Sensitive Monitoring of Humic Acid in Various Aquatic ... - J-Stage

1 downloads 0 Views 111KB Size Report
Oct 10, 2007 - Environments with Acidic Cerium Chemiluminescence Detection ... redox reaction between humic acid and cerium(IV) in the acidic condition.
ANALYTICAL SCIENCES OCTOBER 2007, VOL. 23 2007 © The Japan Society for Analytical Chemistry

1189

Sensitive Monitoring of Humic Acid in Various Aquatic Environments with Acidic Cerium Chemiluminescence Detection Xianglei CHENG,*,** Lixia ZHAO,* Xu WANG,* and Jin-Ming LIN*† *State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, P. O. Box 2871, Beijing 100085, China **School of Public Health Science, Nanchang University, Nanchang 330006, P. R. China

In this research, a simple, sensitive chemiluminescence (CL) method for the determination of humic acid (HA) in water samples was first developed based on the redox reaction between humic acid and cerium(IV) in the acidic condition. Different with the former redox CL reaction which occurred in alkaline solution, no enhancers were needed and neither precipitation nor a second contamination would occur in the present CL system. Comparing with other spectrometric methods, we find that the proposed analysis system had better applicability and accuracy. Under the optimal experiment conditions, the CL peak height was linear with the concentration of HA in the range of 0.03 to 10.0 μg mL–1. The detection limit is 0.01 μg mL–1 (S/N = 3), and the relative standard deviation was 2.3% for 0.5 μg mL–1 HA solution with eleven repeated measurements. The present CL method was successfully applied to the determination of HA in tap water, spring water and river water samples with good recovery from 90.0 to 110.0%. A possible CL mechanism was proposed based on the results of UV and fluorescence spectrometry and the CL spectrum of HA. It was speculated that the semiquinone radicals in the excited state were the emitters. (Received April 16, 2007; Accepted May 24, 2007; Published October 10, 2007)

Introduction Humic substances are high molecular weight, heterogeneous organic materials that are the major constituents of soils and aquatic environments.1,2 Humic acid (HA), as the major fractions of humic substances, contributes vital properties to natural water, including sequestration, mobilization, and oxidative or reductive transformation of organic xenobiotic molecules, trace gases, and trace metal contaminants. It is reported that the halogens can interact with HA in drinking water treatment to produce halogenated carcinogens such as chloroform and bromoform, which are then directly introduced into drinking water with obvious health consequences.3 Thus, researchers are very concerned in the field of analytical chemistry about developing a highly sensitive analytical method to the determination of HA in different types of water samples. Since HA concentration in natural water samples is low, different powerful analytical methods, including UV and fluorometric methods,4–7 have been developed for the determination of HA. Associated analytical problems are sensitivity and interference, particularly from iron and polyphenolic substances (lignin). Hence, high-performance size exclusion chromatography8 and capillary zone electrophoresis9 with fluorescence detection, ELISA10 and environmental scanning electron microscopy11 were adopted to improve the analytical capability. However, these procedures resulted in †

To whom correspondence should be addressed. Present address: Department of Chemistry, Tsinghua University, Beijing 100084, China. E-mail: [email protected]

time-consuming, high cost and laborious processes. In addition, analytical methods for humic acid determination based on UVVis absorption and fluorescence are not accurate by direct analysis or by using “standard” humic acid from different sources because their molar absorptivities differ by up to 270%.12 The determination of HA by CL method has been investigated due to the high sensitivity by using KMnO4,13 N-bromosuccinimide.14 However, these methods caused precipitation13 and the contamination of bromide and hypobromite which might produce brominated carcinogens. Recently, Magdaleno et al.12 compared the CL determination of alkaline H2O2 in the presence of HCHO and peroxymonosulfate in basic medium, but the accuracy of their method was not confirmed. Tian et al.15 reported 1,3-dibromine-5,5dimethylhydantoin in the presence of glycine in alkaline medium. The method was similar to the above mentioned method of Ref. 13, unfortunately, the procedure was tedious because of the addition of glycine in sample solutions before analysis and also caused contamination of bromide and hypobromite. In contrast with the above CL system, Ce(IV) is not toxic and its CL redox reactions could effectively avoid the interference of chloride16 and the contamination of halogenated carcinogen. The Ce(IV)-involved CL reactions have been studied and widely used for the detection of various compounds, as has been reviewed by Lin.17 Ce(IV) was found to be easily reacted with the carbonyl and phenolic compounds.18,19 Based on these results, we conclude that the reaction between Ce(IV) and HA might take place because HA is rich in hydroxyls and carbonyls. Thus, a CL method based on Ce(IV) would be suitable for the determination of organic compounds in water. Although HA is defined as the fractions that are not soluble in water under acid condition (generally pH < 2), it does not mean the impossibility of a redox reaction of HA in acidic conditions. In this paper, we

1190

Fig. 1 Schematic diagram of the flow-injection CL system for the determination of HA. Acidic Ce(IV), 0.05 mol L–1 Ce(IV) sulfate solution in 0.05 mol L–1 sulfuric acid; carrier, water; D, detector; F, flow cell; P, peristaltic pump; PC, personal computer; W, waste; flow rate, 2.0 mL min–1.

have, for the first time, proposed the CL spectrum of HA in the acidic Ce(IV) solution and, according to the UV and fluorometric spectra, a possible CL mechanism was put forward. Under the optimal experimental conditions, the developed CL system could be used with good recovery and low enough limit of detection for the determination of HA in tap water, spring water and river water samples.

Experimental Reagents Chemicals of analytical grade were used as received. Ultrapure water was obtained from a compact ultrapure water system (18.3 MΩ cm–1, Barnstead, Iowa, USA). HA (salt sodium, technique grade) was purchased from Aldrich and Acros and was used without further purification. HA from Tianjin Jingke Institute of Fine Chemicals (Tianjin, China) was also used in the comparison experiments. Ce(IV) sulfate tetrahydrate and sulfuric acid were obtained from Beijing Chemical Industry Plant (Beijing, China). The working Ce(IV) solutions were prepared daily in sulfuric acid prior to use. Standard solutions of HA were prepared daily by further dilution of its stock solution (1000 mg L–1) which was stored at –4˚C. HA of 0.1000 g from Tianjin Jingke Institue of Fine Chemicals was dissolved in 10 mL 0.1 M NaOH solution and then diluted to 100 mL with ultrapure water. Apparatus and procedure A batch style BPCL luminescence analyzer (Institute of Biophysics, Chinese Academy of Science, Beijing, China) was used to obtain batch CL signals. A schematic diagram of the flow-injection CL system in this work is shown in Fig. 1. Reagent solution (acidic Ce(IV) solution) and carrier solution (water) were delivered by peristaltic pumps (P). The Ce(IV) solution was mixed in a mixing valve (V) with the sample solution (100 μL) that was injected into the carrier stream through the sample injection valve. The mixing coil (F) was made by coiling a piece of glass tubing (0.8 mm i.d. and 15.0 mm length) into a spiral disk shape and this was placed close to the photomultiplier tube. The CL emission was recorded with a flow injection CL analyzer (D, Lumiflow LF-800, Microtec NITI-ON, Funabashi, Japan) controlled by a personal computer. PTFE tubing (0.8 mm i.d.) was used to connect all components in the flow system. Calibration graphs were constructed by plotting the CL intensity (peak height) versus the concentration of HA. The fluorescence and absorption spectra were monitored using an F-2500 fluorescence spectrometer (Hitachi, Tokyo, Japan) and a Shimadzu UV-2401 UV-visible recording spectrophotometer (Shimadzu, Kyoto, Japan), respectively. The CL spectrum was obtained with a series of interference filters

ANALYTICAL SCIENCES OCTOBER 2007, VOL. 23

Fig. 2 Batch CL intensity–time profile. 100 μL Ce(IV) in 0.1 mol L–1 sulfuric acid solution + 100 μL HA; Ce(IV), 0.01 mol L–1; HA, 10 μg mL–1.

by the flow injection method. The filters were inserted between the mixing coil and the photomultiplier tube (PMT). The sample solutions were collected in a clean polyethylene flask and stored at –15˚C. The water samples were centrifuged for 10 min at 4000 rpm and then the supernatant were processed according to Martin and Pierce.20 Solutions used for calibration were prepared by mixing appropriate amounts of stock HA solution (1000 mg L–1) in volumetric flasks and diluting the mixtures to 50 ml with water.

Results and Discussion The batch studies of CL The CL profiles in the batch method for the mixtures of acidic Ce(IV) and HA solutions are shown in Fig. 2. While water was being added into the acidic Ce(IV) solution, no CL signals were recorded except the noise. As shown in Fig. 2, adding 100 μL of 10 μg mL–1 HA solution into 100 μL of Ce(IV) in acidic solution, a strong CL emission was recorded. The CL signal reaches a maximum within 0.3 s and then decreases rapidly. This means that the redox CL reaction of acidic Ce(IV) and HA solution is a fast reaction system and could be applied to the routine, rapid determination of HA. Effect of Ce(IV) concentration on the CL intensity The results of experiments showed that Ce(IV) was the essential factor which dramatically influenced the CL signals of the present system. Then, on the basis of the primary experiments, the concentration of Ce(IV) contributing to the CL signals was studied in detail ranging from 1.0 × 10–3 – 0.5 mol L–1. Examination of Fig. 3 suggested that the relative CL intensity increased with the concentration of the Ce(IV) when the concentration of Ce(IV) was below 0.05 mol L–1. Moreover, the CL intensity decreased when the concentration of Ce(IV) continued increasing. This could be attributed to the absorption of CL when the Ce(IV) ion concentration was high and the color of solution grew deep. The results illustrated that the maximum CL signal was found with Ce(IV) at 0.05 mol L–1. Effect of sulfuric acid concentration on the CL intensity It is well known that only in acidic condition could Ce(IV) be a good oxidant. Although HA is not soluble in water under acid condition when pH value is below 2, the present CL reaction is so quick (as shown in the section of “The batch studies of CL”) that the fractions of precipitation of HA in low concentration could not effect the accuracy of the CL signals. Hence, hydrochloric acid, nitric acid, sulfuric acid and phosphoric acid were used as

ANALYTICAL SCIENCES OCTOBER 2007, VOL. 23

1191 Table 1 Comparison the present CL method for the determination of HA with other reported methods Method

Fig. 3 The effect of Ce(IV) concentration on the CL intensity. Sulfuric acid, 0.01 mol L–1; HA, 2.0 μg mL–1.

This method Automated fluorometry Cloud-point extraction flow-injection UV High-performance size exclusion chromatography– fluorometry Peroxymonosulfate CL method

Detection limit/ Dynamic linear Ref. range/µg mL–1 µg mL–1 0.01 0.5 0.005

0.03 – 10.0 0.5 – 20 0.005 – 1

4 7

0.05

0.05 – 10

8

0.24

0.5 – 20

12

Performance of the proposed method Under the recommended conditions given above and following the procedure described in the experimental section, the calibration graph was linear from 0.03 to 10.0 μg mL–1. The maximum peak height increased linearly with the increasing of HA concentration, as expressed by the equations ΔI = 1.16 × 103CHA + 88.2 (r = 0.9960). The detection limit is 0.01 μg mL–1 (S/N = 3), and the relative standard deviation was 2.3% for 0.5 μg mL–1 HA solution with eleven repeated measurements. Fig. 4 The effect of the concentration of sulfuric acid on the CL intensity. Ce(IV), 0.05 mol L–1; HA, 2.0 μg mL–1.

the reaction media to study their effect on the CL intensity. The results showed that the maximum CL signal was attained in sulfuric acid. The influence of different concentrations of sulfuric acid from 0.005 to 0.5 mol L–1 was studied through the same flow manifold as demonstrated in Fig. 4. The present CL reaction occurred in acidic medium, so the S/N ratio increased with the increasing of the concentration of sulfuric acid ranging from 0.005 to 0.05 mol L–1. After that, the S/N ratio decreased. This was due to two factors. One was that the solubility of HA decreased as the concentration of H2SO4 increased, the other was an excess of H2SO4 did not favor the occurrence of the CL reaction. Thus, a concentration of 0.05 mol L–1 was selected for this work. Effect of the flow rate The influence of flow rate was studied. This parameter was critical owing to its influence over the point in the flow manifold in which the excited molecule emits the light and consequently over the signal magnitude: too low flow rates could result in maximum emission before the flow cell and too fast flow rates in maximum emission after it. In order to determine the optimum operating conditions of the flow system, we measured the signal for 0.5 μg mL–1 HA with respect to the flow rate of the reagent solutions. The flow rates of the CL system were investigated from 0.5 to 4.0 mL min–1. In the range of 0.5 to 2.0 mL min–1, the CL intensity increases with increasing flow rate and shows a maximum value. After that, the CL signals decrease slowly with the increase of flow rate. The flow rate of 2.0 mL min–1 was selected as an appropriate condition considering both suitable analytical efficiency and lower solution consumption.

Interference studies The tolerable concentration ratios with respect to 0.5 μg mL–1 HA for interference at 5% level were examined. One thousandfold excess K+, NO3–, Na+, Cl–, Ca2+, NH4+, SO42–; 500-fold Ag+, Pb2+, Zn2+, Cu2+, Mg2+; 100-fold Al3+, Ni2+, Cr3+, EDTA, HPO42–, HCO3–, CO32–, Br–, Cd2+, Co2+, F–, Mn2+; 50-fold Fe3+; 5-fold NO2–, C2O42–; 2-fold orthocresol, paracresol, resorcinol, hydroqinone, 3-chlorophenol, p-chlorophenol, aniline, 2,4dinitrophenol, o-nitrophenol, p-nitrophenol; 1-fold Fe2+; 0.1fold phenol, pyrocatechin, m-cresol, salicylic acid; 0.01-fold pyrogallol, had almost no effect on the determination of HA. These phenolic compounds and Fe2+ could not influence the accuracy of quantity of HA because their content was only μg L–1 in natural aquatic environments.21,22 As compared in Table 1, the present method has a wider linear range and lower detection limit than the methods previously reported. Although Ref. 7 has a much lower detection limit than that of the proposed paper, others hydrophobic compounds could be extracted by cloud-point extraction methodology and a second environmental contamination would happen. The applicability of the present method Under the optimal experimental conditions, 1.0 μg mL–1 HA from different chemical companies was investigated by comparing their CL intensity and CL spectra. The results indicated that the peak height of HA from Acros, Aldrich and Tianjin was 1365, 1350 and 1302, respectively. To our pleasure, all of the CL spectra of different HA were very similar to each other with the CL maximum emission wavelength from 510 to 540 nm, as illustrated in Fig. 5. Therefore, the proposed method could be used in various HA samples from different areas and sources with a better accuracy than other spectrometric methods. Applications Various water samples including tap water, spring water and river water were determined. Tap water and spring water samples were directly injected into the analysis system after the

1192

ANALYTICAL SCIENCES OCTOBER 2007, VOL. 23

Table 2 Determination of HA by the present CL method (n = 5) Sample

Content

Tap water

0.11

Spring water

0.86

Qinghe river water

3.37

Added/ µg mL–1

Founded

Recovery, %

RSD, %

0.1 1.0 8.0 0.5 3.0 8.0 1.5 5.0 8.0

0.22 1.01 8.73 1.49 3.75 8.89 4.85 8.19 10.7

110.0 90.0 107.8 126.0 96.3 100.8 98.7 96.4 91.6

4.8 3.0 1.5 1.9 1.9 1.2 2.1 2.2 1.4

process according to Martin and Pierce20 by using a six-way valve, while in the case of river water, the process of centrifugation as described in the Procedure section was needed. Spiked water samples with standard solutions were used to evaluate the utility of the present method. Table 2 shows the average values of the detections. It can be seen that good recoveries were obtained for the three water samples and a relatively high recovery (126.0%) for spring water, due to its higher concentration of mineral materials which might have some adsorptive interaction with HA. So, the proposed method was applicable for detection of HA in water. The CL mechanism studies of the present system In order to investigate the possible CL mechanism of HA under the condition of acidic Ce(IV) solution, we measured the CL spectrum of HA with cut-off filters. The CL maximum emission wavelength of HA was between 510 and 540 nm as illustrated in Fig. 5. CL spectra of 0.01% Fluka HA (irraditated) measured by Slawinski et al.,23,24 with cut-off filters and a calibrated photomultiplier operating in the single photo counting mode, revealed three main emission bands at 480 – 500, 570 and 610 – 650 nm. They thought that the excitation resulting from the oxidation or an energy transfer from 1O2 and the sensitized CL from the fluorescing part of HA’s molecules contributed considerably to these emissions. Experiments in the present work showed that the CL intensity remained almost unchanged when dissolved oxygen was removed by a flow of nitrogen, which was consistent with the results reported by Fujimori et al.25 This result strongly suggested that a quite different CL mechanism from Slawinski’s was involved in this work. Alwarthan26 suggested that the emitter was an excited state of Ce(III) in a Ce(IV)-tryptophan. However, the maximum CL emission wavelength was at 350 nm. Thus, to elucidate the mechanism of HA-CL with acidic Ce(IV) solution, one should study the fluorescent and UV spectra of all the compounds involved in the reaction. The fluorescence spectra of HA, HA in acidic solution, Ce(IV) and HA-Ce(IV) in acidic solution were recorded, respectively, as shown in Fig. 6. In the presence of 0.01 mol L–1 H2SO4, the mixing of HA and Ce(IV) solution resulted in the disappearance of fluorescence excitation and emission of HA at 276 and 458 nm, as shown in Figs. 6c and 6d. To our surprise, none of new fluorescence excitation and emission wavelengths could be found after the reaction of acidic Ce(IV) and HA solutions. Since the major oxygen-containing functional groups in HA were carboxyls, hydroxyls and carbonyls, we speculate that the products of HA during the redox reaction of Ce(IV) change into quinones which had no fluorescence characteristics. In that case, the UV spectra of the present experiment were studied.

Fig. 5 The CL spectrum of HA-Ce(IV) in acidic condition. (a) HA from Acros; (b) from Aldrich; (c) from Tianjin. Ce(IV), 0.05 mol L–1; sulfuric acid, 0.05 mol L–1; HA, 1.0 μg mL–1; flow rate, 2.0 mL min–1.

Fig. 6 The fluorescence excitation and emission spectra of HA and the products of the present CL reaction. (a) Excitation spectra of HA; (b) emission spectra of HA; (c) excitation spectra of the products with the emission wavelength at 458 nm; (d) emission spectra of the products with the excitation wavelength at 276 nm. Ce(IV), 0.05 mol L–1; sulfuric acid, 0.05 mol L–1; HA, 1.0 μg mL–1.

The absorption spectra of HA after mixing with acidic Ce(IV) solution clearly showed that a reaction occurred between HA and Ce(IV) and that a new compound was formed with the maximum absorption wavelength at 222 nm (Fig. 7c). The absorption spectra of Ce(IV) (Fig. 7a) and HA (Fig. 7b) were also recorded. These showed a big difference from the new compound, which indicated that a similar reaction to that reported by Cui et al.27 might occur. The results of the above experiments and the reactions could be summarized in the following scheme:

ANALYTICAL SCIENCES OCTOBER 2007, VOL. 23

1193

References

Fig. 7 The UV adsorption spectra of (a) 0.05 mol L–1 Ce(IV); (b) 1.0 μg mL–1 HA; (c) 0.01 mol L–1 Ce(IV) in 0.05 mol L–1 sulfuric acid + 1.0 μg mL–1 HA with 0.01 mol L–1 Ce(IV) in 0.05 mol L–1 sulfuric acid as reference solution.

According to the former report,28,29 humic substances were known to be rich in stable free radicals, while the free radical contents of humic materials were caused predominantly or entirely by the semiquinone radicals (II). Moreover, the report of Senesi and his collaborators,28 mentioned that the oxidation solutions were effective in increasing the free radicals. Additionally, according to Ref. 27, Ce(IV) could rapidly react with phenolic compounds with o-benzoquinone as products. Thus, it could be concluded that: (i) in the presence of Ce(IV), large amounts of semiquinone radicals (II) were formed; (ii) the excited state semiquinones (III) were generated during the reaction of HA with acidic Ce(IV) solution; (iii) the excited state semiquinones (III) turned into quinone (IV) and CL signals were produced.

Conclusions A flow-injection CL method for the determination of HA in tap water, spring water and river water samples was developed. Compared with other methods, the proposed method had the advantages of simplicity, rapidity, and high sensitivity in the determination of HA in natural water. Better applicability and accuracy were manifested in this CL analysis system when used in the determination of HA from different areas and origins. The CL spectrum of HA in the redox reaction without irradiation was first brought forward. Based on the studies of UV, fluorescence and CL spectra, a possible CL mechanism was proposed and the intermediate emitters were regarded as excited semiquinone radicals. The system has been successfully applied to various aquatic environmental samples with good recovery and selectivity. These successful demonstrations indicate that the proposed CL method may have a wide use in the detection of HA in various aquatic environments.

Acknowledgements The authors would like to express their thanks for the financial supports from the National Natural Science Foundation of China (Nos. 20437020, 20575073).

1. P. A. Maurice and K. Namjesnik-Dejanovic, Environ. Sci. Technol., 1999, 33, 1538. 2. J. R. Lead, J. Hamilton-Taylor, N. Hesketh, M. N. Jones, A. E. Wilkinson, and E. Tipping, Anal. Chim. Acta, 1994, 294, 319. 3. F. J. Lu, Appl. Organomet. Chem., 1990, 4, 191. 4. G. L. Brun and D. L. D. Milbum, Anal. Lett., 1977, 10, 1209. 5. M. H. Sorouradin, M. Hiraide, and H. Kawaguchi, Anal. Sci., 1993, 9, 863. 6. T. Nomizu, M. Sanji, M. Hiraide, and A. Mizuika, Anal. Sci., 1989, 5, 363. 7. J. C. A. de Wuilloud, R. G. Wuilloud, B. B. M. Sadi, and J. A. Caruso, Analyst, 2003, 128, 453. 8. S. Nagao, T. Matsunaga, Y. Suzuki, T. Ueno, and H. Amano, Water Res., 2003, 37, 4159. 9. J. D. Hewitt and L. B. McGown, Appl. Spectrosc., 2003, 57, 256. 10. P. Ulrich, M. G. Weller, D. Knopp, and R. Niessner, Anal. Sci., 1993, 9, 795. 11. P. S. Redwood, J. R. Lead, R. M. Harrison, I. P. Jones, and S. Stoll, Environ. Sci. Technol., 2005, 39, 1962. 12. G. B. Magdaleno and N. Ciochev, Anal. Chim. Acta, 2005, 552, 141. 13. D. F. Marino and J. D. Ingle, Jr., Anal. Chim. Acta, 1981, 124, 23. 14. J. Michalowshki, P. Halaburda, and A. Kojlo, Anal. Chim. Acta, 2001, 438, 143. 15. Y. H. Tian, Q. J. Song, and Z. Z. Hua, Anal. Lett., 2005, 38, 2439. 16. S. I. Montalvo and J. D. Ingle, Jr., Talanta, 1993, 40, 167. 17. J.-M. Lin, “Chemiluminescence-basic Principles and Applications”, 2004, Chemical Industry Press, Beijing, 121. 18. Z. K. He, X. Y. Li, H. Meng, S. F. Lu, G. W. Song, L. J. Yuan, and Y. E. Zeng, Anal. Lett., 1998, 31, 1553. 19. Q. L. Zhang, H. Cui, A. Myint, M. Lian, and L. J. Liu, J. Chromatogr., A, 2005, 1095, 94. 20. D. F. Martin and R. H. Pierce, Jr., Environ. Lett., 1971, 1, 49. 21. M. Saraji and M. Bakhshi, J. Chromatogr., A, 2005, 1098, 30. 22. S. Kawakubo, Y. Okazaki, and M. Iwatsuki, Anal. Sci., 2003, 19, 373. 23. J. Slawinski, W. Puzyna, and D. Slawinska, Photochem. Photobiol., 1978, 28, 75. 24. J. Slawinski, W. Puzyna, and D. Slawinska, Photochem. Photobiol., 1978, 28, 459. 25. K. Fujimori, N. Takenaka, H. Bandow, and Y. Maeda, Anal. Sci., 2001, 17, 161. 26. A. A. Alwarthan, Anal. Chim. Acta, 1995, 317, 233. 27. H. Cui, Q. L. Zhang, A. Myint, X. W. Ge, and L. J. Liu, J. Photochem. Photobiol. A, 2006, 181, 238. 28. N. Senesi, Y. Chen, and M. Schnitzer, Soil Biol. Biochem., 1977, 9, 397. 29. G. G. Choudhry, “Humic Substances: Structural, Photophysical, Photochemical and Free Radical Aspects and Interactions with Environmental Chemicals”, 1984, Gordon and Breach Science Publishers, New York, 64.