

DOI: 10.1021/ac5042457 Anal. Chem. 2015, 87, 1366−1372