Anal. Chem. 2007, 79, 7176-7181
Technical Notes
Indirect Determination of Sulfide at Ultratrace Levels in Natural Waters by Flow Injection On-Line Sorption in a Knotted Reactor Coupled with Hydride Generation Atomic Fluorescence Spectrometry Yan Jin, Hong Wu, Ye Tian, Luhong Chen, Jiongjia Cheng, and Shuping Bi*
School of Chemistry and Chemical Engineering, State Key Laboratory of Coordination Chemistry and Key Laboratory of MOE for Life Science, Nanjing University, Nanjing 210093, China
A simple and sensitive nonchromatographic approach for indirect determination of sulfide at ultratrace levels in natural waters based on its selective precipitation with Hg2+ on the inner wall of a knotted reactor (KR) was developed for flow injection on-line sorption coupled with hydride generation atomic fluorescence spectrometry (HG-AFS). With the Hg2+ pH kept at 2.0, the HgS precipitation was formed in the KR after a reaction time of 120 s. A 10% (v/v) HCl was introduced to elute the remnant inorganic mercury and to merge with the KBH4 solution (0.05% m/v) for HG-AFS detection. Under the optimal experimental conditions, the sample throughputs were 20 h-1. The detection limit was found to be 0.05 µg L-1, and the relative standard deviation (RSD, n ) 11) for determination of 2.0 µg L-1 sulfide was 3.3%. The developed method was successfully applied to the determination of sulfide in a variety of natural water samples and wastewater samples with the gas-phase separation and sorption apparatus. Because its toxicity and causticity increase awfully when protonated, sulfide has gained sufficient attention from the biological and industrial points of view.1-3 Thus, it is important to develop a rapid and sensitive method for immediate sulfide monitoring in natural waters. Analytical methods for sulfide determination include spectroscopy, chromatography, and electrochemical detection. The methylene blue (MB) method plays a great role in spectroscopically determining sulfide, and the performance is improved when combined with flow injection4 such as flow-injection solid-phase * Corresponding author. Phone: 86-25-86205840. Fax: 86-25-83317761. E-mail:
[email protected]. (1) Howard, E. C.; Henriksen, J. R.; Buchan, A.; Reisch, C. R.; Burgmann, H.; Welsh, R.; Ye, W. Y.; Gonzalez, J. M.; Mace, K.; Joye, S. B.; Kiene, R. P.; Whitman, W. B.; Moran, M. A. Science 2006, 314, 649-652. (2) Meinrat, O. A. Science 2007, 315, 50-51. (3) Pierce, D. T.; Applebee, M. S.; Lacher, C.; Bessie, J. Environ. Sci. Technol. 1998, 32, 1734-1737.
7176 Analytical Chemistry, Vol. 79, No. 18, September 15, 2007
spectrophotometry,4e when a detection limit was achieved as low as 1.7 µg L-1. Some other methods are based on the measurement of generated hydrogen sulfide to quantify sulfide and gain high sensitivity when measured by sulfur-specific flame photometric detection5 with a detection limit of 64 µg L-1. Direct spectroscopic detection such as ultraviolet spectrophotometery6 also obtained outstanding analytical performance for sulfide determination. Chromatography was employed for sulfide speciation. Some researches reported sulfide speciation in natural waters taken by ion chromatography (IC)7,8 as well as estuarine water9 and seawaters10 measured by high-performance liquid chromatography (HPLC), which could detect sulfide as low as 0.067 µg L-1.9 Electrochemical detection also offers more choices for sulfide measurement,11 and the detection limit can even reach 0.016 µg L-1.11h (4) (a) Ferrer, L.; Estela, J. M.; Cerda, V. Anal. Chim. Acta 2006, 573, 391398. (b) Ferrer, L.; de Armas, G.; Miro, M.; Estela, J. M.; Cerda, V. Analyst 2005, 130, 644-651. (c) Ferrer, L.; de Armas, G.; Miro, M.; Estela, J. M.; Cerda, V. Talanta 2004, 64, 1119-1126. (d) De Armas, G.; Ferrer, L.; Miro, M.; Estela, J. M.; Cerda, V. Anal. Chim. Acta 2004, 524, 89-96. (e) Cassella, R. J.; Teixeira, L. S. G.; Garrigues, S.; Costa, A. C. S.; Santelli, R. E.; de la Guardia, M. Analyst 2000, 125, 1835-1838. (f) Singh, V.; Gosain, S.; Mishra, S.; Jain, A.; Verma, K. K. Analyst 2000, 125, 1185-1188. (5) Howard, A. G.; Yeh, C. Y. Anal. Chem. 1998, 70, 4868-4872. (6) Guenther, E. A.; Johnson, K. S.; Coale, K. H. Anal. Chem. 2001, 73, 34813487. (7) Miura, Y.; Fukasawa, K.; Koh, T. J. Chromatogr., A 1998, 804, 143-150. (8) Montegrossi, G.; Tassi, F.; Vaselli, O.; Bidini, E.; Minissale, A. Appl. Geochem. 2006, 21, 849-857. (9) Tang, D. G.; Santschi, P. H. J. Chromatogr., A 2000, 883, 305-309. (10) Gru, C.; Sarradin, P. M.; Legoff, H.; Narcon, S.; Caprais, J. C.; Lallier, F. H. Analyst 1998, 123, 1289-1293. (11) (a) Salimi, A.; Roushani, M.; Hallaj, R. Electrochim. Acta 2006, 51, 19521959. (b) Zen, J. M.; Chang, J. L.; Chen, P. Y.; Ohara, R.; Pan, K. C. Electroanal. 2005, 17, 739-743. (c) Hassan, S. S. M.; Marzouk, S. A. M.; Sayour, H. E. M. Anal. Chim. Acta 2002, 466, 47-55. (d) Prodromidis, M. I.; Veltsistas, P. G.; Karayannis, M. I. Anal. Chem. 2000, 72, 39954002. (e) Muller, B.; Stierli, R. Anal. Chim. Acta 1999, 401, 257-264. (f) Giuriati, C.; Cavalli, S.; Gorni, A.; Badocco, D.; Pastore, P. J. Chromatogr., A 2004, 1023, 105-112. (g) Giovanelli, D.; Lawrence, N. S.; Jiang, L.; Jones, T. G. J.; Compton, R. G. Analyst 2003, 128, 173-177. (h) Alfarawati, R.; Vandenberg, C. M. G. Mar. Chem. 1997, 57, 277-286. 10.1021/ac070699s CCC: $37.00
© 2007 American Chemical Society Published on Web 08/18/2007
However, sulfide concentration of some natural waters is even below 0.5 µg L-1,4e and methods above cannot guarantee the sensitivity. In addition, the necessary instrumentation to apply these techniques of sulfide determination is not always available in all analytical laboratories and the cost is an obstacle. In recent years, hydride generation atomic fluorescence spectrometry (HGAFS) has received special attention due to its benefits of lower instrumental cost, higher sensitivity, shorter analysis time, easy operation, and less interferences. Because of low cost, nearly unlimited lifetime, and ease of construction, the open tube knotted reactor (KR) made from PTFE tubing was considered as not only the ideal collector in FI on-line preconcentration12 techniques but also the filter13 used for on-line elimination of interferences. Since an indirect method with high sensitivity in determinating Hg2+ by HG-AFS has been developed based on the HgS precipitation14,15 for immediate sulfide measurement at trace levels in natural waters, we will develop a novel indirect method by coupling FIAKR with HG-AFS to realize the ultratrace determination of sulfide in natural waters. EXPERIMENTAL SECTION Apparatus. A nondispersive atomic fluorescence spectrometer, model AF-610A (Beijing Rayleigh Analytical Instrument Co., Beijing, China), was used and controlled by a computer. A highintensity mercury hollow cathode lamp (Beijing Tian-gong Analytical Instrumental Factory, Beijing, China) was used as the radiation source at 253.7 nm. The gas-liquid separator made of glass is of a three-stage separation design. An argon flow was used to sweep the separated mercury vapor from the gas-liquid separator to the quartz tube atomizer. The operating parameters of the AFS instrument are selected according to the manual16 as follows: negative high voltage of photomultiplier, 280 V; lamp primary current, 40 mA; lamp boost current, 0 mA; flow rate of carrier gas (Ar), 600 mL min-1; atomizer temperature, 200 °C; atomizer height, 7 mm; signal recording mode, peak area. The HG-AFS was coupled with a model FIA-3100 flow injection system (Vital Instrumentals Co. Ltd., Beijing, China), equipped with two peristaltic pumps and a standard rotary injection valve (8-channel 16-port multifunctional injector). The rotation speed of the two peristaltic pumps, their stop and run intervals, and the actuation of the injection valve were all programmed. A 722s UV-vis spectrophotometer (Shanghai Precision Instruments Co. Ltd., China) was used for MB method validation. Reagents. All chemicals were at least of analytical grade and were purchased from Shanghai Chemicals Co. (Shanghai, China) (12) (a) Fang, Z. L.; Sperling, M.; Welz, B. J. Anal. At. Spectrom. 1991, 6, 301306. (b) Tao, G. H.; Hansen, E. H. Analyst 1994, 119, 333-337. (c) Yan, X. P.; Kerrich, R.; Hendry, M. J. Anal. Chem. 1998, 70, 4736-4742. (d) Lara, R. F.; Wuilloud, R. G.; Salonia, J. A.; Olsina, R. A.; Martinez, L. D. Fresenius J. Anal. Chem. 2001, 371, 989-993. (e) Yan, X. P.; Adams, F. J. Anal. At. Spectrom. 1997, 12, 459-464. (f) Wu, H.; Jin, Y.; Han, W. Y.; Miao, Q.; Bi, S. P. Spectrochim. Acta, Part B 2006, 61, 831-840. (g) Wu, H.; Jin, Y.; Shi, Y. Q.; Bi, S. P. Talanta 2007, 71, 1762-1768. (h) Wu, H.; Jin, Y.; Luo, M. B.; Bi, S. P. Anal. Sci. 2007, in press. (13) Burguera, J. L.; Burguera, M.; Rondon, C. E. Anal. Chim. Acta 1998, 366, 295-303. (14) Chen, D. F.; Wang, C. L.; Li, A. Q.; Bi, S. P. Chin. J. Anal. Chem. 1985, 13, 119-122. (15) Afkhami, A.; Khalafi, L. Microchim. Acta 2005, 150, 43-46. (16) Beijing Rayleigh Analytical Instrument Co. AF-610A Atomic Fluorescence Spectrometry Operation Manual; Beijing Rayleigh Analytical Instrument Co.: Beijing, China, 1997.
Figure 1. The schematic modified gas-phase separation and sorption apparatus.19
unless otherwise stated. Doubly quartz deionized water (DDW) was used. The stock standard solution of inorganic mercury (1000 mg L-1) was supplied by the National Research Center for Standard Materials (NRCSM, Beijing, China). The stock standard solution of sulfide (1000 mg L-1) was prepared daily by dissolving the appropriate amount of crystal Na2S‚9H2O and diluting it to volume with DDW.17 Solutions of Hg2+ were prepared daily by stepwise dilution of the 10 mg L-1 (as Hg) intermediate standard solutions, prepared weekly, with the pH kept at 2.0 by acidification with 6 mol L-1 HCl. For the MB method,18 0.2% (m/v) N,Ndimethyl-p-phenylenediaminehydrochloride and 12.5% (m/v) NH4Fe(SO4)2 were prepared by dissolving the proper quantity in 20% (v/v) and 2.5% (v/v) H2SO4, respectively. The 0.05% (m/v) KBH4 solution for Hg2+ determination was prepared daily by dissolving the appropriate amount of KBH4 in 0.2% (m/v) NaOH solution. Sample Pretreatment. Three mineral water samples were purchased from market. The river water, lake water, well water, and tap water were collected locally. Immediately after sampling, all the water samples were filtered through a 0.45 µm membrane.12e The 25 mL filtered samples were diluted to 100 mL with DDW and then determined at once. The synthetic wastewater samples17 was devised to contain (mg L-1 in parentheses) phenol (500), CH3COONa (500), NaCl (500), KC1 (500), CaCl2 (500), KSCN (500), Na2CO3 (500), and (NH4)2SO4 (150), in addition to sulfide with the concentration of 1.00, 2.00, and 5.00 mg L-1. The synthetic wastewater samples were treated with the modified gas-phase separation and sorption apparatus15,19 as follows shown in Figure 1: To the 150 mL flask containing 50 mL synthetic wastewater was added 12 mL of 6 mol L-1 HCl slowly and N2 was imported for about 30 min. At this moment, the output was absorbed by 25 mL of 0.5 mol L-1 NaOH in the tube and then the solution was kept at a volume of 50 mL. The final solution was diluted appropriately before determination. The previous report15 showed that all sulfide in the water samples could be extracted by this method. Procedure. The operation program of the FI coupled HGAFS (Figure 2) on-line indirect determination of sulfide is illustrated in Table 1 and runs through two primary steps, including background and sample determination. Step I, background determination, injecting DDW and Hg2+, (a) prefilling (Figure 2a): the injector valve was in the injection (17) Kuban, V.; Dasgupta, P. K.; Marx, J. N. Anal. Chem. 1992, 64, 36-43. (18) Qiu, W. J.; Li, G. G.; Chen T. Phys. Test. Chem. Anal., Part B 1995, 31, 49-50. (19) Xue, X. H.; Liu, C. H.; Huang, H. The Administration and Technique of Environmental Monitoring 2002, 14, 30-31.
Analytical Chemistry, Vol. 79, No. 18, September 15, 2007
7177
RESULTS AND DISCUSSION The Scheme of the System. Considering the system of FIKR coupled with HG-AFS for the indirect determination of sulfide, the principle was shown by eqs 1-5 as following. In step I, the signal IFHg2+ of the background determination when injecting DDW and Hg2+ can be expressed by eqs 1 and 2:
Figure 2. FI manifold for on-line sorption coupled with HGAFS.12c,12e,12f (a) Prefill/reaction/washing step, (b) sampling step. P1, P2, peristaltic pump; W, waste; KR, knotted reactor (150 cm long × 0.5 mm i.d. PTFE tubing); V, injector valve; GLS, gas-liquid separator; AFS, atomic fluorescence spectrometry.
position, so that the DDW and Hg2+ were pumped to fill the PTFE tubing before entering the injection valve with pump 1 off and pump 2 on. This prefilling stage was used only when a new sample was introduced but omitted for replication. (b) Sampling (Figure 2b): the injection valve was in the fill position when injecting DDW and Hg2+ with pump 1 off and pump 2 on, ensuring that the KR was saturated with DDW and Hg2+. (c) Washing and determination (Figure 2a): with pump 1 on and pump 2 off, HCl (10% v/v) was imported to wash the Hg2+ in the KR and to merge with the KBH4 solution (0.05% m/v) just before entering the GLS to generate volatile mercury, which was swept into the AFS system for detection by an argon flow. In this step, a signal of background marked IFHg2+ was acquired. Step II, sample determination, injecting sample and Hg2+, (d) prefilling (Figure 2a): the injector valve was in the injection position, so that the sample and Hg2+ were pumped to fill the PTFE tubing before entering the injection valve with pump 1 off and pump 2 on. (e) Sampling (Figure 2b): the injection valve was in the fill position when injecting sample and Hg2+ with pump 1 off and pump 2 on, making sure that the KR was filled with sample and Hg2+. (f) Reaction of HgS precipitation (Figure 2a): this step guaranteed the reaction between Hg2+ and sulfide in samples for a period of time tR with pump 1 and pump 2 off. (g) Washing and determination (Figure 2a): after 120s, with pump 1 on and pump 2 off, HCl was imported to wash remnant unreacted Hg2+ in the KR, and a signal IF′Hg2+ was acquired. The ∆IF indicating the sulfide was concluded by subtracting IFHg2+ from IF′Hg2+, and another sample could be detected repeatedly by step II. In this way a linear regression equation “∆IF - C S/ 2-” was made. 7178
Analytical Chemistry, Vol. 79, No. 18, September 15, 2007
WHg2+ )
πr2LKRvHg2+ / C vsample + vHg2+ Hg2+
(1)
IFHg2+ )
Kπr2LKRvHg2+ / C vsample + vHg2+ Hg2+
(2)
/ -1) represents the concentration of Hg2+; W 2+ CHg Hg 2+ (µg L (µg) is the quantity of mercury to be determined; vHg2+, vsample (mL min-1) represent the Hg2+ and sample flow rate, respectively; K (µg-1) indicates the ratio coefficient between the mercury quantity and the signal in HG-AFS; r and LKR (cm) are the inner diameter and length of the KR, respectively. In step II, the signal IF′Hg2+ of the sample determination when injecting sample and Hg2+ can be expressed by eqs 3 and 4:
πr2LKR (C/ v - δCS/2-vsample) vsample + vHg2+ Hg2+ Hg2+
(3)
Kπr2LKR IF′Hg2+ ) (C/ v - δCS/2-vsample) vsample + vHg2+ Hg2+ Hg2+
(4)
W′Hg2+ )
W′Hg2+ (µg) is the quantity of mercury to be determined in step II, C S/2- (µg L-1) represents the concentration of sulfide in the samples, and δ indicates the kinetic calibration coefficient of the reaction. ∆IF is concluded by eq 5, and a linear regression equation “ ∆IF - C S/2-” can be gained in accordance with the results of the experiments (Figure 3):
∆IF ) IFHg2+ - IF′Hg2+ )
Kδ πr2LKR C/ 1 + vHg2+/vsample S2-
(5)
Optimization of Experimental Parameters. Considering the equations, the main parameters considered to affect the system / 2+ contain the acidity of the reaction, tR, CHg 2+ , vsample/vHg , eluent HCl concentration, and KBH4 concentration. (a) The acidity of the reaction: The acidity of the reaction was controlled by adjusting the pH of Hg2+. The experiments (Figure 4a) showed that ∆IF reached the highest value when the pH of Hg2+ was 2.0. Since the reaction could not progress entirely under these pH conditions due to the reaction kinetic reasons, only a part of sulfide participated in the HgS precipitation. δ, explaining the kinetic principle of the reaction in the FI-KR-HG-AFS system, was 0.34 by summarizing a number of results (n ) 40). (b) tR: The reaction between Hg2+ and sulfide needs a period of time, so that tR was controlled to achieve the best efficiency. With Hg2+, 10.0 µg L-1 and S2- 2.0 µg L-1, the results (Figure 4b) showed that 120 s was / 2+ most proper value for tR. (c) CHg 2+: The concentration of Hg was investigated (Figure 4c). ∆IF increased but remained constant
Table 1. Operating Program of the FI On-Line Sorption System for AFS Sulfide Determination figures
valve position
pump active
a
2a
inject
P2
b
2b
fill
c
2a
inject
d
2a
inject
P2
e
2b
fill
P2
f
2a
inject
g
2a
inject
function
pumped medium
Step I. Background Determination prefill DDW 10 µg L-1 Hg2+ P2 sampling DDW 10 µg L-1 Hg2+ P1 washing 10 %(v/v) HCl 0.05 %(m/v) KBH4
P1
Step II. Sample Determination prefill sample 10 µg L-1 Hg2+ sampling sample 10 µg L-1 Hg2+ reaction eluting
/ -1; v -1 Figure 3. CHg sample ) 6.0 mL min ; vHg2+ ) 1.2 2+ ) 10.0 µg L -1 mL min ; tR ) 120 s; flow rate: HCl ) 10.7 mL min-1; KBH4 ) 8.0 mL min-1. Sampling time ) 10 s; LKR ) 150 cm. Other parameters are based on the operation manual.
/ as CHg 2+ increased, which indicated that reaction was progressing fully. Because too much Hg2+ may cause problems with / -1. (d) v complete elution, CHg sample/ 2+ was chosen as 10.0 µg L -1 vHg2+: With vHg2+ fixed at 1.2 mL min , the influence of sample flow rate on the signal intensity was studied (Figure 4d). When ∆IF reached the highest point, vsample/vHg2+ was 5.0. Thus, the sample flow rate of 6.0 mL min-1 was selected. (e) Eluent HCl concentration: The flow rate of the eluent was fixed at 10.7 mL min-1 as recommended by the instrument manufacturer. The results showed that 10% (v/v) HCl could provide both sufficiently strong elution capability to completely wash the remnant Hg2+ in the KR and required an acidic medium for chemical vapor generation. (f) KBH4 concentration: The influences of KBH4 concentration on the chemical vapor generation of mercury were investigated. The flow rate of KBH4 was fixed at 8.0 mL min-1 as recommended by the instrumental instruction. The maximum ∆IF was found when using KBH4 concentrations of 0.05% (m/v). Analytical Performance. Under the optimum conditions described above, the analytical characteristic data of the FI online KR sorption system for HG-AFS indirect determination of sulfide are summarized in Table 2. The developed method has high sensitivity and flexible linear range. By control of the content of Hg2+, the linear range shown
10(v/v) HCl 0.05 %(m/v) KBH4
time/s
flow rate/mL min-1
10
6.0 1.2 6.0 1.2 10.7 8.0
10 15
10 10
6.0 1.2 6.0 1.2
120 15
10.7 8.0
in Table 3 would be expanded at the expense of straight elution, corresponding to the samples with different requirement. Compared with other methods,2,4,5,10,11 as a rapid and simple alternative, this FIA-KR system can achieve better sensitivity than other approaches, which provided an effective means for immediate sulfide detection in natural waters at ultratrace levels. Interferences. The influences of potential coexisting ions on the signal of 2.0 µg L-1 sulfide were investigated. The tolerance ratios (Cion/Csulfide) when interference concentration is varying the analyte signal by 10% are presented as follows: CO32-, NO3-, Na+, and Cl- (5000); PO43- (3000); SCN- and K+ (2500); SO32- and I(1000); Br-, F-, NO2-, Mg2+, Al3+, Fe3+, Cu2+, Zn2+, and Ni2+ (500); SO42-, Co2+, Sn2+, Fe2+, S2O32-, CN-, and Ca2+ (250); Ag+ (75). Compared with other methods,20 this technology can eliminate the interferences such as S2O32-, I-, CN-, SCN-, PO43-, CO32-, and Al3+. Practical Application. The developed method was applied to the sulfide determination in synthetic wastewater samples with the gas-phase separation and sorption apparatus. Compared with the classical MB method, the results listed in Table 4 demonstrates the validity of the developed method. Several types of natural water samples: tap water, lake water, river water, mineral water and groundwater were determined by the proposed method. The analytical results were shown in Table 5, which showed that sulfide content in some natural water were at µg L-1 levels4e,9,20a and results of tap water were consistent with the sulfide levels in China21 (2.37 µg L-1). The recoveries of the spiked samples varied from 92 to 106% for sulfide. CONCLUSIONS In this work, a flow injection on-line sorption system coupled with hydride generation atomic fluorescence spectrometry using a knotted reactor for indirect fast determination of sulfide was (20) (a) Ghasemi, J.; Mohammadi, D. E. Microchem. J. 2002, 71, 1-8. (b) Safavi, A.; Ramezani, Z. Talanta 1997, 44, 1225-1230. (c) Safavi, A.; Moradlou, O.; Maesum, S. Talanta 2004, 62, 51-56. (d) Safavi, A.; Karimi, M. A. Talanta 2002, 57, 491-500. (e) Florou, A. B.; Prodromidis, M. I.; Karayannis, M. I.; Tzouwara-Karayanni, S. M. Talanta 2000, 52, 465472. (f) Singh, V.; Gosain, S.; Mishra, S.; Jain, A.; Verma, K. K. Analyst 2000, 125, 1185-1188. (21) Sun, D. M.; Ruan, D. W.; Wang, L. H. Chin. J. Anal. Chem. 2004, 32, 179182.
Analytical Chemistry, Vol. 79, No. 18, September 15, 2007
7179
Table 2. Analytical Performance of the FI On-Line Sorption System items
parameters
sampling time/s sample consumption/mL reagent consumption/mL 0.05% m/v KBH4 10% v/v HCl sampling frequency/h-1 linear range/µg L-1 regression equation (7 standards; ∆IF, signal intensity; CS/, µg L-1) correlation coefficient detection limit (3s)/ng L-1 precision (RSD, n ) 11)/%
10 1.0 4.0 5.3 20 0.1-2.5 ∆IF ) 12.5 + 97.5 CS/ 0.9986 50.0 3.3
Table 3. The Characteristics of Linear Regression Equations Hg2+ content µg L-1
regression equation (CS/2-, µg L-1)
5.0 10.0 20.0
∆IF ) 32.2C S/2- + 8.3 ∆IF ) 97.5C S/2- + 12.5 ∆IF ) 67.4C S/2- + 6.4
detection limits linear range µg L-1 µg L-1 0.18 0.05 0.10
0.2-2.0 0.1-2.5 0.2-5.0
Table 4. Analytical Results for Synthetic Wastewater Samples results (n ) 5) synthetic waste water
S2- content mg L-1
this work mg L-1
sample 1 sample 2 sample 3
1.00 2.00 5.00
0.97 ( 0.01 1.89 ( 0.03 4.90 ( 0.06
MB method18 mg L-1 0.95 ( 0.02 1.85 ( 0.04 4.88 ( 0.06
Table 5. Analytical Results for Natural Water Samples recovery verification (n ) 5)
Figure 4. Effect of important parameters on the ∆IF of 2.0 µg L-1 sulfide. All other conditions were shown in Figure 2. (a) The acidity / 2+ of system; (b) tR; (c) CHg 2+; and (d) vsample/vHg .
developed, which has the following prominent advantages: (1) Simplicity and low cost make this method very convenient for sulfide determination. FI on-line KR sorption is easy to be coupled to AFS due to the favorable characteristics of the KR such as low cost, no need for packing materials, low hydrodynamic impedance, and ease of operation. HgS produced by each sampling laid over the inner surface of KR and the thickness of the precipitation cover was 9.6 × 10-12 mm, neither adsorbing Hg2+ nor blocking the tube, leading to more than 400 times for practical use. (2) This method can provide inspiration for developing a series of novel methods for the indirect determination of environmental pollutants. The precipitations between metal and sulfide containing Ag2S 7180 Analytical Chemistry, Vol. 79, No. 18, September 15, 2007
sample
S2- content µg L-1
S2- added µg L-1
S2- found µg L-1
recovery %
ground water tap water mineral water 1 mineral water 2 mineral water 3 lake water river water
2.55 ( 0.01 3.68 ( 0.02 1.16 ( 0.01 0.37 ( 0.02 2.11 ( 0.01 1.40 ( 0.02 2.50 ( 0.01
0.50 0.50 0.50 0.50 0.50 0.50 0.50
0.48 ( 0.01 0.46 ( 0.02 0.53 ( 0.02 0.47 ( 0.03 0.53 ( 0.02 0.52 ( 0.02 0.45 ( 0.02
96 92 106 94 106 104 90
(Ksp ) 6.3 × 10-50), CuS (Ksp ) 6.3 × 10-36), SnS (Ksp ) 1.0 × 10-25), PbS (Ksp ) 8.0 × 10-28), CdS (Ksp ) 8.0 × 10-27), R-ZnS (Ksp ) 1.6 × 10-24), R-CoS (Ksp ) 4.0 × 10-21), and FeS (Ksp ) 6.3 × 10-18) can all be considered for indirect determination.22 The reason why we choose HgS is based on its extremely low Ksp (Ksp ) 4.0 × 10-53) to achieve high sensitivity in the measurement, which greatly extends the use of highly sensitive atomic spectroscopy. Some other methods can also be developed for the FIA-KR-HG-AFS indirect determination of anions such (22) Dean, J. A. Lang’s Handbook of Chemistry, 13th ed.; McGraw-Hill: New York, 1991.
as SCN-, I-, and Br- (when chelating with Hg2+, Hg(SCN)2, log β4 ) 21.23; HgI42-, log β4 ) 29.83; HgBr42-, log β4 ) 21.00)22 and some other more complex organic matrices like 1-(2-pyridylazo)2-naphthol (PAN) and 1-(2-thiazolylazo)-p-cresol (TAC). In addition, based on the high sensitivity of electrochemical determination, some simple and portable electrochemical sensors coupled with FIA-KR for monitoring environmental water samples can be devised. ACKNOWLEDGMENT This project is supported by the National Natural Science Foundation of China (Grant Nos. 20575025 and NFFTBS-
J0630425), Research Funding from the State Education Administration of China for the Ph.D. program (Grant 20050284030), Natural Science Foundation of Jiangsu Province (Grant BK 2005209). The Grant of Analytical Measurements of Nanjing University and assistance from Mr. Tao XC are appreciated. Dr. Wu Hong thanks the support from the Xuzhou Normal University.
Received for July 11, 2007.
review
April
10,
2007.
Accepted
AC070699S
Analytical Chemistry, Vol. 79, No. 18, September 15, 2007
7181
