Single-Drop Microextraction Combined with Low-Temperature

Xiaoxiao YuBeibei ChenMan HeHan WangSongbai TianBin Hu .... Simultaneous derivatization and extraction of nitrophenols in soil and rain samples using ...
0 downloads 0 Views 131KB Size
Download PDF
Anal. Chem. 2004, 76, 2910-2915

Single-Drop Microextraction Combined with Low-Temperature Electrothermal Vaporization ICPMS for the Determination of Trace Be, Co, Pd, and Cd in Biological Samples Linbo Xia, Bin Hu,* Zucheng Jiang, Yunli Wu, and Yu Liang

Department of Chemistry, Wuhan University, Wuhan 430072, P. R. China

A new method of single-drop microextraction combined with low-temperature electrothermal vaporization (LTETV)ICPMS was proposed for the determination of trace Be, Co, Pd, and Cd with benzoylacetone (BZA) as both extractant and chemical modifier. Several factors that influence the microextraction efficiency, such as sample flow rate, microdrop volume, and extraction time, were investigated and the optimized microextraction conditions were established. Be, Co, Pd, and Cd in the postextraction organic phase were directly determined by LTETV-ICPMS with the use of BZA as chemical modifier. The chemical modification of BZA in LTETV-ICPMS was studied, and the factors affecting the formation of chelates and vaporization/transportation of chelates were investigated. Under the optimized conditions, the detection limits of the method were 0.12, 0.99, 1.5, and 0.27 pg/mL for Be, Co, Pd, and Cd, and the relative standard deviations for 0.1 ng/mL Be, Co, Pd, and Cd were 16, 14, 14, and 11%, respectively. After 10 min of extraction, the enrichment factors were 160 (Be), 125 (Co), 40 (Pd), and 180 (Cd). The proposed method was applied to the determination of trace Be, Co, Pd, and Cd in biological reference materials, and the determined values were in good agreement with the certified values. Liquid-liquid extraction (LLE) is one of the oldest and classic preconcentration and analyte isolation techniques in analytical chemistry. Although it offers high reproducibility and high sample capacity, it is a time-consuming procedure, which has the tendency for emulsion formation and poor potential for automation.1 The great need for change in analytical sample preparation procedures has led to the development of new methods, whose main advantages are their speed, the negligible volume of solvents, and their ability to allow analytes to be detected at very low concentrations. In the past few years, efforts have been directed toward miniaturizing the LLE procedure by greatly reducing the solventto-aqueous phase ratio, leading to the development of solvent microextraction methodologies.2,3 In 1996, Liu and Dasgupa4 first reported a novel drop-in-drop system where a microdrop of a water-immiscible organic solvent * Corresponding author. Tel: +86-27-87218764. Fax: +86-27-87647617. E-mail: [email protected].

2910 Analytical Chemistry, Vol. 76, No. 10, May 15, 2004

suspended in a larger aqueous drop, extracted sodium dodecyl sulfate ion pairs. The system possessed several advantages such as low consumption of organic solvent and the facility of automatic backwash. At the same time, Jeannot and Cantwell5 introduced a new single-drop microextration (SDME) technique, where a microdrop (8 µL) of organic solvent containing a fixed amount of internal standard was left suspended at the end of a Teflon rod immersed in a stirred aqueous solution containing 4-methylacetophenone. After extracting a prescribed period of time, the microdrop was retracted back into the microsyringe needle and transferred into the GC for further analysis. The same group6 extended their work and used the above drop-based technique to extract unbound progesterone from a protein aqueous solution. He and Lee7 investigated the extraction of 1,2,3-trichlorobenzene by using two different models of microextraction, called static and dynamic SDME. Dynamic SDME was later applied to the analysis of 10 chlorobenzenes in water samples.8 Ma and Cantwell9 subsequently combined solvent microextraction with simultaneous back-extraction into a single drop and achieved sample cleanup and preconcentration prior to HPLC analysis. De Jager and Andrews10 reported preliminary work on SDME for the analysis of 11 organic chlorine pesticides. Parameters such as extraction solvent, sample volume, drop size, stirring speed, and extraction time were monitored. Liu and Lee introduced a new approach to single-drop microextraction, which was termed continuous-flow microextraction.11 A polyetheretherketone tubing, connected to the extraction chamber, served for both the sample delivery of the pumped aqueous solution and the introduction of the extractant solvent. Li and Keller developed a method that involved the use of a nanoliter droplet containing organic solvents at the tip of a small capillary extraction in 2001.12 The single-drop microextraction is a novel sample preparation technique, which offers an attractive alterative to traditional and (1) Prosen, H.; Zupancic, L. Trens Anal. Chim. 1999, 18, 272-279. (2) Lucy, C. A.; Cantwell, F. F. Anal. Chem. 1989, 61, 101-107. (3) Lucy, C. A.; Yeung, K. K.-C. Anal. Chem. 1994, 66, 2220-2225. (4) Liu, H. G.; Dasgupta, P. K. Anal. Chem. 1996, 68, 1817-1821. (5) Jeannot, M. A.; Cantwell, F. F. Anal. Chem. 1996, 68, 2236-2240. (6) Jeannot, M. A.; Cantwell, F. F. Anal. Chem. 1997, 69, 235-239. (7) He, Y.; Lee, H. K. Anal. Chem. 1997, 69, 4634-4640. (8) Wang, Y.; Kwok, Y. C.; He, Y.; Lee, H. K. Anal. Chem. 1998, 70, 46104614. (9) Ma, M. A.; Cantwell, F. F. Anal. Chem. 1999, 71, 388-392. (10) de Jager, L. S.; Andrews, A. R. J. Chromatographia 1999, 50, 733-738. (11) Liu, W. P.; Lee, H. K. Anal. Chem. 2000, 72, 4462-4467. (12) Keller, B. Q.; Li, L. Anal. Chem. 2001, 73, 2929-2936. 10.1021/ac035437e CCC: $27.50

© 2004 American Chemical Society Published on Web 04/07/2004

recently developed extraction techniques. The method has the advantages of high extraction speed and extreme simplicity and is inexpensive and solventless. Psillakis and Kalogerakis recently gave a review on the SDME principle, instrumental design, and application.13 Electrothemal vaporization (ETV) is one of the sample introduction techniques currently employed in plasma atomic emission spectrometry and mass spectrometry. Compared to the conventional pneumatic nebulization sample introduction, it has some merits of high transport efficiency, low sample consumption, low absolute detection limit, and the ability to analyze both liquid and solid samples. Therefore, it has consistently received considerable attention in the atomic/mass spectrometry.14-16 Recent studies on ETV emphasize the importance of chemical modifiers in the successful application of this analytical technique. The research17 has shown that the use of a chemical modifier can effectively suppress the formation of refractory carbides, eliminate the memory effect, improve transportation efficiency, and therefore improve analytical performance of the method. It should be noted that considerable development was achieved in the exploration and application of organic chelating reagents as chemical modifiers in ETV-ICP-AES/MS. Tao18-20 reported the vaporization of oxinates of refractory elements, and these oxinates of V, Cr, and Al could be introduced into the ICP at ∼1000 °C. The research and application of PMBP chelates of rare earth elements were investigated in ETV-ICP-AES, and it was found that PMBP chelates of La, Y, Eu, and Sc could be vaporized from graphite tube at 1000, 1200, 1200, and 900 °C, respectively.21-23 Liao and Jiang used EDTA as a chemical modifier to determine Cd, Hg, and Pb in fish by ETV-ICPMS.24 A relative low vaporization temperature was used that separated the analyte from the major matrix components and increased the ion signals significantly. Lu and Jiang25 reported the use of a mixture of organic acids as the chemical modifier for ETV-ICPMS determination of Zn, Cd, Tl, and Pb in several soil samples. A new approach to matrix modification in which polyhydroxy compounds are used as complexing agents was proposed in 2000.26 The effects as chemical modifiers of eight polyhydroxy components on the sensitivity enhancement of trace elements of interest by ETV-ICPMS are investigated. The successful application of the above-mentioned organic chelating reagents as chemical modifiers in low-temperature ETV-ICP-AES/MS indicates that research and exploitation of organic chelating reagents as effective chemical modifiers may be the potential and prospective development in this regard. (13) Psillakis, E.; Kalogerakis, N. Trends Anal. Chem. 2002, 21, 53-63. (14) Aziz, A.; Broekaert, J. A. C.; Leis, F. Spectrochim. Acta. 1982, 37B, 369373. (15) Ng, K. C.; Caruso, J. A. Appl. Spectrosc. 1985, 39, 719-726. (16) Matusiewicz, H. J. Anal. At. Spectrom. 1986, 1, 171-175. (17) Wu, Y. L.; Hu, B. J. Anal. At. Spectrom. 2002, 17, 121-124. (18) Tao, S.; Kumamaru, T. Appl. Spectrosc. 1996, 50, 785-791. (19) Tao, S.; Kumamaru, T. J. Anal. At. Spectrom. 1996, 11, 111-116. (20) Kumamaru, T.; Notake, H.; Tao, S.; Okamoto, Y. Anal. Sci. 1997, 885889. (21) Peng, T. Y.; Jiang, Z. C. J. Anal. At. Spectrom. 1998, 13, 75-78. (22) Peng, T. Y.; Jiang, Z. C. Chem. J. Chin. Univ. 1998, 19, 699-702. (in Chinese). (23) Chen, S. Z.; Peng, T. Y.; Jinag, Z. C. Anal. Lett. 1999, 32, 411-416. (24) Liao, H. C.; Jiang, S. J. J. Anal. At. Spectrom. 1999, 14, 1583-1588. (25) Lu, H. H.; Jiang, S. J. Anal. Chim. Acta. 2001, 429, 247-255. (26) Wei, W. C.; Yang, M. H. J. Anal. At. Spectrom. 2000, 15, 1466-1473.

Use of organic chelating reagents as chemical modifiers in the ETV-ICP-AES/MS has the following advantages: refractory elements can be vaporized at lower temperature, which is beneficial to prolong the life of evaporator, and by combining this technique with a chemical separation/preconcentration, a much lower detection limit could be obtained. As described previously, SDME is a miniaturized sample pretreatment technique and ETV is a microamount sample introduction technique; therefore, it will be a perfect combination if SDME is combined with ETV-ICPMS. Be, Co, and Cd have been recognized as human carcinogen metals or possible carcinogens by international agencies27 (International Agency for Research on Cancer, European Union). And palladium is often used as a catalyst in industrial chemical and pharmaceutical synthesis. Because of the allergenic potential of palladium, this metal is of great concern to environmental and occupational medicine. Specific and sensitive analytical methods for effective environmental and biological monitoring need to be elaborated for the above four elements. The aim of this work is to combine SDME with the low-temperature ETV-ICPMS and to develop a new method of SDME-LTETV-ICPMS for the determination of the trace Be, Co, Pd, and Cd in biological standard reference materials. The factors that influence the efficiency of the extraction, the formation of chelates, and the vaporization and transportation of chelates were systematically investigated. Two modes of SDME, continuous-flow and cycle-flow microextraction were preliminarily investigated, and the results showed that both extraction modes provided good extraction efficiency in 10 min. EXPERIMENTAL SECTION Standard Solution and Reagents. All laboratory ware was made of polyethylene or polypropylene material and thoroughly cleaned by soaking in nitric acid (1 + 1) for at least 24 h. Immediately prior to use, all acid-washed ware was rinsed with doubly distilled water. The stock solutions of Cd and Co (1 mg/mL) were prepared from analytical reagent grade CdCl2‚21/2H2O and CoCl2‚6H2O, by dissolving the appropriate amounts in 10 mL of concentrated HCl and making up the volume in 1 L. The stock standard solution of Be (1 mg/mL) was prepared from analytical reagent grade BeSO4‚4H2O by dissolving the appropriate amount in 1 mL of concentrated H2SO4 and making up to volume in 100 mL. A Pd stock standard solution (1 mg/mL) was prepared by dissolving 0.1000 g of Pd metal in 50 mL of aqua regia [a mixture (3 + 1) of concentrated HCl and HNO3], and the solution was evaporated on a water bath to near-dryness. The solution was diluted to 100 mL with 10 mL of concentrated HCl and water successively. The working solutions of the metal ions were made by suitable dilution of the stock solutions with doubly distilled water. The 0.1 mol L-1 solutions of benzoylacetone (BZA) were prepared by dissolving appropriate amount of BZA (Fluka Chemie, AG CH-9470) in 25 mL of benzene. All reagents used were of Specpure grade or at least of analytical reagent grade. Doubly distilled water was used throughout this work. (27) Beyersmann, D. Toxicol. Lett. 2002, 127, 63-68.

Analytical Chemistry, Vol. 76, No. 10, May 15, 2004

2911

Figure 1. Typical signal profiles of Be and Co: drying, 60 °C, ramp, 4 s, hold, 4 s; vaporization time, 4 s. (A) 20 pg Be with 0.1 µmol of BZA as chemical modifier vaporized at 900 °C. (A′) Residual signal of empty firing at 2500 °C. (B) 20 pg Co with 0.1 µmol of BZA as chemical modifier vaporized at 900 °C. (B′) Residual signal of empty firing at 2500 °C. (C) 20 pg of Be without BZA as chemical modifier vaporized at 2500 °C. (C′) Residual signal of empty firing at 2500 °C. (D) 20 pg of Co without BZA as chemical modifier vaporized at 2500 °C. (D′) Residual signal of empty firing at 2500 °C. (E) 20 pg of Be without BZA as chemical modifier vaporized at 900 °C. (E′) Residual signal of empty firing at 2500 °C. (F) 20 pg of Co without BZA as chemical modifier vaporized at 900 °C. (F′) Residual signal of empty firing at 2500 °C.

Instrumentation. Microextration System. The continuous-flow microextraction system is similar to that in ref 11 with minor modification. A PFA tubing (0.30-mm i.d.) was used to connect the extraction chamber (∼0.2-mL volume), and an HL-2 pump (Qingfu, Shanghai) and a 10-µL microsyringe (Gaoxin, Shanghai) were employed to introduce the extracting solution of BZA. For the cycle-flow microextraction system, the waste outlet of tubing was put into the sample reservoir. ETV-ICPMS Apparatus. The analytes (Be, Co, Pd, Cd) in an organic phase were determined with an Angilent system equipped 2912

Analytical Chemistry, Vol. 76, No. 10, May 15, 2004

with a modified commercially available WF-4C graphite furnace (Beijing Second Optics, Beijing, China) as the electrothermal vaporizer. The original silica windows at the two ends of the graphite furnace were removed and replaced by two PTFE cylinders, as detailed in previous work.28 Optimization of the ICPMS instrument (i.e., lens settings, resolution, oxide, and doubly charged ion formation) was performed with a conventional nebulization method (PN-ICPMS) prior to connect with the ETV (28) Chen, S. Z.; Hu, B. J. Anal. At. Spectrom. 1999, 14, 1723-1726.

Figure 2. Effect of vaporization temperature on signal intensity. Conditions: Be, Co, Pd, and Cd 20 pg mixed with 0.1 µmol of BZA; drying, 60 °C, ramp, 4 s, hold, 4 s; vaporization time, 4 s.

Figure 3. Effect of flow rate on signal intensity in cycle-flow microextraction. Conditions: Be, Co, Pd, and Cd 0.1 ng/mL, aqueous volume, 1 mL; drop volume, 4 µL; extraction time, 10 min.

device. Pyrolytic graphite-coated graphite tubes were used throughout the work. The operating conditions for ICPMS and ETV are as follows: rf power, 1250 W; outer gas flow rate, 15 L min-1; intermediate gas flow rate, 0.9 L min-1; nebulizer gas flow rate, 0.7 L min-1; sampling depth, 7.0 mm; sampler/skimmer diameter orifice, nickel 1.0 mm/0.4 mm; scanning mode, peak-hopping; dwell time, 10-40 ms; integration mode, peak area; sample volume, 3-5 µL; carrier gas flow rate, 0.4 L min-1; drying step, 60 °C ramp 4 s/hold 4 s; vaporization step, 900 °C, hold 4 s. Procedure. Extraction Procedures. (1) Continuous-Flow Microextraction. When the extraction chamber has been filled with the sample solution, a 10-µL microsyringe was used to take extraction solution and push out 3-5 µL of extraction solution to form a drop above the PFA tubing outlet in extraction chamber. As the fresh sample solution flowed around the microdroplet continuously, the trace analytes were kept extracting into the solvent drop from the sample solution. After the extraction was finished, the postextraction microdroplet was withdrawn by the microsyringe. The droplet was determined by low-temperature ETV-ICPMS. (2) Cycle-Flow Microextraction. To compare with the continuous-flow microextraction, a cycle-flow microextraction system was

Figure 4. Effect of extraction time on signal intensity in cycle-flow system. Conditions: Be, Co, Pd, and Cd 0.1 ng/mL, aqueous volume, 1 mL; drop volume, 4 µL; flow rate, 0.2 mL/min.

Figure 5. Effect of extraction time on signal intensity in continuousflow system. Conditions: Be, Co, Pd, and Cd 0.1 ng/mL, drop volume, 4 µL; flow rate, 0.2 mL/min.

employed, which was similar to the continuous-flow microextraction system with a modification that the waste outlet of tubing was put into the sample reservoir, and the sample volume was set as 1 mL. Therefore, after the whole extraction system (both extraction chamber and connection tube) filled with the sample solution, the volumes of sample solution in the sample reservoir, connecting tubing, and extraction chamber were about 0.2, 0.6, and 0.2 mL, respectively. The extraction was performed in the same way as continuous-flow microextraction, and after the extraction was finished, the droplet was injected to ETV-ICPMS for further analysis. (3) ETV Analysis. After the ETV unit was connected to the ICPMS and the system was stabilized, 3-5 µL of sample was injected into the graphite furnace. During the drying step of the temperature program, the dosing hole of the graphite furnace was kept open to remove the water and other vapors. When the dosing hole was sealed with a graphite probe 5-10 s prior to the hightemperature vaporization step, the vaporized analytes were swept into the plasma excitation source by a carrier gas of argon and the peak-hop transient mode of data acquisition was used to detect the ions selected. Analytical Chemistry, Vol. 76, No. 10, May 15, 2004

2913

Table 1. Limits of Detection and the Relative Standard Deviations (RSDs) Obtained in the Analysis of Be, Co, Pd, and Cd by ETV-ICPMS actual detection limit enrichment RSD (%) instrument methoda methodb factor (fold) element (ng/mL) (pg/mL) (pg/mL) a b a b Be Co Pd Cd a

0.019 0.124 0.056 0.048

0.12 0.99 1.5 0.27

0.072 0.56 0.83 0.16

160 125 40 180

260 215 70 300

16 14 14 11

16 14 14 12

Cycle-flow extraction method. b Continuous-flow extraction method.

RESULT AND DISCUSSION Chemical Modification of BZA. Volatility Behaviors of BZA Chelates of Be and Co. BZA is an excellent chelating reagent of metals due to its good thermal stability and unique volatility. For example, the melting points of benzoylacetonates of beryllium(II) and cobalt(II) are 213 and 181 °C (0.25 mmHg), and their sublimation temperatures are 105 and 162 °C, respectively.29 Therefore, there is feasibility to vaporize chelates of Be(II), Co(II), and BZA for gaseous introduction in ETV-ICPMS. Figure 1 is the typical signal profile of Be and Co obtained by ETV-ICPMS with/without the use of BZA as chemical modifier. As could be seen, an intense and sharp signal profile could be detected at 900 °C (Figure 1A, B) with the addition of BZA as chemical modifier, and no memory effect could be observed (Figure 1A′, B′) by empty firing at 2500 °C. On the contrary, almost no signal could be found for the same concentration of Be and Co aqueous standard solution at the same vaporization temperature of 900 °C (Figure 1E, F) without BZA as modifier, and a severe memory effect could be observed at 2500 °C (Figure 1E′, F′). It should be pointed out that without the use of BZA as chemical modifier, the aqueous solution of the same concentration of Be and Co could be quantitatively vaporized at 2500 °C (Figure 1C, D), and almost no memory signal could be detected (Figure 1C′, D′). These results demonstrated that, using the chemical modifier BZA, the analytes Be and Co reacted with BZA to form the Be- and Co-BZA chelates, so the analytes were vaporized at a lower temperature and introduced into the ICP as the gaseous BZA chelates, at the same time, the memory effect could be effectively eliminated. There are many factors affecting the vaporization behaviors of Be-, Co-, Pd-, and Cd -BZA chelates; therefore, it is necessary to investigate these influencing factors and optimize the experimental conditions. Optimization of ETV Parameters for Analytes. Under the selected drying temperature of 60 °C, the effect of drying time on the vaporization behaviors of Be-, Co-, Pd-, and Cd-BZA chelates was investigated. It was found that the drying time had no significant influence on the signal intensity of analytes and it kept constant after the drying time was more than 3 s. The drying time of 4 s was chosen in this work. The effect of vaporization temperature on the signal intensity was studied, and the results are shown in Figure 2. At the temperature of 300 °C, a weaker signal intensity could be detected for all the analytes. With an increase of vaporization temperature, (29) Berg, E. W.; Truemper, J. T. Anal. Chim. Acta 1965, 32, 245-252.

2914 Analytical Chemistry, Vol. 76, No. 10, May 15, 2004

the signal intensity increased. The maximum analytical signal intensity was obtained at 800 and 700 °C for Be, Pd and Co, and Cd, respectively. With further increase in the vaporization temperature to the highest temperature tested, 1100 °C, the analytical signal intensity of Co and Pd were kept constant. However, for Be and Cd, the maximum analytical signal intensity remained constant until the vaporization temperatures of 1000 and 800 °C was reached. With further increase in the vaporization temperature, the Be-BZA and Cd-BZA chelates began to decompose and the signal intensity decreased. Compared with using no BZA, the vaporization temperatures of Be, Co, Pd, and Cd with the use of BZA as chemical modifier decreased from 2500 to 800 °C, 2500 to 700 °C, 2700 to 800 °C, and 1600 to 700 °C, respectively. The effect of vaporization time on the vaporization behaviors of Be-, Co-, Pd-, and Cd-BZA chelates were studied, and the experimental results showed that the highest signal intensity could be achieved after the vaporization time was larger than 3 s. Based on the above results, the drying temperature of 60 °C, and drying time of 4 s, the vaporization temperature of 900 °C and vaporization time of 4 s were selected as the compromised ETV conditions for ETV-ICPMS simultaneous determination of Be, Co, Pd, and Cd in this work. Single-Drop Microextraction of Analytes. Several factors that influence the microextraction efficiency, such as sample flow rate, microdrop volume, and extraction time, were studied in the following work. Influence of the Sample Flow Rate. The flow rate of sample solution would affect extraction dynamics remarkably since the thickness of the interfacial layer surrounding the microdroplet will vary with the change of flow rate. The effect of sample flow rate was evaluated with the change of flow rate from 0.05 to 0.5 mL/min in both cycle-flow microextraction and continuous-flow microextraction systems. Figure 3 shows the concentration changes of analytes in postextraction organic phase with the change of sample flow rates in cycle-flow microextraction. The final concentration of analytes (Be, Co, Cd) in organic phase increased with the increase of sample flow rate, which indicated that higher flow rate provided more effective microextraction of analytes. However, when the flow rate is above 0.2 mL/min, although extraction efficiency was better, air bubble formation frequently occurred, leading to quantitation problems. On the contrary, the final concentration of Pd in organic phase decreased with the increase of sample flow rate, which indicated that lower flow rate provided more effective microextraction for Pd. The possible reason for this is that Pd-BZA-benzene extraction dynamic is slower than the others. The similar results could be obtained in continuous-flow microextraction. Therefore, a sample flow rate of 0.2 mL/min was selected in this work. Influence of Droplet Volume. Both cycle-flow microextraction and continuous-flow microextraction were employed to investigate the influence of droplet volume on the extraction efficiency of Be, Co, Pd, and Cd. The microextraction was performed for 10 min at a sample flow rate of 0.2 mL/min. It shows that no notable effect of drop volume on extraction efficiency was observed with the volume of organic solvent microdrop ranging from 3 to 5 µL in both microextraction systems. In our experiments, it was also found that the larger drops cannot tolerate the

Table 2. Analytical Results (Mean ( SD, n ) 3) for Be, Co, Pd, and Cd in Synthetic Water Samples (ng/mL) added (ng/mL)

determined (ng/mL)

Be

Co

Pd

Cd

Be

Co

Pd

Cd

0.5 1.0 5.0

0.5 1.0 5.0

0.5 1.0 5.0

0.5 1.0 5.0

0.52 ( 0.09 1.04 ( 0.10 5.05 ( 0.09

0.53 ( 0.07 1.06 ( 0.09 5.06 ( 0.10

0.46 ( 0.07 0.94 ( 0.07 4.94 ( 0.09

0.52 ( 0.06 0.98 ( 0.06 5.05 ( 0.08

Table 3. Analytical Results (Mean ( SD, n ) 3) for Be, Co, Pd, and Cd in Human Urine and Hair Reference Samples measured value sample

Be

GBW09101 human hair (µg/g) GBW09103 human urine (µg/mL)

0.036 ( 0.007

Co 0.130 ( 0.011

high sample flow rate; the microdrops easily fall off the needle of the microsyringe. On the basis of our observations, a droplet of 4 µL was selected in this work. Influence of Extraction Time. Although the maximum sensitivity is attained when microextraction is at equilibrium, it takes a longer time for microextraction to reach complete equilibrium. This may result in drop dissolution and have a high incidence of drop loss and, thus, poor accuracy and precision. To investigate the influence of extraction time on the microextraction, an extraction of 1 mL of sample solution was performed at a sample flow rate of 0.2 mL/min. Figure 4 is the effect of extraction time on the cycle-flow microextracrtion of Be, Co, Pd, and Cd. As can be seen, the total amount of analytes extracted into organic phase increased with the increase of extraction time to 15 min, and after 15-min extraction, the speed of the increasing intensity became slower. To trade off the analytical speed and the highest extraction efficiency, the extraction time of 10 min was employed for further study. The influence of extraction time was also investigated in continuous-flow microextraction. As shown in Figure 5, it takes a longer time to reach equilibrium compared with cycle-flow microextraction. Reproducibility, Linearity, and Enrichment Factor. Calibrations were carried out by introducing 4 µL of the organic phase into ETV using standard solutions with the preconcentration system. The reproducibility of the method was studied for eight replicate experiments for an aqueous sample spiked at 0.1 ng/ mL of the analytes (Be, Co, Pd, Cd) extracted by BZA and benzene solution. The relative standard deviations (RSDs) for two microextraction modes were from 11 to 16% and from 12 to 16%, respectively. Linearity was obtained over the range of 0.01-50 ng/mL and the coefficient of correlation (r2) ranged from 0.9897 to 0.9901. The enrichment factor, defined as the ratio of ETVICPMS signals after extraction and that before extraction, was used to evaluate the extraction efficiency. It should be noted that >40-fold enrichment for the cycle-flow microextraction system and >70-fold for the continuous-flow microextraction system were achieved in only 10 min. The results, together with the limits of detection are listed in Table 1 for comparison.

Pd

certified value Cd

Be

0.089 ( 0.016 0.048 ( 0.007

0.031 ( 0.002

Co 0.135 ( 0.008

Pd

Cd 0.095 ( 0.012 0.053 ( 0.003

The limit of detection, defined as three times the standard deviation of the method blank, was determined by analyzing three replicates of the method blank. Samples Analysis. Synthetic Water Samples. Synthetic water samples were analyzed to check the accuracy and precision of the proposed method, and the analytical results were listed in Table 2. The calibration was obtained by standard solutions subjected to microextraction as well. As could be seen, the concentrations of Be, Co, Pd, and Cd obtained with the proposed method were in good agreement with the expected values, when their concentrations ranging from 0.5 to 5.0 ng/mL. Biological Samples. To further verify the accuracy of the method, the developed method was applied to the determination of Be, Co, Pd, and Cd in human hair (GBW09101) and human urine (GBW09103) biological standard reference materials, and the analytical results were given in Table 3. For determination of Be, Co, Pd, and Cd in human hair, 10.0 mg of human hair standard reference material (GBW09101, provided by Perambulation Institute of Physical Geography and Geochemistry of Geological and Mineral Ministry, Langfang, China) was weighed and transferred into a 10-mL PTFE beaker. Then 1 mL of concentrated HNO3 and 0.2 mL of HClO4 were added to the beaker, and the sample was digested by warming at low temperature; at the same time, 0.4 mL of 30% H2O2 was added drop by drop to the beaker. After the sample was dissolved completely, the resulting sample solution were heated to near dryness, then dissolved with buffer solution, and finally diluted to 2 mL with doubly distilled water for further determination. Human urine (GBW09103, provided by School of Public Health of Beijing Medical University, China) was directly determined after being diluted 10 times by buffer solution. ACKNOWLEDGMENT National Nature Science Foundation of China (Grant 20375030 and 20175014) and Wuhan Municipal Science & Technology Committee are acknowledged for financial support. Received for review December 4, 2003. Accepted March 2, 2004. AC035437E

Analytical Chemistry, Vol. 76, No. 10, May 15, 2004

2915