Plasma Jet Desorption Atomization-Atomic Fluorescence

Nov 16, 2012 - A novel plasma jet desorption atomization (PJDA) source was developed for atomic fluorescence spectrometry (AFS) and coupled on line wi...
0 downloads 0 Views 822KB Size
Download PDF
Letter pubs.acs.org/ac

Plasma Jet Desorption Atomization-Atomic Fluorescence Spectrometry and Its Application to Mercury Speciation by Coupling with Thin Layer Chromatography Zhifu Liu,†,‡ Zhenli Zhu,*,† Hongtao Zheng,§ and Shenghong Hu†,‡ †

State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Wuhan, China, 430074 Faculty of Earth Sciences, China University of Geosciences, Wuhan, China, 430074 § Faculty of Material Science and Chemistry, China University of Geosciences, Wuhan, China, 430074 ‡

S Supporting Information *

ABSTRACT: A novel plasma jet desorption atomization (PJDA) source was developed for atomic fluorescence spectrometry (AFS) and coupled on line with thin layer chromatography (TLC) for mercury speciation. An argon dielectric barrier discharge plasma jet, which is generated inside a 300 μm quartz capillary, interacts directly with the sample being analyzed and is found to desorb and atomize surface mercury species rapidly. The effectiveness of this PJDA surface sampling technique was demonstrated by measuring AFS signals of inorganic Hg2+, methylmercury (MeHg), and phenylmercury (PhHg) deposited directly on TLC plate. The detection limits of the proposed PJDA-AFS method for inorganic Hg2+, MeHg, and PhHg were 0.51, 0.29, and 0.34 pg, respectively, and repeatability was 4.7%, 2.2%, and 4.3% for 10 pg Hg2+, MeHg, and PhHg. The proposed PJDA-AFS was also successfully coupled to TLC for mercury speciation. Under optimized conditions, the measurements of mercury dithizonate (HgD), methylmercury dithizonate (MeHg-D), and phenylmercury dithizonate (PhHg-D) could be achieved within 3 min with detection limits as low as 8.7 pg. The combination of TLC with PJDA-AFS provides a simple, cost-effective, relatively highthroughput way for mercury speciation.

M

effect”, because a new plate is typically used for each TLC run. Although mainly used for the separation of organic compounds, procedures for separation of mercury species on silica gel plates have been developed.13,14 Densitometric detection13 and enzymatic determination14 have been used to detect the separated mercury compounds on a TLC plate; the sensitivity and detection limits of these detection methods, however, are still unsatisfied. Sensitive and element-specific detection techniques for TLC speciation would improve the performance of this hyphenated technique. TLC has been off-line coupled with atomic absorption spectrometry (AAS) for determination of methylmercury.15 After TLC separation, the adsorbent was scraped into an ignition tube and then heated to generate mercury vapor for detection by AAS. However, this off-line “scrape” method is tedious and time-consuming, and the recovery is poor. Direct analysis of the analytes separated on a TLC plate provides a simple means to achieve high speed and compoundspecific detection with minimal or no sample pretreatment.

ercury is one of the most toxic elements impacting human health and ecosystem, and it exists in a variety of chemical forms. Because the toxicity, metabolism, and bioavailability of mercury are greatly dependent on its chemical form,1,2 there is an increasing concern in speciation analysis of mercury. Cold vapor atomic fluorescence spectrometry (CVAFS) is one of the most commonly used methods for mercury determination due to its high sensitivity, simplicity, low running costs, short warm-up time, and ease of operation. However, it provides information only on the elemental composition of the sample; it needs to be coupled to chromatographic methods to give speciation information. Gas chromatography (GC)3−5 and high performance liquid chromatography (HPLC)6−10 are the most commonly applied separation method for mercury speciation. However, for samples with high biological and environmental matrix, purification and dilution are usually needed before GC and HPLC analysis. Thin layer chromatography (TLC) is a rapid, simple, robust, and economical separation technique for a wide range of organic and inorganic mixtures.11,12 In comparison with chromatographic columns (e.g., GC and HPLC), TLC can be used for direct analysis of crude samples with minimal purification procedures. In addition, TLC has no “memory © 2012 American Chemical Society

Received: October 10, 2012 Accepted: November 16, 2012 Published: November 16, 2012 10170

dx.doi.org/10.1021/ac3028504 | Anal. Chem. 2012, 84, 10170−10174

Analytical Chemistry

Letter

Figure 1. Schematic diagram of the TLC-PJDA-AFS setup.

atic optimization of the parameters of TLC-PJDA-AFS was performed, and analytical figures of merit were determined.

Two innovative ideas, extraction device and laser ablation, have been reported to directly couple TLC with atomic spectrometer. Meermann et al.16 developed a TLC-inductively coupled plasma mass spectrometry (ICP-MS) system for iodine speciation by means of an extraction device as an interface, which allows a nearly complete recovery of compounds from TLC within short extraction times. However, an extraction device can suffer from clogging of the capillary, which is used to transfer the analyte-containing liquid to the device. Owing to the innately high spatial resolution and scanning capability, laser ablation is a useful technique for rapid and continuous characterization of analytes directly from the surfaces of TLC plates. Speciation of arsenic17 and chromium18 have been achieved by the combination of TLC and laser ablation (LA)ICPMS. LA has also been coupled with AFS for the determination of mercury in solid samples.19,20 The combination of TLC and LA-AFS for mercury speciation appears as a logical choice; a possibility that, however, to the best of our knowledge, is yet unexplored. In addition, the LA systems are complex and expensive. Plasma based surface sampling/ionization techniques,21−28 which offer the advantages of inexpensive instrumentation, rugged construction, and low power consumption, have received increasing attention. Several plasma based ionization techniques, including direct analysis in real time (DART),21 flowing afterglow-atmospheric pressure glow discharge,22 low temperature plasma (LTP),23 plasma assisted desorption ionization,28 and microplasma discharge ionization,25 have been developed for direct surface sample analysis. DART has also been developed as interface to couple TLC with MS.29 Nevertheless, these techniques focus on the analysis of organic compounds. Recently, Zhang et al.30 achieved depth profiling of thin layer films by combining LTP with ICPMS. The plasma was generated in a quartz capillary, with a plasma jet (PJ) used to ablate the samples for elemental analysis. Therefore, PJ provides a new alternative to LA to couple TLC with atomic spectrometer. In the present work, we demonstrated that mercury species could be readily desorbed and atomized simultaneously by plasma jet desorption atomization (PJDA) source, which was generated inside a quartz capillary with inner diameter of 300 μm at ambient condition. The PJDA source provides a novel interface to couple TLC on line with AFS for the speciation analysis of three chemical forms of mercury: inorganic Hg2+, methylmercury (MeHg), and phenylmercury (PhHg). System-

■

EXPERIMENTAL SECTION

Instrumentation. Figure 1 shows a schematic diagram of the TLC-PJDA-AFS experimental setup. After development and air drying, a cut TLC plate (50 mm × 25 mm, L × W) was stabilized on a sled by double-sided adhesive tape and then positioned into a sampling chamber. The sampling chamber is a polymethyl methacrylate (PMMA) cuboid chamber. In this study, two sampling chambers were used. A larger volume chamber A (100 mm × 30 mm × 30 mm, L × W × D) with a sled (50 mm × 25 mm × 25 mm, L × W × H) was used in the initial experiments to investigate the feasibility of PJ as mercury desorption and atomization source and optimize the experiment conditions. A compact chamber B (100 mm × 30 mm × 10 mm, L × W × D) with a sled (50 mm × 25 mm × 7 mm, L × W × H) was used for the speciation of mercury. The sled was translated by a motor-controlled handle, and the sample was scanned at a rate of 0.3−0.6 mm/s. The PJDA source, which is built similar to the configuration of a dielectric barrier discharge (DBD) ionization source,27,31 consists of a 30 mm long fused silica capillary (i.d., 300 μm; o.d., 1.0 mm) and two copper rings separated about 12 mm and wrapped outside the capillary. An AC high voltage, about 2.5 kV, at a frequency of 10 kHz was applied to the electrodes. The AC voltage was provided by a CTP-2000 K discharge power supply (Suman Electronics Co., Ltd., NanJing, China). The capillary was immobilized on the sampling chamber at an angle of 45° to the surface plane. Highpurity Ar was used as both discharge and carrier gas at flow rates of 100 and 800 mL min−1, respectively. The generated mercury atom by PJ was introduced by Ar to quartz furnace where it was excited by a mercury hollow cathode lamp (HCL), and generated Hg fluorescence was detected by a solar-blind photomultiplier tube (PMT). A commercial double-channel nondispersion atomic fluorescence spectrometer (AFS 9130, Beijing Titan Instruments Co. Ltd., Beijing, China) was used in the present study. The operating parameters of AFS were as follows: wavelength, 253.7 nm; PMT voltage, 280 V; HCL current, 30 mA; shield gas (Ar), 900 mL min−1. TLC Separation. TLC separation was carried out on a glassbacked silica gel high performance TLC (HPTLC) plate (2.5 cm × 7.5 cm, thickness 0.20−0.25 mm, Qingdao Shenghai Fine Silica Gel Chemical Co., Ltd.). The mercury species were reacted with dithizone (0.02% m/v in chloroform) to allow 10171

dx.doi.org/10.1021/ac3028504 | Anal. Chem. 2012, 84, 10170−10174

Analytical Chemistry

Letter

Figure 2. Typical temporal profile of background-subtracted fluorescence signal of 0.05 ng of (a) Hg2+, (b) MeHg, and (c) PhHg and 0.1 ng of (d) Hg-D, (e) MeHg-D, and (f) PhHg-D by PJDA-AFS.

an interface for the direct coupling of TLC with AFS. As a proof-of-principle experiment to illustrate the desorption capabilities of PJ, Hg2+, MeHg, and PhHg were employed as model analytes. Each compound (1 μL) was spotted on TLC plate, followed by air drying, and then desorbed by PJ and continually detected by AFS. The temporal profile of fluorescence signals from 0.05 ng of Hg2+, MeHg, and PhHg with quartz tube atomizer at room temperature were showed in Figure 2a,b,c. In comparison to the capillary DBD ionization source27,31 and the LTP,23,26 the results demonstrated that PJ has two functions: (1) as a desorption source for the mercury species adsorbed on the TLC plates and (2) as an atomization source for the desorbed mercury species. In addition, the appearance of Hg fluorescence signal was almost instantaneous and synchronized with the power applied to the PJ, which indicated that desorption and atomization of analytes was achieved simultaneously. The sampling efficiency of the PJDA system was evaluated roughly by comparison with conventional cold vapor (CV) system (NaBH4−HCl reduction) with signalto-sample volume (S/V) ratio as the parameter using Hg2+ as a model analyte. With peak area as the analytical parameter, the S/V ratio was found to be higher for the PJ system by a factor of approximately 2.7 in comparison with conventional CV. It can be concluded that the PJ system is a high-efficiency desorption atomization technique for mercury. It provides a new efficient way for direct determination of mercury on sample surface with AFS. In addition, the analysis time for one sample is only about 20 s; therefore, further development into a high-throughput method for rapid analysis of mercury is, in principle, highly possible.

visualization of the spots. The dithizone excess was then eliminated by 0.2 mol L−1 NaOH. The sample solution, which contained the Hg−dithizone complexes, was spotted (1 μL) on the HPTLC plate and developed in n-hexane/triethylamine (5:1, v/v) solution, which was the optimum condition (see the Supporting Information for Figure S1), for 20 min. After development, the three sample spots were completely resolved, with Rf values (defined as the migration distance of analyte to that of solvent front) of 0.08 for mercury dithizonate (Hg-D), 0.63 for methylmercury dithizonate (MeHg-D), and 0.47 for phenylmercury dithizonate (PhHg-D). Reagents. All chemicals used in this work were of at least analytical-reagent grade. The stock standard solution of inorganic mercury (GBW08617) and methylmercury (GBW08675) were supplied by National Research Center for Certified Reference Materials (Beijing, China). Phenylmercurychloride was purchased from Laboratories of Dr. Ehrenstorfer (Augsburg, Germany). The mass of all the three mercury species were expressed as Hg in the present study. Tetrachloromethane and dithizone were obtained from Tianjin Baishi Chemical Industry Co., Ltd. (Tianjin, China) and Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China), respectively. HPLC grade n-hexane and triethylamine were purchased from CNW Technologies GmbH (Dusseldorf, Germany).

■

RESULTS AND DISCUSSION Capillary dielectric barrier discharge allows the plasma species to be extracted by the combined action of the gas flow and the electric field, with a plasma jet extending beyond the quartz capillary and suitable for direct surface sampling. In this study, PJ was used as sampling tool for mercury species and served as 10172

dx.doi.org/10.1021/ac3028504 | Anal. Chem. 2012, 84, 10170−10174

Analytical Chemistry

Letter

Clearly, the PJDA-AFS is capable of direct sampling and atomization for those mercury species adsorbed on TLC plates. Several TLC methods for separation of mercury species have been reported.13,15 Because the mercury species are colorless, they were usually transformed to colored mercury−dithizone complexes by reaction with dithizone before TLC analysis. Therefore, desorption and atomization of mercury−dithizone complexes (Hg-D, MeHg-D, and PhHg-D) on TLC surface with the proposed PJDA source was also investigated. As shown in Figure 2d,e,f, similar results were also observed. However, the signals of mercury−dithizone complexes were slightly lower than the signal of the corresponding unmodified mercury species. This difference may be ascribed to the different sampling/atomization efficiency. The results demonstrated that PJDA-AFS can be used not only for mercury analysis but also for mercury−dithizone complex analysis. On the basis of these results, mercury speciation can be achieved if we combine TLC with PJDA-AFS. After optimization (see the Supporting Information for Figures S2, S3, and S4), the performance of the TLC-PJDAAFS setup for mercury speciation was investigated. After TLC development, the plate was positioned on the sled of the sampling chamber A and was linearly scanned by the PJDA. The elemental Hg vapor generated by PJDA was then detected by AFS. Figure 3b shows the chromatograms of the Hg-D, PhHg-D, and MeHg-D with the sampling chamber A. The results showed that baseline separation of Hg-D and PhHg-D can be achieved. However, MeHg-D and PhHg-D are not well separated which is maybe caused by the relatively large internal volume of the sampling chamber A (about 58 mL). Therefore, a more compact cell B (internal volume of about 21 mL) was fabricated. It was found that the resolution was significantly improved. In addition, the resolution can be further improved by reducing the scan rate from 0.4 to 0.2 mm/s. However, peak distortion was observed at a scan rate of 0.2 mm/s. Therefore, 0.3 mm/s was selected as the scan speed. The improved chromatogram of Hg-D, PhHg-D, and MeHg-D using compact cell B was shown in Figure 3c. The results showed that PhHg-D and MeHg-D was almost baseline separated using the compact sampling cell, and scanning of the separated Hg-D, PhHg-D, and MeHg-D bands required only 3 min. The mercury dithizonates separated by TLC was also scanned by LAICPMS with laser spot size of 160 μm for comparison. Baseline separation of PhHg-D and MeHg-D was also not completely achieved (Supporting Information, Figure S5), which indicated that memory effect of mercury also degraded the resolution. The analytical characteristics of PJDA-AFS and TLC-PJDAAFS were evaluated under the optimal operating conditions (Table 1). The peak height was used as analytical signal throughout this work. Linear correlation coefficients for calibration curves of analytes were better than 0.99 for both PJDA-AFS and TLC-PJDA-AFS (Supporting Information, Figure S6). Repeatability, expressed as the relative standard deviation of the peak height, was better than 8% (n = 5). The limits of detection (LODs), using the definition 3σ/m (σ is the standard deviation corresponding to 10 blank measurements and m is the slope of the calibration graph), were lower than 0.51 pg for PJDA-AFS and 8.7 pg for TLC-PJDA-AFS, respectively. In addition, the detection limits may be improved by reapplying the sample on the TLC several times, which is a commonly used method in TLC. The reliability of the method was checked by recovery experiments. Analytical data of the recovery of Hg2+, MeHg,

Figure 3. (a) Picture of developed lane on TLC plate showing complete development of three mercury species spotted (10 ng each compound); (b) chromatogram of Hg-D (0.5 ng), PhHg-D (0.5 ng), and MeHg-D (1 ng) separated by TLC with the sampling chamber A at scan rate of 0.4 mm/s; (c) chromatogram of Hg-D (0.5 ng), PhHgD (0.5 ng), and MeHg-D (1 ng) separated by TLC with the compact sampling chamber B at scan rate of 0.3 mm/s.

Table 1. The Analytical Characteristics of PJDA-AFS and TLC-PJDA-AFS for the Determination of Mercury Species method PJDA-AFS

TLC-PJDAAFS

mercury species

R2

linear range (ng)

LOD (pg)

RSD (%, n = 5)

Hg MeHg PhHg Hg-D MeHg-D PhHg-D

0.9998 0.9998 0.9987 0.9919 0.9986 0.9979

0.005−0.2 0.005−0.2 0.005−0.2 0.1−2 0.1−2 0.1−2

0.51 0.29 0.34 3.1 8.7 6.0

4.7 2.2 4.3 5.8 4.7 7.9

and PhHg spiked in lake water and urine samples were summarized in Table 2. The results showed that good recoveries (96−104%) were obtained for all samples, which demonstrated the applicability of the proposed method to the speciation analysis of mercury in water and urine samples.

■

CONCLUSIONS A new plasma jet desorption atomization-atomic fluorescence spectrometry was developed and successfully coupled with TLC for rapid speciation analysis of mercury. The plasma jet 10173

dx.doi.org/10.1021/ac3028504 | Anal. Chem. 2012, 84, 10170−10174

Analytical Chemistry

Letter

Table 2. Determination of Mercury in Urine and Water Samples by TLC-PJDA-AFS founda (ng)

added (ng) sample

Hg

2+

MeHg

PhHg

0.5 1.0

0.5 1.0

0.5 1.0

0.5 1.0

0.5 1.0

0.5 1.0

urine

lake water

a

Hg ND 0.48 0.99 ND 0.49 1.01

2+

MeHg ND 0.49 1.02 ND 0.49 0.99

± 0.02 ± 0.04 ± 0.03 ± 0.05

± 0.03 ± 0.05 ± 0.02 ± 0.06

recovery (%) Hg2+

MeHg

PhHg

± 0.03 ± 0.06

96 99

98 102

104 98

± 0.03 ± 0.08

98 101

98 99

102 103

PhHg ND 0.52 0.98 ND 0.51 1.03

Mean and standard deviation of three experiments. ND = not detected. (10) Margetinova, J.; Houserova-Pelcova, P.; Kuban, V. Anal. Chim. Acta 2008, 615, 115−123. (11) Sherma, J. Anal. Chem. 2002, 74, 2653−2662. (12) Cheng, S.-C.; Huang, M.-Z.; Shiea, J. J. Chromatogr., A 2011, 1218, 2700−2711. (13) Bruno, P.; Caselli, M.; Traini, A. J. High Resolut. Chromatogr. 1985, 8, 135−139. (14) Shekhovtsova, T. N.; Muginova, S. V.; Bagirova, N. A. Mendeleev Commun. 1997, 119−120. (15) Margler, L. W.; Mah, R. A. J. Assoc. Off. Anal. Chem. 1981, 64, 1017−1020. (16) Meermann, B.; Möller, I.; Nowak, S.; Luftmann, H.; Karst, U. J. Anal. At. Spectrom. 2010, 25, 1654−1658. (17) Resano, M.; Ruiz, E. G.; Mihucz, V.; Móricz, Á . M.; Záray, G.; Vanhaecke, F. J. Anal. At. Spectrom. 2007, 22, 1158−1162. (18) Lafleur, J. P.; Salin, E. D. Anal. Chem. 2008, 80, 6821−6823. (19) Beaudin, L.; Johannessen, S. C.; Macdonald, R. W. Anal. Chem. 2010, 82, 8785−8788. (20) Rico, C.; Fernandez-Romero, J.; de Castro, M. D. L. Fresenius J. Anal. Chem. 1999, 365, 320−324. (21) Cody, R. B.; Laramee, J. A.; Durst, H. D. Anal. Chem. 2005, 77, 2297−2302. (22) Andrade, F. J.; Shelley, J. T.; Wetzel, W. C.; Webb, M. R.; Gamez, G.; Ray, S. J.; Hieftje, G. M. Anal. Chem. 2008, 80, 2654− 2663. (23) Harper, J. D.; Charipar, N. A.; Mulligan, C. C.; Zhang, X.; Cooks, R. G.; Ouyang, Z. Anal. Chem. 2008, 80, 9097−9104. (24) Na, N.; Zhao, M.; Zhang, S.; Yang, C.; Zhang, X. J. Am. Soc. Mass Spectrom. 2007, 18, 1859−1862. (25) Symonds, J. M.; Galhena, A. S.; Fernandez, F. M.; Orlando, T. M. Anal. Chem. 2010, 82, 621−627. (26) Jafari, M. T. Anal. Chem. 2011, 83, 797−803. (27) Michels, A.; Tombrink, S.; Vautz, W.; Miclea, M.; Franzke, J. Spectrochim. Acta, Part B 2007, 62, 1208−1215. (28) Ratcliffe, L. V.; Rutten, F. J. M.; Barrett, D. A.; Whitmore, T.; Seymour, D.; Greenwood, C.; Aranda-Gonzalvo, Y.; Robinson, S.; McCoustra, M. Anal. Chem. 2007, 79, 6094−6101. (29) Morlock, G.; Ueda, Y. J. Chromatogr., A 2007, 1143, 243−251. (30) Xing, Z.; Wang, J.; Han, G.; Kuermaiti, B.; Zhang, S.; Zhang, X. Anal. Chem. 2010, 82, 5872−5877. (31) Hayen, H.; Michels, A.; Franzke, J. Anal. Chem. 2009, 81, 10239−10245.

can desorb and atomize mercury species on TLC surface rapidly; therefore, direct detection of the separated mercury species by AFS has been readily achieved with the proposed PJDA sampling technique. The present method offers advantages of being simple and cost-effective and having fast analysis speed and ease of implementation. Because TLC is the simplest chromatographic method and the experimental setup of plasma jet is simple, it provides a possibility to design a portable, miniature, and mobile TLC-PJDA-AFS instrument for field speciation analysis. It is promising for rapid screening of mercury species in clinical samples, pharmaceutical samples, and industrial samples. Means to further improve the readout resolution will need to be addressed. Also, further characterizations of the proposed method are needed to pave the way to real application.

■

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

*Phone: +86-27-6788-3455. Fax: +86-27-6788-3456. E-mail: [email protected] or [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by National Nature Science Foundation China (Nos. 20905066, 41173018) and the Fundamental Research Funds for the Central Universities, China University of Geosciences (Wuhan).

■

REFERENCES

(1) Clevenger, W. L.; Smith, B. W.; Winefordner, J. D. Crit. Rev. Anal. Chem. 1997, 27, 1−26. (2) Clarkson, T. W. Crit. Rev. Clin. Lab. Sci. 1997, 34, 369−403. (3) Ebdon, L.; Foulkes, M. E.; Le Roux, S.; Munoz-Olivas, R. Analyst 2002, 127, 1108−1114. (4) Gao, Y.; De Galan, S.; De Brauwere, A.; Baeyens, W.; Leermakers, M. Talanta 2010, 82, 1919−1923. (5) Krystek, P.; Ritsema, R. Anal. Bioanal. Chem. 2005, 381, 354− 359. (6) Yin, Y.-g.; Wang, Z.-h.; Peng, J.-f.; Liu, J.-f.; He, B.; Jiang, G.-b. J. Anal. At. Spectrom. 2009, 24, 1575−1578. (7) Yin, Y.; Liu, J.; He, B.; Shi, J.; Jiang, G. Microchim. Acta 2009, 167, 289−295. (8) Yin, Y.; Liu, J.; He, B.; Shi, J.; Jiang, G. J. Chromatogr., A 2008, 1181, 77−82. (9) He, Q.; Zhu, Z.; Hu, S.; Jin, L. J. Chromatogr., A 2011, 1218, 4462−4467. 10174

dx.doi.org/10.1021/ac3028504 | Anal. Chem. 2012, 84, 10170−10174