- Email: [email protected]
Spectrofluorimetric determination of copper(II) by its static quenching effect on the fluorescence of 4,5-dihydroxy-1,3-benzenedisulfonic acid
Talanta 55 (2001) 163– 169 www.elsevier.com/locate/talanta
Spectrofluorimetric determination of copper(II) by its static quenching effect on the fluorescence of 4,5-dihydroxy-1,3-benzenedisulfonic acid Hye-Seon Kim, Hee-Seon Choi Department of Chemistry, The Uni6ersity of Suwon, P.O. Box 77, Suwon, 445 -743, South Korea Received 16 January 2001; received in revised form 29 March 2001; accepted 30 March 2001
Abstract A spectrofluorimetric method has been developed for the determination of trace Cu(II) in real samples with 4,5-dihydroxy-l,3-benzenedisulfonic acid (Tiron) as a fluorimetric reporter. Tiron is very soluble in water and is a good fluorimetric reagent. However, as Tiron was complexed with Cu(II), the fluorescence intensity decreased proportionally to the concentration of Cu(II) by a static quenching effect. The excitation wavelength and the fluorescence wavelength of Tiron were 294 and 350 nm, respectively, as it was caused by a quenching effect from Cu(II) at pH 8.0. The highest sensitivity was shown at Tiron concentration of 5.0 × 10 − 5 M. To enhance the quenching effect, the Cu(II)–Tiron complex solution was heated up to 80°C for 90 min. As for Cu(II), the interference by Co(II) was very serious, which was eliminated by oxalate ion. The linear response to Cu(II) was shown at the concentration range between 5.0 ×10 − 7 and 1.0×10 − 5 M. With this proposed method, the detection limit of Cu(II) was 3.83( 90.09)×10 − 7 M. Recoveries of Cu(II) in the diluted brass samples and the stream water samples were almost 100%. Based on results from the experiment, this proposed technique could be applied to the practical determination of Cu(II) in real samples. © 2001 Elsevier Science B.V. All rights reserved. Keywords: 4,5-Dihydroxy-1,3-benzenedisulfonic acid (Tiron); Copper; Quenching effect; Brass; Stream water
1. Introduction With spectrophotometry, electroanalytical chemistry, chromatography and radioactive analysis, many new and modified techniques that can be used to determine metal ions in real samples have been studied. While these methods have * Corresponding author. Fax: + 82-31-2229385. E-mail address: [email protected] (H.-S. Choi).
some pros and cons, spectrofluorometry has merit in a sense that it is more sensitive, convenient, and simpler than other techniques. Spectrofluorometry was used as follows; to intensify the fluorescence to combine fluorophore and metal ions for determining metal ions [1–3]; to determine metal ions in micellar medium [4,5]; to investigate the fluorescence of ternary complex [6,7]; to determine the metal ions and some biological materials in real samples [8–10], and so on.
0039-9140/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S0039-9140(01)00405-2
164
H.-S. Kim, H.-S. Choi / Talanta 55 (2001) 163–169
Metal ions have often been determined with the technique that enhances the fluorescence intensity of fluorophore bound with metal ions by their concentrations. With the complexing agent, the fluorophore, combined with metal ion to form complex, the rigidity of fluorophore increased, and it resulted in increased quantum yield of fluorescence. In contrast, the quenching effect has also been used to determine metal ion where fluorescence intensity of fluorophore decreases as the concentration of metal ion increases. The fluorescence quenching method has been used for the determinations of organic materials [11,12], metal ions [13– 16], inorganic anion [17] and biological materials [18]. The 4,5-dihydroxy-l,3-benzenedisulfonic acid (Tiron) has been used as a doping species in ion selective electrode [19], as an activator for catalytic reaction [20,21], as a color-developing reagent for separated metal ions [22,23] and as a complexing agent in the determination of DNA and RNA by fluorescence quenching [24]. Tiron, an excellent fluorophore, combines with Cu(II) to form water-soluble chelate. But Cu(II)– Tiron complex does not fluoresce. In this study, Tiron as fluorophore was used to determine Cu(II) by static quenching phenomenon.
2. Experimental
2.2. Apparatus Perkin–Elmer model LS-50 spectrofluorometer was used to measure the fluorescence of Tiron. Both spectral bandwidths of excitation and fluorescence spectra were 2.5 nm. All fluorescence measurements were accomplished in 1 cm quartz cell without removing the oxygen in the sample because spectral difference was not observed on deaeration with N2. A GBC model 903 flame atomic absorption spectrometer and a HP 4500 ICP/MS spectrometer were also used to determine Cu(II) in real samples. To adjust the pH, a Bantex model 300A digital pH meter with a combined calomel and glass electrode was used.
2.3. Calibration cur6e After the aliquots of Cu(II) stock solution were each placed in several 10 ml volumetric flasks for the concentration of Cu(II) ranging from 1.0× l0 − 7 to 1.0× l0 − 5 M, 5.0 ml of pH 8.0 borate buffer solution and 1.0 ml of 5.0× l0 − 4 M Tiron were added, filling deionized water to the mark. These solutions were heated up to 80°C in waterbath for 90 min, cooled down to room temperature and used to measure fluorescence intensity at 294 nm of excitation wavelength and at 350 nm of fluorescence wavelength.
2.4. Reco6ery yields and determinations of Cu (II) in the diluted brass samples and the stream water
2.1. Reagents and solution The 4,5-dihydroxy-1,3-benzenedisulfonic acid (Tiron), manufactured by Aldrich Co., was used without further purification and with appropriate concentration whenever necessary. Standard Cu(II) stock solution was made 1.0× 10 − 3 M by preparing from copper(II) nitrate (Aldrich Co.). Buffer solution of pH 8.0 was prepared by pH meter with 0.025 M sodium tetraborate and 0.1 M HCl. All chemicals used were of analytical and reagent grade, while the deionized water by a Barnstead catridge deionization system (Barnstead Co.) was used throughout all experimental procedures.
A 0.500 g brass sample cleaned by acetone and deionized water was taken into the 250 ml beaker, added by 10 ml of 6M HNO3, heated in fumehood to dissolve the brass sample completely, cooled down to the room temperature, and diluted to 1000 ml in a volumetric flask. Plus, 1.0 ml of this brass solution was transferred to a 1000 ml volumetric flask, and diluted by filling to the mark to get 5.0× 10 − 5% (w/v) brass sample. Suwon stream water was used after the suspended matters or particles were filtered out by glass filter (1-G-4). To investigate the recovery yield of Cu(II) in a diluted brass sample, three 3.0 ml brass samples of 5.0×10 − 5% were transferred to
H.-S. Kim, H.-S. Choi / Talanta 55 (2001) 163–169
one of each 10 ml volumetric flask, and then 0.0, 1.0 and 1.5 ml of 1.0×10 − 5 M Cu(II) standard solution were added to each flask, followed by the procedures which are same as calibration curve procedures. For stream water, the experimental procedures were same as those of diluted brass samples.
3. Results and discussion
3.1. Excitation and fluorescence spectra of Tiron The excitation spectra and the fluorescence spectra of 5.0×10 − 5 M Tiron in various concentrations of Cu(II) are shown in Fig. 1. It was found that fluorescence intensity of excitation (uex = 294 nm) and fluorescence (uf =350 nm) spectra decreased, as the concentration of Cu(II) increased. It was thought to be because static quenching occurred quantitatively in a given concentration range of Cu(II) as Cu(II)– Tiron complex was formed. It was also observed under experiment that fluorescence intensity of Tiron fluctuated in some degree (R.S.D.=2.0%), causing large relative errors in determination of ana-
165
lytes. However, the fluctuation of fluorescence intensity of Tiron decreased by 10 times (R.S.D.=0.2%) when Cu(II) was contained in Tiron solution. While both fluorescence intensities of Tiron only and Tiron in Cu(II) solution decreased slowly in time, the difference between these fluorescence intensities remained constant, which meant that the decreased amount of fluorescence intensity might be equal if measured at any time. However, it was necessary that fluorescence should be measured after waiting for a while because fluorescence intensity somewhat fluctuated if its fluorescence was measured as soon as the Tiron solution was prepared.
3.2. Concentration of Tiron Fig. 2 shows how fluorescence intensity of Tiron changed by the concentration of Tiron in a given concentration range of Cu(II). Since the curve with the steepest decrease of fluorescence intensity by a given concentration range of Cu(II), that is, the highest value of slope, represented the highest sensitivity, it was investigated what concentration of Tiron had the highest slope. When the concentration of Tiron was 5.0×10 − 5 M, its
Fig. 1. Excitation and emission spectra of Tiron (5.0 ×10 − 5 M) as the concentration of Cu2 + . Concentrations of Cu2 + from top spectrum were 0.0, 5.0 ×l0 − 7, 1.0× 10 − 6, 3.0× 10 − 6, 6.0× 10 − 6 and 1.0 × 10 − 5 M, respectively.
166
H.-S. Kim, H.-S. Choi / Talanta 55 (2001) 163–169
Fig. 2. Effect of the concentration of Tiron on the fluorescence quenching curves of Tiron in the presence of Cu(II).
slope was the highest, and its linearity was good (R 2 =0.9971). It was inferred that at lower concentrations of Tiron the slope was gentle because Tiron was not quantitatively complexed in a given concentration range of Cu(II). Also, it was supposed that at higher concentrations of Tiron the relative magnitude of quenching effect by Cu(II) compared with the lower concentration of Tiron decreased as the traction of Tiron that could be complexed with Cu(II) was lower.
3.3. pH Since various buffer solutions, i.e. different electrolytes were used to adjust the pH, the excitation wavelength and the fluorescence wavelength might change. Particularly, fluorescence intensity could be altered because the Tiron complex-forming ability in Cu(II) solution was dependent on pH. So, pH should be controlled to be optimum value. It was investigated how fluorescence intensity of Tiron in a given concentration of Cu(II) solution would change. The steepest slope that represented the highest decreasing effect of fluorescence intensity was shown at pH 8.0. At lower pH, as the oxygen atom in chelating site of Tiron had more affinity power with proton at
higher concentration of proton, Tiron could not play a role as ligand well to complex with Cu(II). At higher pH, as Cu(II)–Tiron complexation was competed with copper hydroxide precipitation, the slower slope was shown.
3.4. Temperature As temperature is an important factor in Cu(II)–Tiron complex formation, it was investigated at room temperature how the fluorescence intensity of Tiron in a given concentration range of Cu(II) decreased when heating to various temperatures during 90 min, and its results were shown in Fig. 3. In temperature experiment, Cu(II)–Tiron complex was formed more quantitatively as the temperature was higher, e.g. at 80°C.
3.5. Heating time Tiron and Cu(II) should be heated to form a stable complex quantitatively and rapidly at a given temperature for appropriate time. Fig. 4 shows that the fluorescence intensity of Tiron decreased in a given concentration range of Cu(II)
H.-S. Kim, H.-S. Choi / Talanta 55 (2001) 163–169
at 80°C as heating time increased. The highest slope was obtained at 120 min of heating time with its very poor linearity (R 2 =0.9800). While its slope at 90 min of heating time was somewhat slower than at 120 min, its linearity was very good
167
(R 2 = 0.9956). At heating time less than 90 min, Cu(II)–Tiron complex was not formed quantitatively, and at more than 90 min, it was supposed that Cu(II)–Tiron complex might be unstable and dissociate.
Fig. 3. Fluorescence quenching curves of Tiron in Cu(II) solution at various temperatures.
Fig. 4. Effect of heating time on the fluorescence quenching curves of Tiron in the presence of Cu(II).
H.-S. Kim, H.-S. Choi / Talanta 55 (2001) 163–169
168
Table 1 Tolerance limitsa of interfering species for the determination of Cu(II) (1.0×10−6 M) in 5.0×10−5 M Tiron solution Mole ratiob
Interfering species
500
− − 2+ NH+ , F−, SO2− 4 , NO3 , Mg 4 , HCO3 − + + 2− − Cl , Na , K , HPO4 , I , NH3
250 100 50 30 10 5 3
Ca2+ Al3+, Ni2+
micellar medium’. Despite the increased slope in 0.1% Tween 80 medium, it was difficult to use since the fluorescence wavelength of Tiron was overlapped by that of Tween 80. It was supposed that its enhanced slope was due to self-quenching of Tween 80. Thus, it was known that the more efficient decreasing effect of fluorescence intensity of Tiron in micellar medium could not occur.
3.7. Interference Fe3+ Cd2+, Zn2+ Pb2+ Co2+
a
Tolerance limit is the maximum concentration in which there is less than 3% effect on fluorescence intensity. b Mole ratio of interfering species to Cu(II).
3.6. Surfactants Although it was expected that the quenching effects decreased by external conversion at high viscosity of micellar medium, it might also be expected that the static quenching effect increased by more stable complex forming in micellar medium. To obtain more efficient quenching effect, various surfactants, such as cetyltrimethylammonium bromide as cationic, sodium dodecylsulfate as anionic, and Triton X-100 and Tween 80 as nonionic surfactants were used. Comparison of the decreasing magnitudes of fluorescence intensity showed less difference for two cases of ‘in no micellar medium’ and ‘in
The interfering effects were investigated which might be caused by other metal ions that can form good complex with Tiron such as Fe(III), Al(III), Cd(II), Co(II), Ni(II), Pb(II), and Zn(II), various species like I− and NH3 that can combine with Cu(II),the analyte ion, and probable species 2+ + such as NO− , Na+, Cl−, and Ca2 + 3 , NH4 , Mg that can exist in real samples. To observe the interfering effect on the determination of Cu(II), the fluorescence intensity of Tiron was measured after the probable interfering ions were added to 1.0×l0 − 6 M Cu(II) and 5.0×10 − 5 M Tiron solution by 1, 3, 5, 10, 30, 50, 100, 250 and 500 times mole of Cu(II). Tolerance limits on these species were investigated on the fluorescence intensity of 5.0× 10 − 5 M Tiron solution containing 1.0×10 − 6 M Cu(II) at maximum amount corresponding to 3% relative error, and were shown in Table 1. Co(II) interfered with, but the interfering effect by Co(II) was completely eliminated by adding sodium oxalate 20 times moles of Tiron.
Table 2 Analytical data of Cu(II) in real samples Real samples
Stream water
Brass sample
a b
Spiked (M)
0.00 1.00×10−6 1.50×10−6 0.00 1.00×10−6 1.50×10−6
Measured (M)
Recovery (%)
Other methodsa
R.S.D. (%)
This methodb
R.S.D. (%)
1.68×10−6
2.3
1.65×10−6 2.68×10−6 3.12×10−6 5.18×10−6 6.15×10−6 6.62×10−5
2.8 2.7 3.2 3.6 3.2 2.9
5.22×10−6
1.5
ICP/MS data on stream waters and FAAS data on brass samples. These values were averaged from seven stream water and seven brass samples.
103 98 97 96
H.-S. Kim, H.-S. Choi / Talanta 55 (2001) 163–169
3.8. Application to brass sample and stream water A calibration curve was constructed at optimum conditions according to experimental procedure in Section 2.3. The linear response range of Cu(II) was 5.0×l0 − 7M – 1.0 ×l0 − 5 M. The correlation coefficient of 0.9948 showed that this calibration curve had good linearity. The detection limit [25] was 3.83(90.09) ×10 − 7 M. With a given amount of Cu(II) spiked in diluted brass sample and stream water, recovery yields were determined using this calibration curve, listed in Table 2, and were found to be nearly 100%. Cu(II) in diluted brass sample and stream water was also determined by flame atomic absorption spectrometry and ICP/MS spectrometry. Results from these techniques were also included in Table 2 and at the 95% confidence level, no difference between results from the proposed and reference methods had been established. Therefore, this proposed technique could be applied to the determination of Cu(II) in real samples.
References [1] R.D. Bautista, A.I. Jimie´ nez, F. Jime´ nez, J.J. Arias, Talanta 43 (1996) 421. [2] S. Nalini, T.V. Ramakrishna, Talanta 43 (1996) 1437. [3] F.H. Hernandez, J.M. Escriche, Analyst 109 (1984) 1585. [4] T. Taketatsu, Talanta 29 (1982) 397. [5] F.H. Hernandez, J.M. Escriche, M.T.G. Andieu, Talanta
.
169
33 (1986) 537. [6] P.R. Haddad, P.W. Alexander, L.E. Smyihe, Talanta 23 (1976) 275. [7] P.C. Ioannou, E.S. Lianidou, D.G. Konstantianos, Anal. Chim. Acta 300 (1995) 237. [8] C.C. Blanco, A.M.G. Campana, F.A. Barrero, M.R. Ceba, Talanta 42 (1995) 1037. [9] A. Pal, N.R. Jana, T.K. Sau, M. Bandyopadhyay, T. Pal, Anal. Comm. 33 (1996) 315. [10] Z. Jiang, X. Liu, A. Liang, Anal. Chim. Acta 346 (1997) 249. [11] J.H. Ayala, A.M. Afonso, V. Gonza´ lez-Diaz, Microchem. J. 60 (1998) 101. [12] A.L. Hunt, J.F. Alder, Anal. Chim. Acta 387 (1999) 207. [13] T.M.A. Razek, S. Spear, S.S.M. Hassen, M.A. Arnold, Talanta 48 (1999) 269. [14] K Cha, K. Park, Talanta 46 (1998) 1567. [15] N. Jie, Q. Zhang, Y. Yang, X. Huang, Talanta 46 (1998) 215. [16] B. Juskowiak, Anal. Chim. Acta 320 (1996) 115. [17] N. Jie, D. Yang, Q. Jiang, Q. Zhang, L. Wei, Microchem. J. 62 (1999) 371. [18] C.Q. Ma, K.A. Li, S.Y. Tong, Anal. Chim. Acta 333 (1996) 83. [19] J. Migdalski, T. Blaz, A. Lewenstam, Anal. Chim. Acta 322 (1996) 141. [20] T. Fujimoto, N. Teshima, M. Kurhara, S. Nakano, T. Kawashima, Talanta 49 (1999) 1091. [21] J. Gao, X. Zhang, W. Yang, B. Zhao, J. Hon, J. Kang, Talanta 51 (2000) 447. [22] M. Busch, A. Seubert, Anal. Chim. Acta 399 (1999) 223. [23] J.F. van Staden, H. du Plessis, R.E. Taljaard, Anal. Chim. Acta 357 (1997) 141. [24] Y. Zhao, D. Wang, X. Guo, J. Xu, Anal. Chim. Acta 353 (1997) 329. [25] D.A. Skoog, F.J. Holler, T.A. Nieman, Principles of Instrumental Analysis, 5th edn, Saunders College Publishing, 1998, p. 13.
Spectrofluorimetric determination of copper(II) by its static quenching effect on the fluorescence of 4,5-dihydroxy-1,3-benzenedisulfonic acid Hye-Seon Kim, Hee-Seon Choi Department of Chemistry, The Uni6ersity of Suwon, P.O. Box 77, Suwon, 445 -743, South Korea Received 16 January 2001; received in revised form 29 March 2001; accepted 30 March 2001
Abstract A spectrofluorimetric method has been developed for the determination of trace Cu(II) in real samples with 4,5-dihydroxy-l,3-benzenedisulfonic acid (Tiron) as a fluorimetric reporter. Tiron is very soluble in water and is a good fluorimetric reagent. However, as Tiron was complexed with Cu(II), the fluorescence intensity decreased proportionally to the concentration of Cu(II) by a static quenching effect. The excitation wavelength and the fluorescence wavelength of Tiron were 294 and 350 nm, respectively, as it was caused by a quenching effect from Cu(II) at pH 8.0. The highest sensitivity was shown at Tiron concentration of 5.0 × 10 − 5 M. To enhance the quenching effect, the Cu(II)–Tiron complex solution was heated up to 80°C for 90 min. As for Cu(II), the interference by Co(II) was very serious, which was eliminated by oxalate ion. The linear response to Cu(II) was shown at the concentration range between 5.0 ×10 − 7 and 1.0×10 − 5 M. With this proposed method, the detection limit of Cu(II) was 3.83( 90.09)×10 − 7 M. Recoveries of Cu(II) in the diluted brass samples and the stream water samples were almost 100%. Based on results from the experiment, this proposed technique could be applied to the practical determination of Cu(II) in real samples. © 2001 Elsevier Science B.V. All rights reserved. Keywords: 4,5-Dihydroxy-1,3-benzenedisulfonic acid (Tiron); Copper; Quenching effect; Brass; Stream water
1. Introduction With spectrophotometry, electroanalytical chemistry, chromatography and radioactive analysis, many new and modified techniques that can be used to determine metal ions in real samples have been studied. While these methods have * Corresponding author. Fax: + 82-31-2229385. E-mail address: [email protected] (H.-S. Choi).
some pros and cons, spectrofluorometry has merit in a sense that it is more sensitive, convenient, and simpler than other techniques. Spectrofluorometry was used as follows; to intensify the fluorescence to combine fluorophore and metal ions for determining metal ions [1–3]; to determine metal ions in micellar medium [4,5]; to investigate the fluorescence of ternary complex [6,7]; to determine the metal ions and some biological materials in real samples [8–10], and so on.
0039-9140/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S0039-9140(01)00405-2
164
H.-S. Kim, H.-S. Choi / Talanta 55 (2001) 163–169
Metal ions have often been determined with the technique that enhances the fluorescence intensity of fluorophore bound with metal ions by their concentrations. With the complexing agent, the fluorophore, combined with metal ion to form complex, the rigidity of fluorophore increased, and it resulted in increased quantum yield of fluorescence. In contrast, the quenching effect has also been used to determine metal ion where fluorescence intensity of fluorophore decreases as the concentration of metal ion increases. The fluorescence quenching method has been used for the determinations of organic materials [11,12], metal ions [13– 16], inorganic anion [17] and biological materials [18]. The 4,5-dihydroxy-l,3-benzenedisulfonic acid (Tiron) has been used as a doping species in ion selective electrode [19], as an activator for catalytic reaction [20,21], as a color-developing reagent for separated metal ions [22,23] and as a complexing agent in the determination of DNA and RNA by fluorescence quenching [24]. Tiron, an excellent fluorophore, combines with Cu(II) to form water-soluble chelate. But Cu(II)– Tiron complex does not fluoresce. In this study, Tiron as fluorophore was used to determine Cu(II) by static quenching phenomenon.
2. Experimental
2.2. Apparatus Perkin–Elmer model LS-50 spectrofluorometer was used to measure the fluorescence of Tiron. Both spectral bandwidths of excitation and fluorescence spectra were 2.5 nm. All fluorescence measurements were accomplished in 1 cm quartz cell without removing the oxygen in the sample because spectral difference was not observed on deaeration with N2. A GBC model 903 flame atomic absorption spectrometer and a HP 4500 ICP/MS spectrometer were also used to determine Cu(II) in real samples. To adjust the pH, a Bantex model 300A digital pH meter with a combined calomel and glass electrode was used.
2.3. Calibration cur6e After the aliquots of Cu(II) stock solution were each placed in several 10 ml volumetric flasks for the concentration of Cu(II) ranging from 1.0× l0 − 7 to 1.0× l0 − 5 M, 5.0 ml of pH 8.0 borate buffer solution and 1.0 ml of 5.0× l0 − 4 M Tiron were added, filling deionized water to the mark. These solutions were heated up to 80°C in waterbath for 90 min, cooled down to room temperature and used to measure fluorescence intensity at 294 nm of excitation wavelength and at 350 nm of fluorescence wavelength.
2.4. Reco6ery yields and determinations of Cu (II) in the diluted brass samples and the stream water
2.1. Reagents and solution The 4,5-dihydroxy-1,3-benzenedisulfonic acid (Tiron), manufactured by Aldrich Co., was used without further purification and with appropriate concentration whenever necessary. Standard Cu(II) stock solution was made 1.0× 10 − 3 M by preparing from copper(II) nitrate (Aldrich Co.). Buffer solution of pH 8.0 was prepared by pH meter with 0.025 M sodium tetraborate and 0.1 M HCl. All chemicals used were of analytical and reagent grade, while the deionized water by a Barnstead catridge deionization system (Barnstead Co.) was used throughout all experimental procedures.
A 0.500 g brass sample cleaned by acetone and deionized water was taken into the 250 ml beaker, added by 10 ml of 6M HNO3, heated in fumehood to dissolve the brass sample completely, cooled down to the room temperature, and diluted to 1000 ml in a volumetric flask. Plus, 1.0 ml of this brass solution was transferred to a 1000 ml volumetric flask, and diluted by filling to the mark to get 5.0× 10 − 5% (w/v) brass sample. Suwon stream water was used after the suspended matters or particles were filtered out by glass filter (1-G-4). To investigate the recovery yield of Cu(II) in a diluted brass sample, three 3.0 ml brass samples of 5.0×10 − 5% were transferred to
H.-S. Kim, H.-S. Choi / Talanta 55 (2001) 163–169
one of each 10 ml volumetric flask, and then 0.0, 1.0 and 1.5 ml of 1.0×10 − 5 M Cu(II) standard solution were added to each flask, followed by the procedures which are same as calibration curve procedures. For stream water, the experimental procedures were same as those of diluted brass samples.
3. Results and discussion
3.1. Excitation and fluorescence spectra of Tiron The excitation spectra and the fluorescence spectra of 5.0×10 − 5 M Tiron in various concentrations of Cu(II) are shown in Fig. 1. It was found that fluorescence intensity of excitation (uex = 294 nm) and fluorescence (uf =350 nm) spectra decreased, as the concentration of Cu(II) increased. It was thought to be because static quenching occurred quantitatively in a given concentration range of Cu(II) as Cu(II)– Tiron complex was formed. It was also observed under experiment that fluorescence intensity of Tiron fluctuated in some degree (R.S.D.=2.0%), causing large relative errors in determination of ana-
165
lytes. However, the fluctuation of fluorescence intensity of Tiron decreased by 10 times (R.S.D.=0.2%) when Cu(II) was contained in Tiron solution. While both fluorescence intensities of Tiron only and Tiron in Cu(II) solution decreased slowly in time, the difference between these fluorescence intensities remained constant, which meant that the decreased amount of fluorescence intensity might be equal if measured at any time. However, it was necessary that fluorescence should be measured after waiting for a while because fluorescence intensity somewhat fluctuated if its fluorescence was measured as soon as the Tiron solution was prepared.
3.2. Concentration of Tiron Fig. 2 shows how fluorescence intensity of Tiron changed by the concentration of Tiron in a given concentration range of Cu(II). Since the curve with the steepest decrease of fluorescence intensity by a given concentration range of Cu(II), that is, the highest value of slope, represented the highest sensitivity, it was investigated what concentration of Tiron had the highest slope. When the concentration of Tiron was 5.0×10 − 5 M, its
Fig. 1. Excitation and emission spectra of Tiron (5.0 ×10 − 5 M) as the concentration of Cu2 + . Concentrations of Cu2 + from top spectrum were 0.0, 5.0 ×l0 − 7, 1.0× 10 − 6, 3.0× 10 − 6, 6.0× 10 − 6 and 1.0 × 10 − 5 M, respectively.
166
H.-S. Kim, H.-S. Choi / Talanta 55 (2001) 163–169
Fig. 2. Effect of the concentration of Tiron on the fluorescence quenching curves of Tiron in the presence of Cu(II).
slope was the highest, and its linearity was good (R 2 =0.9971). It was inferred that at lower concentrations of Tiron the slope was gentle because Tiron was not quantitatively complexed in a given concentration range of Cu(II). Also, it was supposed that at higher concentrations of Tiron the relative magnitude of quenching effect by Cu(II) compared with the lower concentration of Tiron decreased as the traction of Tiron that could be complexed with Cu(II) was lower.
3.3. pH Since various buffer solutions, i.e. different electrolytes were used to adjust the pH, the excitation wavelength and the fluorescence wavelength might change. Particularly, fluorescence intensity could be altered because the Tiron complex-forming ability in Cu(II) solution was dependent on pH. So, pH should be controlled to be optimum value. It was investigated how fluorescence intensity of Tiron in a given concentration of Cu(II) solution would change. The steepest slope that represented the highest decreasing effect of fluorescence intensity was shown at pH 8.0. At lower pH, as the oxygen atom in chelating site of Tiron had more affinity power with proton at
higher concentration of proton, Tiron could not play a role as ligand well to complex with Cu(II). At higher pH, as Cu(II)–Tiron complexation was competed with copper hydroxide precipitation, the slower slope was shown.
3.4. Temperature As temperature is an important factor in Cu(II)–Tiron complex formation, it was investigated at room temperature how the fluorescence intensity of Tiron in a given concentration range of Cu(II) decreased when heating to various temperatures during 90 min, and its results were shown in Fig. 3. In temperature experiment, Cu(II)–Tiron complex was formed more quantitatively as the temperature was higher, e.g. at 80°C.
3.5. Heating time Tiron and Cu(II) should be heated to form a stable complex quantitatively and rapidly at a given temperature for appropriate time. Fig. 4 shows that the fluorescence intensity of Tiron decreased in a given concentration range of Cu(II)
H.-S. Kim, H.-S. Choi / Talanta 55 (2001) 163–169
at 80°C as heating time increased. The highest slope was obtained at 120 min of heating time with its very poor linearity (R 2 =0.9800). While its slope at 90 min of heating time was somewhat slower than at 120 min, its linearity was very good
167
(R 2 = 0.9956). At heating time less than 90 min, Cu(II)–Tiron complex was not formed quantitatively, and at more than 90 min, it was supposed that Cu(II)–Tiron complex might be unstable and dissociate.
Fig. 3. Fluorescence quenching curves of Tiron in Cu(II) solution at various temperatures.
Fig. 4. Effect of heating time on the fluorescence quenching curves of Tiron in the presence of Cu(II).
H.-S. Kim, H.-S. Choi / Talanta 55 (2001) 163–169
168
Table 1 Tolerance limitsa of interfering species for the determination of Cu(II) (1.0×10−6 M) in 5.0×10−5 M Tiron solution Mole ratiob
Interfering species
500
− − 2+ NH+ , F−, SO2− 4 , NO3 , Mg 4 , HCO3 − + + 2− − Cl , Na , K , HPO4 , I , NH3
250 100 50 30 10 5 3
Ca2+ Al3+, Ni2+
micellar medium’. Despite the increased slope in 0.1% Tween 80 medium, it was difficult to use since the fluorescence wavelength of Tiron was overlapped by that of Tween 80. It was supposed that its enhanced slope was due to self-quenching of Tween 80. Thus, it was known that the more efficient decreasing effect of fluorescence intensity of Tiron in micellar medium could not occur.
3.7. Interference Fe3+ Cd2+, Zn2+ Pb2+ Co2+
a
Tolerance limit is the maximum concentration in which there is less than 3% effect on fluorescence intensity. b Mole ratio of interfering species to Cu(II).
3.6. Surfactants Although it was expected that the quenching effects decreased by external conversion at high viscosity of micellar medium, it might also be expected that the static quenching effect increased by more stable complex forming in micellar medium. To obtain more efficient quenching effect, various surfactants, such as cetyltrimethylammonium bromide as cationic, sodium dodecylsulfate as anionic, and Triton X-100 and Tween 80 as nonionic surfactants were used. Comparison of the decreasing magnitudes of fluorescence intensity showed less difference for two cases of ‘in no micellar medium’ and ‘in
The interfering effects were investigated which might be caused by other metal ions that can form good complex with Tiron such as Fe(III), Al(III), Cd(II), Co(II), Ni(II), Pb(II), and Zn(II), various species like I− and NH3 that can combine with Cu(II),the analyte ion, and probable species 2+ + such as NO− , Na+, Cl−, and Ca2 + 3 , NH4 , Mg that can exist in real samples. To observe the interfering effect on the determination of Cu(II), the fluorescence intensity of Tiron was measured after the probable interfering ions were added to 1.0×l0 − 6 M Cu(II) and 5.0×10 − 5 M Tiron solution by 1, 3, 5, 10, 30, 50, 100, 250 and 500 times mole of Cu(II). Tolerance limits on these species were investigated on the fluorescence intensity of 5.0× 10 − 5 M Tiron solution containing 1.0×10 − 6 M Cu(II) at maximum amount corresponding to 3% relative error, and were shown in Table 1. Co(II) interfered with, but the interfering effect by Co(II) was completely eliminated by adding sodium oxalate 20 times moles of Tiron.
Table 2 Analytical data of Cu(II) in real samples Real samples
Stream water
Brass sample
a b
Spiked (M)
0.00 1.00×10−6 1.50×10−6 0.00 1.00×10−6 1.50×10−6
Measured (M)
Recovery (%)
Other methodsa
R.S.D. (%)
This methodb
R.S.D. (%)
1.68×10−6
2.3
1.65×10−6 2.68×10−6 3.12×10−6 5.18×10−6 6.15×10−6 6.62×10−5
2.8 2.7 3.2 3.6 3.2 2.9
5.22×10−6
1.5
ICP/MS data on stream waters and FAAS data on brass samples. These values were averaged from seven stream water and seven brass samples.
103 98 97 96
H.-S. Kim, H.-S. Choi / Talanta 55 (2001) 163–169
3.8. Application to brass sample and stream water A calibration curve was constructed at optimum conditions according to experimental procedure in Section 2.3. The linear response range of Cu(II) was 5.0×l0 − 7M – 1.0 ×l0 − 5 M. The correlation coefficient of 0.9948 showed that this calibration curve had good linearity. The detection limit [25] was 3.83(90.09) ×10 − 7 M. With a given amount of Cu(II) spiked in diluted brass sample and stream water, recovery yields were determined using this calibration curve, listed in Table 2, and were found to be nearly 100%. Cu(II) in diluted brass sample and stream water was also determined by flame atomic absorption spectrometry and ICP/MS spectrometry. Results from these techniques were also included in Table 2 and at the 95% confidence level, no difference between results from the proposed and reference methods had been established. Therefore, this proposed technique could be applied to the determination of Cu(II) in real samples.
References [1] R.D. Bautista, A.I. Jimie´ nez, F. Jime´ nez, J.J. Arias, Talanta 43 (1996) 421. [2] S. Nalini, T.V. Ramakrishna, Talanta 43 (1996) 1437. [3] F.H. Hernandez, J.M. Escriche, Analyst 109 (1984) 1585. [4] T. Taketatsu, Talanta 29 (1982) 397. [5] F.H. Hernandez, J.M. Escriche, M.T.G. Andieu, Talanta
.
169
33 (1986) 537. [6] P.R. Haddad, P.W. Alexander, L.E. Smyihe, Talanta 23 (1976) 275. [7] P.C. Ioannou, E.S. Lianidou, D.G. Konstantianos, Anal. Chim. Acta 300 (1995) 237. [8] C.C. Blanco, A.M.G. Campana, F.A. Barrero, M.R. Ceba, Talanta 42 (1995) 1037. [9] A. Pal, N.R. Jana, T.K. Sau, M. Bandyopadhyay, T. Pal, Anal. Comm. 33 (1996) 315. [10] Z. Jiang, X. Liu, A. Liang, Anal. Chim. Acta 346 (1997) 249. [11] J.H. Ayala, A.M. Afonso, V. Gonza´ lez-Diaz, Microchem. J. 60 (1998) 101. [12] A.L. Hunt, J.F. Alder, Anal. Chim. Acta 387 (1999) 207. [13] T.M.A. Razek, S. Spear, S.S.M. Hassen, M.A. Arnold, Talanta 48 (1999) 269. [14] K Cha, K. Park, Talanta 46 (1998) 1567. [15] N. Jie, Q. Zhang, Y. Yang, X. Huang, Talanta 46 (1998) 215. [16] B. Juskowiak, Anal. Chim. Acta 320 (1996) 115. [17] N. Jie, D. Yang, Q. Jiang, Q. Zhang, L. Wei, Microchem. J. 62 (1999) 371. [18] C.Q. Ma, K.A. Li, S.Y. Tong, Anal. Chim. Acta 333 (1996) 83. [19] J. Migdalski, T. Blaz, A. Lewenstam, Anal. Chim. Acta 322 (1996) 141. [20] T. Fujimoto, N. Teshima, M. Kurhara, S. Nakano, T. Kawashima, Talanta 49 (1999) 1091. [21] J. Gao, X. Zhang, W. Yang, B. Zhao, J. Hon, J. Kang, Talanta 51 (2000) 447. [22] M. Busch, A. Seubert, Anal. Chim. Acta 399 (1999) 223. [23] J.F. van Staden, H. du Plessis, R.E. Taljaard, Anal. Chim. Acta 357 (1997) 141. [24] Y. Zhao, D. Wang, X. Guo, J. Xu, Anal. Chim. Acta 353 (1997) 329. [25] D.A. Skoog, F.J. Holler, T.A. Nieman, Principles of Instrumental Analysis, 5th edn, Saunders College Publishing, 1998, p. 13.