Mar 15, 2016 - Eastern Regional Research Center, Agricultural Research Service, U.S. Department of Agriculture, 600 East Mermaid Lane,. Wyndmoor ...
Article pubs.acs.org/ac
Ambient-Temperature Trap/Release of Arsenic by Dielectric Barrier Discharge and Its Application to Ultratrace Arsenic Determination in Surface Water Followed by Atomic Fluorescence Spectrometry Xuefei Mao,†,‡ Yuehan Qi,†,‡ Junwei Huang,§ Jixin Liu,*,†,§ Guoying Chen,∥ Xing Na,† Min Wang,*,† and Yongzhong Qian† †
Institute of Quality Standard and Testing Technology for Agro-products, Chinese Academy of Agricultural Sciences, and Key Laboratory of Agro-food Safety and Quality, Ministry of Agriculture, Beijing 100081, China § Beijing Titan Instruments Company, Limited, Beijing 100015, China ∥ Eastern Regional Research Center, Agricultural Research Service, U.S. Department of Agriculture, 600 East Mermaid Lane, Wyndmoor, Pennsylvania 19038, United States S Supporting Information *
ABSTRACT: A novel dielectric barrier discharge reactor (DBDR) was utilized to trap/release arsenic coupled to hydride generation atomic fluorescence spectrometry (HGAFS). On the DBD principle, the precise and accurate control of trap/release procedures was fulfilled at ambient temperature, and an analytical method was established for ultratrace arsenic in real samples. Moreover, the effects of voltage, oxygen, hydrogen, and water vapor on trapping and releasing arsenic by DBDR were investigated. For trapping, arsenic could be completely trapped in DBDR at 40 mL/min of O2 input mixed with 600 mL/min Ar carrier gas and 9.2 kV discharge potential; prior to release, the Ar carrier gas input should be changed from the upstream gas liquid separator (GLS) to the downstream GLS and kept for 180 s to eliminate possible water vapor interference; for arsenic release, O2 was replaced by 200 mL/min H2 and discharge potential was adjusted to 9.5 kV. Under optimized conditions, arsenic could be detected as low as 1.0 ng/L with an 8-fold enrichment factor; the linearity of calibration reached R2 > 0.995 in the 0.05 μg/L−5 μg/L range. The mean spiked recoveries for tap, river, lake, and seawater samples were 98% to 103%; and the measured values of the CRMs including GSB-Z50004-200431, GBW08605, and GBW(E)080390 were in good agreement with the certified values. These findings proved the feasibility of DBDR as an arsenic preconcentration tool for atomic spectrometric instrumentation and arsenic recycling in industrial waste gas discharge.
A
procedure could be carried out under ambient temperature, its applicability to trace-level arsenic determination would be greatly enhanced. Dielectric barrier discharge (DBD) has been well adopted for widespread applications,16 due to its simplicity, low cost, low energy consumption, etc. DBD is usually implemented as an ionization source for mass spectrometry (MS),17,18 as a tool for chemical vapor generation (CVG),19,20 or as an excitation source for atomic emission spectrometry (AES).21 Zhu et al.22−25 used it as an atomizer of hydrides or plasma-assisted chemical vapor generation (CVG) in atomic absorption spectrometry (AAS) or atomic fluorescence spectrometry (AFS). On the other hand, Kratzer et al.26 found that some memory effects occurred to DBD as an atomizer for bismuthane in the presence of oxygen admixed with argon discharge gas. Moreover, their recent work27 demonstrated that
rsenic is a well-known toxic element; carcinogenicity of inorganic arsenic is especially harmful to both human and animals.1 Contaminated drinking water has been considered one of the primary sources for human exposure to As.2 Ultratrace arsenic level in surface water renders it crucial to develop a highly sensitive monitoring method for water samples. A number of preconcentration methods, such as liquid−liquid extraction,3−5 solid phase extraction,6−8 and coprecipitation,9 have been coupled to spectrometric instrumentation to enhance sensitivity. However, applicability of these procedures is limited by time-consuming handling, unsatisfactory enrichment factors, and usage of large volumes of organic solvents leading to secondary waste. Gas phase enrichment (GPE)10 is an interesting approach with proven efficiency, simplicity, and low cost. Among GPE methods, in situ trapping and releasing of arsine on the surface of a graphite furnace11,12 or quartz tube13−15 could enrich arsenic by more than 10-fold. However, the necessity of high temperature in trap/release procedures usually complicates device and process design and limits its applicability. Hence, if the trap/release © 2016 American Chemical Society
Received: February 4, 2016 Accepted: March 15, 2016 Published: March 15, 2016 4147
DOI: 10.1021/acs.analchem.6b00506 Anal. Chem. 2016, 88, 4147−4152
Article
Analytical Chemistry
manufacturer’s recommendations. A flow injection system was assembled in this work comprising a peristaltic pump (PP), a reaction coil (C), and a gas−liquid separator (GLS). The setup of the HG-DBD-AFS system is presented in Figure 1, which was mainly composed of a HG system, gas line
the DBD atomizer could be employed to trap bismuth on the surface of quartz in the presence of high level selenium, arsenic, and antimony hydrides. Compared with those GPE methods using the heated graphite furnace11,12 or quartz tube,13−15 the DBD trapping device could consume less energy due to no electrical heating process, such as only 200 mL/min flow rate, due to the possible fluorescence depression by the excess H2. This result indicated an acceleration role of H2 on the release of arsenic, in line with Kratzer’s findings.27 Thus, 200 mL/min of H2 was chosen as the optimal flow rate, achieving the smallest RSD value (approximately 3%). On the other hand, owing to the consumption of H radicals by O2, O2 should be eliminated. As shown in Figure S-4, released arsenic from DBD decreased with increase of O2. The effect of discharge potential on arsenic release is shown in Figure 5B. It demonstrated that no arsenic was released from DBD at 9.8 kV, the arsenic signals began to decline, possibly caused by unknown side reactions at very strong discharge in H2−Ar atmosphere. In this work, ∼9.5 kV was therefore chosen as the applicable discharge potential. Analytical Performance and Interference. Under the optimized conditions, the analytical figures of merit were evaluated. The linearity of the calibration curve was investigated by measuring a series of standard solutions ranging from 0.05 μg/L to 5 μg/L, and the linear regression coefficient (R2) was >0.995. The limit of detection (LOD) of arsenic for this method was 1.0 ng/L (introduction volume = 20 mL), calculated by taking 3 times the standard deviation of the blank solution divided by the slope from 11 measurements. The RSD of the 0.5 μg/L arsenic standard solution from 11 measurements was 180 s. It could be deduced that a long argon sweep time enables removal of the interfering agent from GLS during the trap step. This interfering agent was very likely water
Figure 4. The AF intensity of released arsenic by different sweep times after switching Ar to downstream GLS. Here, O2 is 40 mL/min for trap, while no extra O2 is used for release; no extra H2 is used for trap, while H2 is 200 mL/min for release; the discharge potential is 9.2 kV for trap and 9.5 for release, respectively. The AF intensity of arsenic by 240 s sweep is set as 100; other results are normalized to this value. 4150
DOI: 10.1021/acs.analchem.6b00506 Anal. Chem. 2016, 88, 4147−4152
Article
Analytical Chemistry
Figure 5. The AF intensity of released arsenic at different H2 flow rates/discharge potentials during release procedure. The optimized trap conditions are adopted. For panel A, the discharge potential is 9.5 kV; no extra O2 is used. The AF intensity of arsenic at 200 mL/min H2 is set as 100; other results are normalized to this value. For panel B, H2 is 200 mL/min; no extra O2. The AF intensity of arsenic at 9.8 kV is set as 100.
Table 2. Spiked Recoveries and Measured Values of Arsenic in Real Water Samples samples tap water river water lake water sea water GSB-Z50004200431b GBW08605c GBW(E)080390d
measda (μg/ L)
added (μg/L)
0.22 ± 0.02 4.2 ± 0.2 8.3 ± 0.1 1.53 ± 0.04 58.5 ± 0.4
10.0 10.0 10.0 10.0
founda (μg/ L)
recovery (%)
± ± ± ±
101 103 98 102
10.1 14.5 18.1 11.7
0.7 0.5 0.9 0.3
499 ± 4 493 ± 19
a Mean value and standard deviation (n = 3). bCertified value: 60.6 ± 4.2 μg/L. cCertified value: 0.500 ± 0.008 μg/g. dCertified value: 0.50 ± 0.02 mg/L.
Figure 6. The AF signals of arsenic without or with DBDR. Legend A, 20 mL of 0.5 μg/L arsenic solution was introduced into the HG system without DBDR. Legend B, 20 mL of 0.5 μg/L arsenic solution was introduced into the HG system, trapped for 200 s, and then released at the optimized conditions.
established for ultratrace arsenic in real samples based on DBD principle. In comparison of other reported atomic spectrometric methods, ultratrace arsenic in surface water could be effectively enriched at a higher sample volume. Meanwhile, lower energy consumption and no elemental residue caused by the temperature gradient are also superior to the in situ trapping devices reported previously. Detailed trapping and releasing mechanisms are not yet fully clarified; a thorough investigation must be performed in the future. In a word, these findings laid the foundation for the future development of DBD as a trap tool in atomic spectrometric instrumentation to improve sensitivity. Arsenic trap/release by DBD has great potential to recycle arsenic from contaminated industrial waste gas, thereby protecting human health and the environment.
we investigated some possible interference from HG-forming elements, such as Se, Sb, Bi, Pb, and Hg. No significant interference was found at
