Article pubs.acs.org/ac
Single Drop Solution Electrode Glow Discharge for Plasma AssistedChemical Vapor Generation: Sensitive Detection of Zinc and Cadmium in Limited Amounts of Samples Zhi-ang Li,† Qing Tan,§ Xiandeng Hou,†,‡ Kailai Xu,† and Chengbin Zheng*,† †
Key Laboratory of Green Chemistry & Technology, Ministry of Education, College of Chemistry, Sichuan University, Chengdu, Sichuan 610064, China ‡ Analytical & Testing Center, Sichuan University, Chengdu, Sichuan 610064, China § Chengdu Environmental Monitoring Center, Chengdu, Sichuan 610072, China S Supporting Information *
ABSTRACT: A simple and sensitive approach is proposed and evaluated for determination of ultratrace Zn and Cd in limited amounts of samples or tens of cells based on a novel single drop (5−20 μL) solution electrode glow discharge assisted-chemical vapor generation technique. Volatile species of Zn and Cd were immediately generated and separated from the liquid phase for transporting to atomic fluorescence or atomic mass spectrometric detectors for their determination only using hydrogen when the glow discharge was ignited between the surface of a liquid drop and the tip of a tungsten electrode. Limits of detection are better than 0.01 μg L−1 (0.2 pg) for Cd and 0.1 μg L−1 (2 pg) for Zn, respectively, and comparable or better than the previously reported results due to only a 20 μL sampling volume required, which makes the proposed technique convenient for the determination of Zn and Cd in limited amounts of samples or even only tens of cells. The proposed method not only retains the advantages of conventional chemical vapor generation but also provides several unique advantages, including better sensitivity, lower sample and power consumption, higher chemical vapor generation efficiencies and simpler setup, as well as greener analytical chemistry. The utility of this technique was demonstrated by the determination of ultratrace Cd and Zn in several single human hair samples, Certified Reference Materials GBW07601a (human hair powder) and paramecium cells.
R
plasma mass spectromtry (ICPMS).8 However, the high cost and the inflexible sample uptake may limit their further applications. Some nonpneumatic nebulization techniques include electrothermal vaporization (ETV),9 laser ablation (LA),10,11 and monodisperse droplet sample introduction systems,12−15 which offer attractive alternatives to micropneumatic nebulization for analyzing limited amounts of precious samples. Although these sampling methods have been successfully used for analysis of samples at the microliter level, they more or less retain some shortcomings, such as serious memory effects and low transport efficiency from an ETV, expensive equipment required for LA, and the increased risk of clogging, difficulty in cleaning, and rather expensive use of monodisperse droplet sample introduction systems. Because of the tremendous potential of analysis at the microliter or single cell level, many studies have focused on improvement of these techniques. Inexpensive diode lasers successfully served to construct an inexpensive and simple sample introduction system for analysis of microamounts of sample.16 Günther et
ecently, there has been a surge in the development of analytical methods to analyze ultratrace elements in very small amounts of samples, even in a single cell.1−4 Pneumatic nebulization coupled to atomic spectrometry is primarily used for routine elemental analysis because of its simplicity, low cost, and robustness. However, there remain a number of serious impediments for further evolution of conventional pneumatic nebulization, including low sensitivity, high sample consumption, and serious matrix interferences. Therefore, it is still attractive to develop new sampling techniques for atomic spectrometry to analyze small amounts of samples with complex matrixes associated with biological, forensic, toxicological, and clinical studies.5 Increasing the volume of sample with dilution can form part of the solution, but this leads to a significant decrease in signal intensity. Many efforts have thus been devoted to develop new techniques for highly sensitive analysis of samples based on micro- or even nanoliter (microgram) levels. Reducing the size of the nebulizer nozzle can be adopted to effectively decrease sample consumption and improve sample introduction efficiency.6,7 Todoli et al. reviewed the application of low-flow nebulizers for the analysis of liquid microsamples by inductively coupled plasma-atomic emission spectroscopy (ICP-AES) or inductively coupled © 2014 American Chemical Society
Received: August 4, 2014 Accepted: November 16, 2014 Published: November 19, 2014 12093
dx.doi.org/10.1021/ac502911p | Anal. Chem. 2014, 86, 12093−12099
Analytical Chemistry
Article
Figure 1. (a) Schematic of the SD-SEGD-CVG system; and (b) photograph of the vapor generator.
al.17 developed an inexpensive and disposable microfluidicsbased droplet dispenser and used it as a sampling system for ICPMS to analyze samples of less than 1 μL or at the single cell level. Matrix and spectral interferences are inevitable in these techniques since sample matrixes are introduced into the atomizer or plasma together with analytes. Chemical vapor generation (CVG) is a completely different sampling approach, which introduces gaseous “analyte”, not the whole sample, to analytical atomic spectrometers. Therefore, CVG significantly alleviates matrix interferences and enhances sample transport efficiency, and thus results in an improvement in detection limits, sensitivity and throughput when coupled to atomic spectrometry.18−21 However, application of CVG for analysis of microamounts of sample is rather limited, apart from a single report using a movable reduction bed generator by Wang et al.22 nearly 20 years ago. Moreover, conventional CVG techniques require mineral acids and reductants, thereby resulting in elevated blanks and serious secondary contamination.23 Therefore, the aim of this work was to develop a novel CVG technique to determine ultratrace elements in limited amounts of sample. Although solution cathode glow discharge assisted Hg, I, and Os vapor generations have been realized by Zhu et al.,24−26 applications to other elements analyses have not yet been reported. We report here a novel CVG method using solution electrode glow discharge (SEGD) energized by a compact ac ozone generation power supply to the highly efficient generation of volatile species of cadmium and zinc from a single drop (5−20 μL) only using hydrogen. To the best of our knowledge, this is the first report of use of SEGD for simultaneous generation of volatile species of Zn and Cd from a liquid drop, and this green analytical method can be utilized for the determination of ultratrace zinc and cadmium in small amounts of samples, such as a single human hair or tens of cells.
water (DIW) from a water purification system (Chengdu Ultrapure Technology Co., Ltd., China) was used to prepare all solutions. Standard solutions were prepared daily by serial dilution of the stock standard solution (1000 mg L−1) of Zn and Cd obtained from the National Research Center of China (NRCC, Beijing, China). HCl, HNO3, MgCl2·6H2O, CaCl2, and other chemicals were purchased from Kelong Chemical Factory (Chengdu, China). High-purity Ar (99.999%) was obtained from Qiaoyuan Gas Co. (Chengdu, China). Hydrogen was produced from a hydrogen generator (SPGH-300, Zhongya Gas & Instrument Research Institute, Beijing, China). Certified Reference Materials (CRMs, human hair GBW07601a) and two human hair samples were obtained from the National Research Center for Standard Materials (Beijing, China) and colleagues, respectively. Paramecium cells were purchased from the College of Life Science, Hebei University, China. Instrumentation. An inductively coupled plasma atomic emission spectrometer (Arcos FHS12, Spectro Analytical Instruments GmbH, Germany), an atomic fluorescence spectrometer (AFS-8220) (Beijing Titan Instruments Co. Ltd., Beijing, China) or an ICPMS (7700x, Agilent Technologies) coupled to a single drop solution electrode glow discharge assisted-CVG (SD-SEGD-CVG) system was used for sample analysis. Optimization of experimental conditions and investigation of the mechanism of the vapor generation was carefully undertaken. A schematic diagram of the SD-SEGD-CVG system interfaced to AFS/ICP-AES/ ICPMS is illustrated in Figure 1a. The system mainly consisted of a peristaltic pump (BT100-02, Baoding Qili Precision Pump Co., Ltd.), a six port injection valve (Genuine Rheodyne Co.) equipped with 5, 10, and 20 μL sampling loops and a SDSEGD-CVG generator that integrates both the generation and gas−liquid separation functions. The closed generator consisted of a quartz tube (8 mm i.d. × 10 mm o.d. × 8.5 cm length), a tapered tungsten electrode, and a stainless steel tube (0.5 mm i.d. × 1.5 mm o.d. × 5 cm length). The photograph of the generator is shown in Figure 1b. The steel tube and the tungsten electrode were inserted into the quartz tube to hang
■
EXPERIMENTAL SECTION Reagents. All reagents used in this work were of at least analytical reagent grade. High purity (18.2 MΩ cm) deionized 12094
dx.doi.org/10.1021/ac502911p | Anal. Chem. 2014, 86, 12093−12099
Analytical Chemistry
Article
well-known that the atomic emission lines of Mg and Ca are probably among the strongest ones from the ICP. Thus, the Cd and Zn would not be only introduced into ICP via aerosols from the discharge sputtering if atomic emission signals of Ca and Mg are not observed during SEGD-CVG. Then, a standard solution containing 100 μg L−1 Cd and Zn spiked with 100 μg L−1 Mg and Ca, respectively, was processed by using 600 mL min−1 Ar and 50 mL min−1 H2 as carrier gas and discharge gas. The volatile species were transported to ICP-AES and atomic emission at 228.8, 213.8, 279.6, and 393.4 nm were monitored for measurements of Cd, Zn, Mg, and Ca, respectively, as shown in Figure 2. The results show that obvious signals of Zn
the sample drop and use as the lower electrode, respectively. The discharge gap defined as the distance between the surface of the liquid drop and the tip of the tungsten electrode was set at about 1 mm. The plasma was ignited and sustained in the gap when a high voltage delivered by a compact ac ozone generation power supply (YG.BP105P; 6 cm long × 4 cm wide × 3 cm high; with a rated output of 4 kV, 20 kHz, and 12 W at 220 V, 50 Hz input; Electronic Equipment Factory of Guangzhou Salvage, Guangzhou, China) was energized between the tungsten electrode and stainless steel tube. The cells were characterized using a Nikon E100 microscope (Nikon, Japan). Cells Sample Preparation. Paramecium cells were incubated in a solution containing 10 mg L−1 Cd(II) for 24 h at 25 °C. The living paramecium cells were successively separated from large particles and free Cd(II) by using two mesh nylon filters (pore size, ∼500 and 50 μm) and quickly washed three times with DIW. The washed cells were then transferred to a quartz bottle with DIW and analyzed within 2 h. The obtained cell solution contains about 50 paramecium cells per 20 μL. A volume of 20 μL of the obtained paramecium solution was transferred into a tube for 30 min of ultrasonic radiation to release the Cd(II) from paramecium prior to analysis by SD-SEGD-CVG-AFS. In order to compare the results obtained by the proposed method and direct determination by ICPMS, paramecium cells were separated from 10 mL of the obtained paramecium solution (containing about 25 000 paramecium cells) by centrifuging for 10 min at 10 000 rpm and then 2 mL of DIW were added. This 2 mL of cell solution was sonicated for 30 min and directly analyzed by pneumatic nebulization ICPMS. The preparations of single human hair samples and CRM were described in Section 1 of the Supporting Information. Analytical Procedure. Standard solution or sample was initially directed to a 20 μL sample loop through a six-port valve with the aid of a peristaltic pump. The valve was activated to pass air so as to flush the analyte solution to form a liquid drop at the outlet of the steel tube. Once a 60 V input voltage was supplied to the electrodes, the discharge plasma was spontaneously ignited and the volatile species of Cd and Zn were immediately generated. The plasma was sustained for 10 s. During this time, the gaseous products were simultaneously separated from the liquid phase and further transported by the carrier gas to the Ar-H2 flame atomizer of AFS or the ICP torch for their measurements. All instrumental parameters and experimental conditions were carefully optimized for maximum response for Zn and Cd and the optimum parameters for AFS, ICP-AES, and ICPMS are summarized in Table S1 (see Section 2 of the Supporting Information). A blank was measured before each run based on use of DIW. The compounds of cadmium are extremely toxic, and some unknown products may be generated during the vapor generation procedure. Thus, the whole manipulations were carried out in a room with good ventilation/exhaust systems.
Figure 2. Comparisons of the signals from 100 μg L−1 of Cd(II), Zn(II), Mg(II), and Ca(II) obtained by the SD-SEGD-CVG-OES. Experimental conditions: sample volume, 20 μL; Ar carrier gas flow rate, 600 mL min−1; and discharge voltage, 60 V.
and Cd can be observed while no obvious signal was observed for Mg and only one-third of signal of Zn was obtained for Ca. Meanwhile, the Ca signal peaks are not simultaneously appeared and do not have the same temporal profiles as those for Cd and Zn. Moreover, only about a quarter of the total volume of liquid drop was consumed when the glow discharge was sustained for 10 s. Therefore, we may conclude that a significant amount of analyte is being transported to the ICP by vapor generation. In order to identify these volatile species, a comparison of the signals from a standard solution containing 10 μg L−1 Cd(II) and 50 μg L−1 Zn(II) obtained by atomic fluorescence spectrometry with and without an Ar-H2 flame was first undertaken, as shown in Figure 3. It should be noted that 20 mL min−1 of H2 was introduced to the SDSEGD-CVG generator together with 500 mL min−1 of Ar as both discharge gas and carrier gas in this investigation. Meanwhile, an auxiliary hydrogen gas was introduced behind the CVG generator and mixed with the carrier gas in order to form a stable argon-hydrogen flame for efficient atomization of
■
RESULTS AND DISCUSSION Characterization of the Volatile Species. Glow discharge sputtering is widely used as a sampling technique in atomic emission spectrometry and mass spectrometry for analysis of solid or liquid samples,27,28 and the initial experiment is thus to minimize the possibility that transport of the analytes and hence the response was originated from formation of a fine aerosol by glow discharge sputtering. It is 12095
dx.doi.org/10.1021/ac502911p | Anal. Chem. 2014, 86, 12093−12099
Analytical Chemistry
Article
the gaseous species. For cadmium, an obvious signal was found when the flame atomizer (see Cd-1 in Figure 3) was not ignited but much lower than that observed with Ar-H2 flame atomization (see Cd-2 in Figure 3), which indicates that both free atomic vapor and molecular volatile species of Cd were generated from the SD-SEGD-CVG system. This agrees with reports using DBD-CVG,29 conventional hydride generation (HG),30 and electrochemical hydride generation (EcHG).31 As no other chemical was involved in this reaction besides H2, the volatile species of Cd was thus presumed to be cold atomic vapor (Cd0) and hydrides of Cd (CdH2). As expected, the small signal of Zn was detected in the absence of an Ar-H2 flame atomization (see Zn-1 in Figure 3), while a significant signal was obtained in the presence of the flame atomizer (see Zn-2 in Figure 3). This implies most of the volatile species of Zn in the SD-SEGD-CVG is not atomic species (Zn0) but hydride. This also agrees well with previous work that reported the volatile specie of zinc generated from conventional HG and electrochemical HG is a hydride.32,33 Previous work29 reported that hydrogen was essential for generation of volatile species of Cd from the DBD-CVG system. Subsequently, the role of H2 played in the proposed method was investigated with formation of a stable Ar-H2
Figure 3. Typical response from 20 μL of 10 μg L−1 of Cd and 50 μg L−1 of Zn by SD-SEGD-CVG-AFS under different experimental conditions. Cd-1 and Zn-1, using 20 mL min−1 H2 and 500 mL min−1 Ar as discharge gas and without the Ar-H2 flame atomizer; Cd-2 and Zn-2, using 20 mL min−1 H2 and 500 mL min−1 Ar as discharge gas and with Ar-H2 flame atomizer; Cd-3 and Zn-3, only using 500 mL min−1 Ar as discharge gas and with an Ar-H2 flame atomizer.
Figure 4. Optimization of the parameters of SD-SEGD-CVG: (a) effect of input voltage on responses; (b) effect of the flow rate of the Ar discharge gas; (c) effect of the discharge gap; and (d) effect of the pH of the solution. 12096
dx.doi.org/10.1021/ac502911p | Anal. Chem. 2014, 86, 12093−12099
Analytical Chemistry
Article
flame, as shown in Figure 3. In contrast to the previous work, the signal for Cd obtained by the proposed method without H2 (see Cd-3 in Figure 3) can be improved about 2-fold compared to that obtained from the case in the presence of H2 (see Cd-2 in Figure 3), which implies that volatile species of Cd are easily generated from the SD-SEGD-CVG even in the absence of hydrogen. Figure 3 also shows that the signal obtained in absence of H2 is just 5% of that obtained in the presence of hydrogen (see Zn-3 in Figure 3). Although the real mechanism of the proposed method remains not completely known and further investigation is needed because the generation of elemental volatile species with this plasma is rather complicated, we speculate that the mechanism for generation of these volatile species may involve reducing species such as H• radicals and electrons, arising from dissociation of H2O and H2 by the plasma.34,35 It is necessary to add hydrogen to provide a stronger reducing atmosphere for vapor generation of zinc because more critical conditions are required for generation of Zn volatile species than that for Cd. Optimization of Single Drop Solution Electrode Glow Discharge CVG. As this is the first attempt to use microliter levels of liquid drops as the electrode for generation of volatile species of Cd and Zn by liquid electrode glow discharge, it is necessary to carefully investigate all physicochemical parameters related to generation and detection of the volatile species. A standard solution containing 10 and 50 μg L−1 Cd and Zn was used to optimize each influencing factor. The effect of input voltage on responses was investigated from 50 to 90 V. The results are summarized in Figure 4a and show response from both Cd and Zn is increased with voltage within the range of 50−60 V, followed by an obvious decrease at higher voltage. The discharge plasma could not be operated when the input voltage was lower than 50 V. Unsurprisingly, the bulk solution was vigorously sputtered and immediately generated excess water vapor when the voltage was higher than 80 V (see the photos in Figure S1 of the Supporting Information). Presumably, the analytes were sputtered prior to the vapor generation and deposited on the surface of the generator and transport tube. Moreover, the discharge became unstable when the voltage was higher than 60 V. Thus, an input voltage of 60 V was used for subsequent experiments. Other operational parameters of SD-SEGD-CVG including discharge gas flow rate, hydrogen flow rate, discharge gap, and pH of solution were also investigated, and the results are summarized in Figure 4. The effects of these parameters are discussed in detail in Section 3 of the Supporting Information. Generation Efficiency. Initially, the overall chemical vapor generation efficiencies of Cd and Zn by SD-SEGD-CVG was estimated via comparison of the resultant ICP-AES signal obtained from the HG and the SD-SEGD-CVG of a standard solution containing 100 μg L−1 of Cd and 500 μg L−1 Zn under their corresponding optimal conditions. According to earlier work,36 the HG of Zn and Cd was carried out using the following experimental conditions: 3.5% (m/v) KBH4 prepared in 0.1% (m/v) KOH as the reducing solution; 5 mg L−1 8hydroxyquinoline and 2 mg L−1 cobalt prepared in standard solution as the “enhancement’’ reagents; and 2.5% (v/v) HCl as the carrier solution. The results are presented in Figure 5 and show that the obtained signals (peak area) of Zn and Cd by the proposed method are much greater than those obtained by flow injection HG-ICP-AES, and this implies that the vapor generation efficiencies of Cd and Zn obtained using SDSEGD-CVG are much higher than those from flow injection
Figure 5. Comparisons of the signals from 100 μg L−1 of Cd(II) and Zn(II) obtained by flow injection HG-ICP-AES and SD-SEGD-CVGICP-AES. Experimental conditions: sample volume, 20 μL; Ar carrier gas flow rate, 500 mL min−1; and discharge voltage, 60 V.
HG. This is probably because the conventional HG operation consists of several stages, including formation, separation, and transport. Feng et al.37 reported that volatile species of Cd and other species are soluble and stable in water. Therefore, the generated species from flow injection HG may remain in the large volumes of carrier solution and reducing solution and cannot be efficiently separated from the liquid phase, thereby resulting in these low efficiencies. Additionally, the low signal obtained from flow injection HG also arises from the serious dilution of analyte by the carrier solution. The high CVG efficiencies of the proposed method with microliter volumes of sample make it more favorable to determine Cd and Zn in limited amounts of sample or even single cells. The exact vapor generation efficiency could be obtained from comparison of the relative mass of analyte in the feed and waste solution after SD-SEGD-CVG of standard solutions. For this purpose, the waste solution remaining from the 20 μL volume of standard solution containing 10 μg L−1 of Cd (0.2 ng) and 100 μg L−1 Zn (2 ng) was collected (about 15 μL) and analyzed by graphite furnace electrothermal atomic absorption spectrometry (GF-AAS). The mass of Cd and Zn in the waste solution are 0.02 ± 0.01 and 0.40 ± 0.12 ng, respectively. Thus, the overall efficiencies were 90 ± 5% and 80 ± 6% for Cd and Zn. These results further confirm that the proposed procedure is a very efficient CVG methodology for both cadmium and zinc. Interferences. With conventional HG, transition and noble metal ions are very easily reduced to their metallic state or to colloidal forms which then scavenge or decompose hydrides prior to gas−liquid separation and result in serious interferences. Guo el al.38 reported that the HG of Cd was significantly 12097
dx.doi.org/10.1021/ac502911p | Anal. Chem. 2014, 86, 12093−12099
Analytical Chemistry
Article
Table 1. Analytical Results of Cd and Zn in CRM GBW07601a and Two Human Hair Samples by Using SD-SEGD-CVG-AFS certified, mg kg−1
a
determineda, mg kg−1
added, mg kg−1
foundeda, mg kg−1
recovery, %
sample
Cd
Zn
Cd
Zn
Cd
Zn
Cd
Zn
Cd
Zn
GBW07601a Human hair A Human Hair B
0.11 ± 0.03
190 ± 9
0.100 ± 0.05 0.09 ± 0.01 0.11 ± 0.04
181 ± 10 158 ± 10 165 ± 7
0.10 0.25
30 75
0.20 ± 0.07 0.34 ± 0.05
186 ± 7 233 ± 8
105 92
92 90
Mean and standard deviation (n = 3).
suppressed by Cu2+ even at concentrations as low as 0.05 mg L−1. In this study, the effect of 11 concomitant ions, including Cu2+, Co2+, and Ni2+, on generation of volatile species of Zn and Cd was investigated and the results are summarized in Table S2 (see Section 4 of the Supporting Information). Despite earlier reports of serious interference arising from Cu2+, no interferences from Cu2+ were detected for either Cd2+ or Zn2+, even at concentrations as high as 10 mg L−1. It is worthwhile to note that no interferences from other tested elements were detected, including Co2+, which is a typical enhancement reagent in conventional HG of Zn and Cd. All these results demonstrate that the current technique has higher tolerable levels of coexisting transition and noble metal ions over conventional HG, EcHG, and DBD plasma CVG. However, the mechanism of this high anti-interference capability is unclear due to the complex nature of microplasma and needs further investigation. This is probably because the tested transition and noble metal ions cannot be reduced to their metallic state or colloidal forms, which scavenge or decompose hydrides and result in significant interferences in conventional HG. Analytical Figures of Merit. Table S3 (see Section 5 of the Supporting Information) summarizes analytical figures of merit achieved under the optimal experimental conditions when coupling the SD-SEGD-CVG to AFS and ICPMS and compares their performance to that of similar analytical methods. The linear correlation coefficients for calibration curves were better than 0.99 regardless of the detector. The limit of detection (LOD), which is defined as the analyte concentration equivalent to 3 standard deviations of 11 measurements of a blank solution, was better than 0.01 μg L−1 (0.2 pg) for Cd and 0.1 μg L−1 (2 pg) for Zn, respectively. The LODs of the proposed method are comparable or better than those obtained from EcHG, Photo-CVG, DBD plasma CVG, or conventional HG using an enhancement reagent. More importantly, the absolute LODs are comparable or better than the previously reported results since only a 20 μL sampling volume is used, which is of benefit to the determination of trace Cd and Zn analysis in limited amounts of sample. It should be noted that good analytical performance could still be obtained even with a sample volume of 5 μL. The precision, which is expressed as the relative standard deviation (RSDs) of six replicate measurements, was better than 3.0% for Zn and Cd at 50 and 5 μg L−1 respectively. Figure S2 (see Section 5 of the Supporting Information) shows the temporal profiles of repetitive flow injection SD-SEGD-CVG-AFS analyses and a typical calibration curve for Cd. Analysis of Limited Amounts of Sample. Single human hair was selected as a model sample to validate the potential of the proposed method on the analysis of limited amounts of sample with complex matrixes because human hair can be used for forensic means as well as an attractive indicator of environmental or occupational exposure.39 However, hair left at the scene of a crime is often only present in very small
amounts (single hairs). The results of two hair samples are summarized in Table 1 with good spike recoveries of analytes (92−105%), confirming accuracy of the proposed method for limited amounts of sample. The utility of the proposed technique was further demonstrated by the determination of Cd and Zn in a CRM (human hair powder, GBW07601a), with analytical results summarized in Table 1. The t test showed that the analytical results obtained by the proposed method were not significantly different from the certified values at the 95% level of confidence. Cells Analysis. Quantitative analysis of elements in individual biological cells facilitates more accurate evaluation of the effects of an element as opposed to the average concentration of tens of thousands of cells because heterogeneity of biological systems causes distribution gradients of metabolites among different cells.40 The feasibility of cell analysis was demonstrated by determination of Cd in single paramecium cells. Figure S3a in the Supporting Information shows the images of replicate injections (n = 10) of 20 μL of the diluted paramecium cell solution and indicate a drop contains 44 ± 2 paramecium cells. The steady-state signals obtained from analysis of this repeatedly injected solution using the SD-SEGD-CVG-AFS are summarized in Figure S3b in the Supporting Information, providing a precision of 5.5% RSD. A single paramecium contained 1.5 ± 0.2 pg Cd(II) and higher than the average concentration obtained by ICPMS (about 1.0 pg). These results indicate the proposed method has great potential for elemental analysis at the single cell level because of its high sensitivity, low blank, and small amounts of sample needed.
■
CONCLUSION
A novel plasma assisted-chemical vapor generation approach based on single drop solution electrode glow discharge was developed to couple with AFS, ICP-AES and ICPMS for ultrasensitive determination of Cd and Zn. Both Cd (II) and Zn(II) can be readily reduced to their volatile species by this SD-SEGD-CVG system only using hydrogen. This approach not only retains the advantages provided by conventional CVG methods but further offers several unique advantages including lower sample and power consumption (
