Letter pubs.acs.org/ac
Nonthermal Optical Emission Spectrometry: Direct Atomization and Excitation of Cadmium for Highly Sensitive Determination Yi Cai,† Ya-Jie Zhang,† De-Fu Wu,‡ Yong-Liang Yu,*,† and Jian-Hua Wang*,† †
Research Center for Analytical Sciences, College of Sciences, Northeastern University, Box 332, Shenyang, Liaoning 110819, China Water Quality Technology Center, Qingdao Jiaming Measurement and Control Technology Co., LTD, Qingdao, Shandong 266000, China
‡
S Supporting Information *
ABSTRACT: The low atomization and excitation capability of nonthermal microplasma, e.g., dielectric barrier discharge (DBD), has greatly hampered its potential applications for the determination of metals in solution. In the present work, an inspiring development is reported for direct atomization and excitation of cadmium in aqueous solution by DBD and facilitates highly sensitive determination. A DBD microplasma is generated on the nozzle of a pneumatic micronebulizer to focus the DBD energy on a confined space and atomize/excite metals in the spray. Meanwhile, an appropriate sample matrix and nebulization in helium further improves the atomization and excitation capability of DBD. With cadmium as a model, its emission is recorded by a CCD spectrometer at 228.8 nm. By using an 80 μL sample solution nebulized at 3 μL s−1, a linear range of 5−1000 μg L−1 along with a detection limit of 1.5 μg L−1 is achieved, which is comparable to those obtained by commercial bulky inductively coupled plasma (ICP)-based instrumentations.
O
ptical emission spectrometry (OES) with inductively coupled plasma (ICP) as excitation source is extensively applied for trace metal analysis.1,2 However, the high temperature of the ICP source makes it impossible to develop miniaturized analytical devices for field analysis. As a typical nonthermal microplasma, dielectric barrier discharge (DBD) had been reported to serve as an excitation source for a miniature OES system for trace mercury determination.3 However, DBD microplasma has been mainly used for the excitation of gaseous species4,5 produced by chemical vapor generation6−11 or electrothermal vaporization.12 The direct atomization and excitation of metals in aqueous solution by DBD remains unpractical due to its very weak atomization and excitation capability, and the detection sensitivity for metals by OES is far away from satisfactory.13−15 In this respect, a novel DBD microplasma integrated on a pneumatic micronebulizer is developed for the first time for direct atomization and excitation of metals in aqueous solution and facilitating highly sensitive determination, with a sensitivity parallel to those achieved by a bulky commercial ICP-OES system. The miniature DBD-OES system consists of a nebulization unit, a DBD microplasma excitation source, and a charge coupled device (CCD) spectrometer (AvaSpec-ULS2048-4USB2, Avantes, Netherlands) as given in Figure 1. A piece of aluminum foil as an electrode is used to surround the outside of the nozzle of a pneumatic micronebulizer (LTA-1, Agilent/ Varian), while a tungsten plate as counter electrode is attached onto the outer surface of a quartz plate. A discharging gap © 2016 American Chemical Society
Figure 1. Setup of the miniature nonthermal DBD-OES system incorporating the photographs of DBD microplasma under light and in the dark.
separated by the quartz plate is set at a distance of 2 mm. A helium stream at a flow rate of 600 mL min−1 is introduced into the micronebulizer for solution nebulization and microplasma generation. Atmospheric-pressure DBD microplasma is directly generated on the nozzle of the micronebulizer, when applying a high-frequency and high-voltage electric field from an ENT106B neon power supply (Guangzhou Xinxing Neon Light Received: March 2, 2016 Accepted: March 31, 2016 Published: March 31, 2016 4192
DOI: 10.1021/acs.analchem.6b00830 Anal. Chem. 2016, 88, 4192−4195
Letter
Analytical Chemistry
low atomization efficiency is among the main reason for causing very low detection sensitivity.13−15 In the present case, the aqueous sample solution is nebulized into fine droplets for enhancing its surface area, and meanwhile, the DBD microplasma is set to pass through the aerosol/spray for performing effective excitation. The effective contact between aqueous solution and microplasma as a key factor is adjusted by nebulizing the solution at various sampling flow rates (Figure 3). It is observed that two maxima of emission intensity of
Supply, China) at a sinusoidal discharging voltage of 2.9 kV with a frequency of ca. 40 kHz. The sample solution is introduced into the micronebulizer by a microsequential injection system at a flow rate of 3 μL s−1, which is nebulized at the nozzle of the micronebulizer and in situ atomized/excited by DBD microplasma. The emission spectra are recorded by the CCD spectrometer combined with a fiber-optic probe (950 μm core i.d. and 20 cm length) furnished with a 74-UV collimating lens (Ocean Optics, Dunedin, FL). The CCD spectrometer with a 25 μm slit and an 1800 lines/mm grating offers a spectral resolution of ca. 0.2 nm. An integration time of 300 ms and an average of 5 scans were employed. The emission spectral characteristics of the DBD-OES system are evaluated by the direct atomization and excitation of cadmium in solution under a helium atmosphere (Figure 2).
Figure 3. Dependence of cadmium optical emission at 228.8 nm on the sampling flow rate, incorporating the photographs of nebulization effect at sampling flow rates of 3, 30, and 70 μL s−1. 80 μL of 200 μg L−1 Cd2+ in 2% (v/v) ethanol and 0.6 g L−1 KCl is used.
cadmium are observed at either a lower or a higher sampling flow rate of 3 or 70 μL s−1, respectively. The total surface area of the fine droplets in the spray is dependent on two factors, i.e., nebulization efficiency and sampling quantity. A higher nebulization efficiency is provided when the sampling flow rate falls within 0.5−3 μL s−1, resulting in the first maximum emission intensity. On the other hand, a larger sampling quantity is given within the flow rate range of 30−70 μL s−1, corresponding to the second maximum. Although the first maximum is only ca. 65% of the second one, a favorable stability for the DBD microplasma is obtained at a lower sampling flow rate, giving rise to better precision for cadmium detection. Therefore, a sampling flow rate of 3 μL s−1 is employed for the ensuing study. We further discovered that the matrix component in sample solution is a critical factor for the improvement of detection sensitivity of cadmium (Figure 4). Some soluble organics, e.g., methanol, ethanol, glycerol, and n-butanol, serve as sensitizers and enhance the cadmium optical emission. On the contrary, however, some other soluble organics, e.g., acetone and formic acid, significantly suppress cadmium optical emission. A certain amount of alcohol in the sample solution would be decomposed to produce hydrogen in the microplasma,19−21 which highly promotes the production of ·H.17,18 The presence of a large amount of ·H would be highly effective in the promotion of cadmium atomization. It is seen that the optical emission intensity of cadmium is significantly enhanced in aqueous solution containing alcohol with respect to that in pure water (Figure S1). Therefore, 2% (v/v) ethanol is added into the sample solution to serve as sensitizer for the improvement of cadmium optical emission. The solution conductivity is another key factor playing an important role in the enhancement of detection sensitivity of
Figure 2. Optical emission spectra of helium DBD microplasma: without introduction of sample solution (A); by introducing 2% (v/v) ethanol and 0.6 g L−1 KCl as blank solution at flow rates of 0.5 μL s−1 (B) or 3 μL s−1 (C); by introducing 300 μg L−1 Cd2+ in 2% (v/v) ethanol and 0.6 g L−1 KCl at a flow rate of 3 μL s−1 (D).
In the absence of liquid phase, a series of nitrogen emission lines are observed in the background emission spectra of helium DBD, due to the diffusion of nitrogen from ambient air into the open DBD microplasma.4,5 When a certain amount of aqueous solution is nebulized into the microplasma, these ambient nitrogen emission lines are significantly weakened and even disappear at a liquid flow rate of ≥3 μL s−1. In this case, the excitation energy of DBD microplasma is focused on the species in the fine droplets. The high-energy electrons generated by DBD collide with the ambient aerosol to produce various active species, e.g., ·H, ·OH, H2O2, and O3.16 Particularly, the presence of ·H highly promotes the atomization of cadmium and its excitation in the microplasma.17,18 As we have observed, the emission line of cadmium in the fine droplets at 228.8 nm is clearly and sensitively isolated from the background emission spectra facilitating its quantification. The background emission at 229.1 nm is chosen for the correction of fluctuations of the analytical emission line. The detection sensitivity of cadmium in aqueous solution is highly dependent on the atomization efficiency of cadmium, which is closely related to the effective contact between the aqueous solution and the microplasma. The probabilities of ·H produced and its reaction with cadmium would be extremely enhanced at a larger contact area,17,18 resulting in improvement on the atomization efficiency. In previous attempts for direct excitation of metals in aqueous solution, the insufficient or too 4193
DOI: 10.1021/acs.analchem.6b00830 Anal. Chem. 2016, 88, 4192−4195
Letter
Analytical Chemistry
between cadmium and ·H.17 With the increase of pH value, ·H would promote redox reaction and cadmium atomization, thereby improving the detection sensitivity of cadmium. On the other hand, however, it is required to prevent cadmium from hydrolysis, which tends to cause a decrease of cadmium atomization efficiency. Generally, a better detection sensitivity of cadmium is obtained in a weak acidic or neutral aqueous medium. A few important parameters governing the performance of the entire system, e.g., discharging voltage, discharging gap, carrier gas, and its flow rate, were scrutinized (Figures S2 and S3). The generation of stable DBD microplasma is crucial to serve as excitation source for quantitative analysis. It is found that no helium DBD microplasma is generated at a discharging voltage of
