Electrothermal Vaporization for Universal Liquid Sample Introduction

May 7, 2014 - solvent and matrix were removed first and subsequently atomized/vaporized analyte with extra energy provided by the W-coil was swept ...
0 downloads 0 Views 1MB Size
Download PDF
Technical Note pubs.acs.org/ac

Electrothermal Vaporization for Universal Liquid Sample Introduction to Dielectric Barrier Discharge Microplasma for Portable Atomic Emission Spectrometry Xiaoming Jiang,† Yi Chen,‡ Chengbin Zheng,‡ and Xiandeng Hou*,†,‡ †

Analytical & Testing Center, Sichuan University, Chengdu, Sichuan 610064, China College of Chemistry and Key Laboratory of Green Chemistry and Technology of MOE, Sichuan University, Chengdu, Sichuan 610064, China

‡

S Supporting Information *

ABSTRACT: Direct introduction of liquid sample into a microplasma for analytical atomic spectrometry can be a problem for its lowered atomization/excitation capability or can even extinguish it. The low power dielectric barrier discharge (DBD) microplasma has been widely used in optical spectrometry, but the number of detectable elements by atomic emission spectrometry (AES) is very limited, partially for the same reason. Here we use electrothermally vaporized analyte-containing species for sample introduction into a DBD microplasma, together with simple heating of the DBD, to enhance its atomization/excitation capability for AES. A compact tungsten coil electrothermal vaporizer (W-coil ETV) was used in this work, onto which a tiny volume of liquid sample was pipetted. Through administrating the heating program for the W-coil, sample solvent and matrix were removed first and subsequently atomized/vaporized analyte with extra energy provided by the W-coil was swept directly into the DBD microplasma for further atomization/excitation. These significantly contribute the stability of the DBD microplasma and save its power for reatomization/excitation of analyte thus improving the detectability. Under optimized experimental conditions, limits of detection of 0.8 μg L−1 (0.008 ng) for cadmium and 24 μg L−1 (0.24 ng) for zinc were obtained, with relative standard deviation (RSD) of 3.2% for 5 μg L−1 Cd and 3.7% for 100 μg L−1 Zn. Its potential application was also demonstrated by successfully analyzing several Certified Reference Materials. Its characteristics including compactness, low power consumption, cost effectiveness, tiny sample requirement, and easy operation make it very promising for field analytical chemistry.

A

atomic spectrometry, DBD has now been widely introduced into analytical spectrometry as an ionizer for mass spectrometry,11,12 chemical vapor generator,13,14 atomizer for atomic absorption spectrometry (AAS)9,10 and atomic fluorescence spectrometry (AFS),15,16 excitation source for molecular emission spectrometry,17 and gas chromatography (GC) detector.18,19 For atomic emission spectrometry (AES), Yu et al.1 and Zhu et al.20 first used DBD as the excitation source for determination of mercury. Their later works used DBD-AES to determinate thimerosal21 and iodine22 via vapor generation. In our recent study, atomic carbon emission was detected in a DBD microplasma, and a new universal GC detector was designed for carbon-containing volatile compounds.23 Like other microplasmas, however, the limited atomization/excitation capability derived from its low power consumption confines further application of DBD in AES. When liquid sample is introduced into the DBD microplasma, most energy is first consumed for evaporation of solvent and

lthough traditional laboratory benchtop atomic spectrometers are relatively mature in both instrumentation and applications to real sample analysis, miniaturized and even portable atomic spectrometric instrumentation attracts increasing attention, for more and more field analytical chemistry (FAC) is needed in numerous areas.1−3 Further, the drive to an energy-saving modern society is definitely another stimulus to develop FAC. In recent years, increasing interests have been focused on microplasma in order to build microplasmabased portable device because of its small size and low power consumption, such as glow discharge, corona discharge, microhollow-cathode discharge, capacitively coupled plasma, and dielectric barrier discharge (DBD), etc.4,5 However, direct introduction of liquid sample into a microplasma for analytical atomic spectrometry can be a problem for its lowered atomization/excitation capability or can even extinguish it. DBD’s unique advantages include simple construction, low temperature and power consumption, long lifetime, and atmospheric operation.6,7 Since Miclea and Franzke et al.8 first used a DBD for atomization of halogenated hydrocarbons for diode laser atomic absorption spectrometry (DLAAS) and Zhang et al.9,10 pioneered DBD’s applications in traditional © 2014 American Chemical Society

Received: February 8, 2014 Accepted: May 7, 2014 Published: May 7, 2014 5220

dx.doi.org/10.1021/ac500637p | Anal. Chem. 2014, 86, 5220−5224

Analytical Chemistry

Technical Note

details of its real device). A coaxial DBD (quartz tube, 7 cm × 3 mm i.d. × 5 mm o.d.) was used. The outer electrode was copper filament which was spiraled around the DBD tube firmly, and a stainless steel wire (10 cm × 1 mm diameter) was used as the inner one. The two electrodes were connected to a high voltage power supply (CTP-2000K, Nanjing SuMan Electronic Co., LTD, China). The discharge area of the DBD was confined in another larger quartz tube which was wrapped with a resistant filament (nichrome wire) for heating the DBD. The W-coil ETV device was the filament carefully extracted from a commercial slide projector bulb (HLX64633, 10 A, 150 W, OSRAM, Munich, Germany). It was housed in a laboratory-made quartz cell, which connected the DBD with a short silica tube. The base of the cell, on which the W-coil was fixed, connected a laboratory-made power supply for the W-coil, and a hole was made through the base for introduction of working gas to sweep the atomized/vaporized analyte into the DBD. On the top of the cell, another hole was made for sample introduction. The optical emission in the discharge region transmitted through a quartz window to a lens fixture for converging into the optical fiber, and finally entered into the CCD spectrometer (Maya 2000 PRO, 25 μm slit, 175−401 nm, Ocean Optics (Shanghai) Co., China). Analytical Procedure. A volume of 10 μL of analyte solution was pipetted onto the W-coil by a microinjector and followed by the heating program of the W-coil ETV. At the right time during the program, igniting the plasma and synchronizing data acquisition were implemented to complete the measurement (see details in the Supporting Information). Reagents and Sample. All reagents used in this work were at least of analytical grade. Stock solution (1000 mg L−1) of Cd(II), Zn(II), and Hg(II) were obtained from National Research Center (NRC) of China (Beijing). Several Certified Reference Materials (CRMs) from NRC of China, including GBW08510 (rice powder), GBW07601 (human hair), and GBW08607 (water) were used to validate the accuracy of the proposed method. Detailed sample preparation can be found in the Supporting Information.

matrix, and the plasma state is also altered thus deteriorating its stability. A large volume of liquid sample can even extinguish the plasma. Ideally, the analyte is transferred into analyte-containing gaseous species prior to introduction to the DBD microplasma. The present direct DBD-AES almost employed mainly cold vapor generation (CVG) for the sample introduction, which limited the number of applicable elements and mainly focused on mercury. Puanngama et al.24 designed a preconcentrator for DBD, with improved stability and thus excellent LODs for mercury because solvent interference was avoided. Tombrink et al.25 first introduced direct liquid analysis by using a capillary DBD, and Zhu et al.26 developed a liquid-film DBD (LFDBD). In their works, microplasma generated between the liquid sample and another meal electrode for excitation of several elements, which solved the liquid sampling problem to some extent. Nevertheless, residual or concomitant moisture and matrix from liquid sample introduction still restrict the analytical performance of DBD microplasma for AES. Electrothermal vaporization (ETV) is recognized as an excellent sample introduction approach for atomic spectrometry, which offers unique advantages such as high sample introduction efficiency, improved detection limits, and most importantly the separation of the analyte from the solvent and matrix by multistep heating the ETV.27,28 Among miniaturized metallic ETV devices, tungsten coil (W-coil) is an especially promising one for its inherent advantages29 and has been widely used in atomic spectrometry.2,30,31 In the present work, therefore, a W-coil ETV was utilized as a sampler to introduce gasified analyte to a DBD microplasma to thoroughly explore its application in AES analysis of liquid samples. The analyte-containing gasified species from the W-coil with “some extra energy” (e.g., free atoms, radicals, and even excited species) but without moisture and matrix immediately entered the DBD microplasma for further atomization/excitation; and simple heating the DBD significantly further saved the energy of the DBD and thus enhanced the final atomic excitation efficiency. Analytical performance including sensitivity and stability could be significantly improved for the high sampling efficiency of W-coil ETV and elimination of interferences from sample solvent and matrix. Trace toxic element cadmium was chosen to optimize the experimental conditions, and trace zinc and mercury were preliminarily investigated as well.

■

RESULTS AND DISCUSSION Since both DBD and W-coil feature small size and simple structure, the interface between them could be as short as possible to improve the sample introduction efficiency of the W-coil; and in fact, we just needed to ensure that the W-coil would not be too close to the DBD and disturbed by the high voltage. Using this W-coil ETV and the DBD excitation source, the atomic emission signal of cadmium, zinc, and mercury

■

EXPERIMENTAL SECTION Instrumentation. Schematic diagram of the experimental setup is shown in Figure 1 (see the Supporting Information for

Figure 1. Schematic diagram of the instrumental arrangement. (a) The system at working state, an inset shows the axial view of the discharge region of the DBD. 5221

dx.doi.org/10.1021/ac500637p | Anal. Chem. 2014, 86, 5220−5224

Analytical Chemistry

Technical Note

(the 23rd data point/acquisition). Three continuous timeresolved emission spectra during the atomization/vaporization containing intact emission signal of analyte are shown in Figure 2b (see details of all spectra of one measurement in the Supporting Information). Analyte atoms were excited between 5.50 and 5.75 s, and the entire emission signal was presented at the acquisition of 5.75 s with only one spectrum containing the emission lines of the analyte obtained. The consistency and minimum background fluctuation of the three continuous spectra except the only difference of emission lines of analyte at 5.75 s also benefited from the removal of the moisture and matrix. So a clean spectrum of analyte in Figure 2c could be simply obtained from the spectrum at 5.75 s minus that at 5.50 s in Figure 2b. Besides, to ensure the whole emission signal could be recorded in just one integration period of the CCD, the integration time was also investigated and 250 ms was used (see details in the Supporting Information). Residual Solvent and Matrix. The heating program of ETV starts with a drying step for removing solvent and then a pyrolysis step for further removing moisture and eliminating sample matrix, which is also a key issue in this work. To demonstrate the effectiveness of coupling W-coil ETV with DBD microplasma and further studying the effect of residual moisture and matrix on the plasma, the pyrolysis current (temperature) of the W-coil was investigated carefully. A low current could not effectively remove most of the moisture and matrix, accompanying great fluctuation of background spectra, while a too high current would cause serious analyte loss, and 4.0 A was finally selected for use (see details in the Supporting Information). Vaporization for Sample Introduction. The analyte is atomized/vaporized off the W-coil during the atomization/ vaporization step, so the atomization/vaporization current is an essential factor for the sample introduction efficiency, analyte atomization/vaporization efficiency, and the extra energy provided to the DBD. A high current would lead to quick release of analyte from the W-coil and fast thermal expansions, and thus the analyte would be transported into the DBD effectively, keeping high population of free analyte atoms to save the energy of the DBD for reatomization/excitation, so high current was in favor in this work. Besides, to further accelerate the release of analyte, a preheating step was first added before the atomization/vaporization for heating the W-coil prior to a desirable temperature. The details can be found in the Supporting Information. Heat-Assisted Atomic Emission Enhancement. To further improve the atomization/excitation capability of the DBD and decrease the influence of moisture, simple heating the DBD was also tested. As shown in Figure 3, it could be concluded that the atomization/excitation capability of the DBD was improved obviously with the temperature. It also benefited from the further evaporation of solvent, so 30 V was adopted, without further increasing the voltage for rated power and safety. Working Gas. The discharging region of DBD needs to be full of working gas to maintain stable discharge, such as helium, argon, nitrogen, and air.7 The W-coil also needs working in an inert gas atmosphere to prevent oxidation. Meanwhile, the working gas also plays a very important role of transporting the atomized/vaporized analyte-containing species from the W-coil to the DBD, namely, as carrier gas. To meet these requirements, we tried helium and argon for their inertness. Previous reports29,31 revealed that a reducing atmosphere was provided

could be easily observed, of which the spectral characteristics and experimental parameters were carefully studied to evaluate the analytical performance of the new instrumentation. Spectral Characteristics. The atomic emission spectrum from this device was investigated in details, as shown in Figure 2.

Figure 2. (a) Synchronization for emission signal acquisition; (b) three continuous time-resolved spectra of Cd, Zn, and Hg at 5.50, 5.75, and 6.00 s; and (c) final processed spectrum of Cd, Zn, and Hg; the wavelengths of the numbered emission lines can be found in the Supporting Information. Experimental details: Cd, 1 mg L−1; Zn, 50 mg L−1; Hg, 10 mg L−1; and CCD integration time, 250 ms.

Figure 2c is the final processed spectrum of Cd, Zn, and Hg mixed solution, from which we can obviously identify the spectral lines by comparing with the reference database,32 including Cd 326.105, Zn 213.856, and Hg 253.652 nm. The release of analyte from the W-coil and retention time for excitation in the DBD is quite short, thus the generated atomic emission signal is transient. Therefore, a synchronization was made between the W-coil heating program and the CCD data acquisition to obtain the spectral data accurately. As shown in Figure 2a, the data acquisition of the CCD begins with the preheating step (1 s delay) through sending a synchronous trigger signal from the power supply of the W-coil to the control software of the CCD at the step beginning. According to calculation and the experimental results, we could confirm that emission signal of analyte could be obtained at 5.75 s 5222

dx.doi.org/10.1021/ac500637p | Anal. Chem. 2014, 86, 5220−5224

Analytical Chemistry

Technical Note

volume or adding a preconcentration step. Meanwhile, because of elimination of solvent and matrix, more stable steady-state signals could be obtained, yielding good RSDs of 3.2% for 5 μg L−1 Cd and 3.7% for 100 μg L−1 Zn, respectively. Real Sample Analysis. The utility and accuracy of the proposed technique were demonstrated by the determination of Cd in several CRMs including rice powder, human hair, and water samples with analytical results summarized in Table 1, Table 1. Analytical Results of Cd in CRMs

Figure 3. Effect of heating DBD. Cd, 100 μg L−1.

a

with addition of hydrogen, so this was also tested in this work, whereas pure He was finally selected for higher sensitivity. Moreover, the transportation efficiency of analyte and its concentration in working gas as well as the residence time in the DBD are significantly influenced by the working gas flow rate. Since the atomized/vaporized products from the W-coil needed to be transported into the DBD immediately, a high flow rate was in favor in this work. The comparison and influence of working gas in detail can be found in the Supporting Information, together with the investigation of discharge voltage, frequency, and power consumption. Analytical Figures of Merit. Analytical figures of merit were evaluated under the optimal experimental conditions. The linear correlation coefficients for the calibration curves were better than 0.99 in trace concentration ranges from 5 to 1000 μg L−1 for Cd and from 100 to 50 000 μg L−1 for Zn, respectively; and the higher concentration ends of the curves were limited by optical saturation of the CCD detector. Figure 4 shows the emission signal profiles of Cd and Zn

sample

certified (μg g−1)

found (μg g−1)a

GBW08510 GBW07601 GBW08607

2.062 ± 0.052 0.11 ± 0.03 0.102 ± 0.002

2.023 ± 0.022 0.107 ± 0.010 0.103 ± 0.006

Mean and standard deviation (n = 3). Standard curve method.

and the t test showed the analytical results by the proposed method had no significant difference with the certified values at the confidence level of 95%.

■

CONCLUSION Using electrothermally vaporized analyte-containing species for sample introduction into a low power DBD microplasma enhanced its atomization/excitation capability for atomic emission. In addition, simple heating of the DBD further improved the atomization/excitation capability most probably through saving energy from the DBD. Compared to conventional DBD-AES, this technique demonstrated improved sensitivity and stability owing to the advantageous features of the W-coil ETV device, which include elimination of interferences from sample solvent and matrix prior to analyte introduction to the DBD and providing extra energy to the DBD. The proposed instrumentation has additional advantages of small size, simple operation, high sensitivity (picogram level for Cd), fast analysis (2 min), tiny sampling volume (10 μL), and low power consumption, and it would have broad applications (Cd, Zn, Hg, and more) in field analytical chemistry.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

Figure 4. Emission profiles and calibration curves in linear concentration ranges of Cd and Zn. Cd, 326.105 nm; and Zn, 213.856 nm.

ACKNOWLEDGMENTS We thank the National Natural Science Foundation of China (Grants 21275103 and 21305094) and Youth Foundation of Sichuan University (Grant 2012SCU11056) for financial support.

(mainly 326.105 and 213.856 nm) and the calibration curves in the linear concentration ranges. The limits of detection (LODs) of 0.8 μg L−1 (0.008 ng) for Cd and 24 μg L−1 (0.24 ng) for Zn were obtained, which is comparable to that of AAS, while better than those of conventional AES,26,33 and the mass LODs even much better than HG-AFS for the small sampling volume, as summarized in the Supporting Information. It should be noted that a sample volume of just 10 μL was used, and the sensitivity and LOD can be further improved by increasing the sampling

■

REFERENCES

(1) Yu, Y. L.; Du, Z.; Chen, M. L.; Wang, J. H. Angew. Chem., Int. Ed. 2008, 47, 7909−7912. (2) Gu, J. Y.; Oliveira, S. R.; Donati, G. L.; Gomes Neto, J. A.; Jones, B. T. Anal. Chem. 2011, 83, 2526−2531. (3) Hou, X. D.; Jones, B. T. Microchem. J. 2000, 66, 115−145.

5223

dx.doi.org/10.1021/ac500637p | Anal. Chem. 2014, 86, 5220−5224

Analytical Chemistry

Technical Note

(4) Miclea, M.; Franzke, J. Plasma Chem. Plasma Process. 2007, 27, 205−224. (5) Yuan, X.; Tang, J.; Duan, Y. X. Appl. Spectrosc. Rev. 2011, 46, 581−605. (6) Kogelschatz, U. Plasma Chem. Plasma Process. 2003, 23, 1−46. (7) Hu, J.; Li, W.; Zheng, C. B.; Hou, X. D. Appl. Spectrosc. Rev. 2011, 46, 368−387. (8) Miclea, M.; Kunze, K.; Musa, G.; Franzke, J.; Niemax, K. Spectrochim. Acta, Part B 2001, 56, 37−43. (9) Zhu, Z. L.; Zhang, S. C.; Lv, Y.; Zhang, X. R. Anal. Chem. 2006, 78, 865−872. (10) Zhu, Z. L.; Zhang, S. C.; Xue, J. H.; Zhang, X. R. Spectrochim. Acta, Part B 2006, 61, 916−921. (11) Gilbert-Lopez, B.; Schilling, M.; Ahlmann, N.; Michels, A.; Hayen, H.; Molina-Diaz, A.; Garcia-Reyes, J. F.; Franzke, J. Anal. Chem. 2013, 85, 3174−3182. (12) Na, N.; Zhang, C.; Zhao, M. X.; Zhang, S. C.; Yang, C. D.; Fang, X.; Zhang, X. R. J. Mass Spectrom. 2007, 42, 1079−1085. (13) Wu, X.; Yang, W. L.; Liu, M. G.; Hou, X. D.; Zheng, C. B. J. Anal. At. Spectrom. 2011, 26, 1204−1209. (14) Zhu, Z. L.; Wu, Q. J.; Liu, Z. F.; Liu, L.; Zheng, H. T.; Hu, S. H. Anal. Chem. 2013, 85, 4150−4156. (15) Zhu, Z. L.; Liu, J. X.; Zhang, S. C.; Na, X.; Zhang, X. R. Spectrochim. Acta, Part B 2008, 63, 431−436. (16) Yu, Y. L.; Du, Z.; Chen, M. L.; Wang, J. H. J. Anal. At. Spectrom. 2008, 23, 493−499. (17) Li, W.; Jiang, X. M.; Xu, K. L.; Hou, X. D.; Zheng, C. B. Microchem. J. 2011, 99, 114−117. (18) Kunze, K.; Miclea, M.; Franzke, J.; Niemax, K. Spectrochim. Acta, Part B 2003, 58, 1435−1443. (19) Li, W.; Zheng, C. B.; Fan, G. Y.; Tang, L.; Xu, K. L.; Lv, Y.; Hou, X. D. Anal. Chem. 2011, 83, 5050−5055. (20) Zhu, Z. L.; Chan, G. C. Y.; Ray, S. J.; Zhang, X. R.; Hieftje, G. M. Anal. Chem. 2008, 80, 8622−8627. (21) He, H. Y.; Zhu, Z. L.; Zheng, H. T.; Xiao, Q.; Jin, L. L.; Hu, S. H. Microchem. J. 2012, 104, 7−11. (22) Yu, Y. L.; Dou, S.; Chen, M. L.; Wang, J. H. Analyst 2013, 138, 1719−1725. (23) Han, B. J.; Jiang, X. M.; Hou, X. D.; Zheng, C. B. Anal. Chem. 2014, 86, 936−942. (24) Puanngam, M.; Ohira, S. I.; Unob, F.; Wang, J. H.; Dasgupta, P. K. Talanta 2010, 81, 1109−1115. (25) Tombrink, S.; Mueller, S.; Heming, R.; Michels, A.; Lampen, P.; Franzke, J. Anal. Bioanal. Chem. 2010, 397, 2917−2922. (26) He, Q.; Zhu, Z. L.; Hu, S. H.; Zheng, H. T.; Jin, L. L. Anal. Chem. 2012, 84, 4179−4184. (27) Hu, B.; Li, S. Q.; Xiang, G. Q.; He, M.; Jiang, Z. C. Appl. Spectrosc. Rev. 2007, 42, 203−234. (28) Vanhaecke, F.; Resano, M.; Verstraete, M.; Moens, L.; Dams, R. Anal. Chem. 2000, 72, 4310−4316. (29) Hou, X. D.; Jones, B. T. Spectrochim. Acta, Part B 2002, 57, 659−688. (30) Parsons, P. J.; Zhou, Y.; Palmer, C. D.; Aldous, K. M.; Brockman, P. J. Anal. At. Spectrom. 2003, 18, 4−10. (31) Wu, P.; Wen, X. D.; He, L.; He, Y. H.; Chen, M. Z.; Hou, X. D. Talanta 2008, 74, 505−511. (32) NIST Atomic Spectra Database, http://physics.nist.gov/ PhysRefData/ASD/lines_form.html. (33) Weagant, S.; Karanassios, V. Anal. Bioanal. Chem. 2009, 395, 577−589.

5224

dx.doi.org/10.1021/ac500637p | Anal. Chem. 2014, 86, 5220−5224