A novel electrochemical microsensor for the determination of NO and its application to the study of the NO donor S-nitrosoglutathione

Abstract

A novel electrochemical microsensor for the determination of NO based on an electropolymerized film of tetraaminophthalocyaninecopper [Cu(TAPc)] was prepared. Its response to NO and its application to the study of an NO donor (S-nitrosoglutathione; GSNO) are also described. The microsensor exhibited an electrocatalytic effect on NO oxidation and showed a low detection limit, high sensitivity and selectivity for NO determination. The oxidation current (measured by differential pulse amperometry) was linear for NO concentrations ranging from 6.2 × 10⁻⁹ to 3.0 × 10⁻⁵ mol L⁻¹ with a calculated detection limit of 4.0 ×10 ⁻⁹ mol L⁻¹ (S/N = 3) and a linear coefficient of 0.9984. Some endogenous electroactive substances in biological tissues, such as dopamine, 5-hydroxytryptamine and nitrite, at concentrations higher than those in biological systems did not interfere with NO determination. The sensor shows promise for the possible in vivo determination of NO. Using the microsensor, the NO release from the NO donor (GSNO) was successfully monitored. This work sets a foundation for the study of the pharmacology and the biological effects in vivo of S-nitrosothiols.

---

Introduction

Nitric oxide as a new vasodilator messenger has been identified as the endothelium-derived relaxing factor (EDRF),¹ being responsible for a number of diseases and a variety of physiological processes. For example, it can play key roles in cellular communication including the central and peripheral nervous systems and in host defense mechanism of eukaryotes.^2,3 Owing to its important physiological roles, nitric oxide has received extensive attention in the fields of life science and related areas.^4,5

There is currently much interest in the chemistry of S-nitrosothiols (RSNO), because of their possible therapeutic use as vasodilators (by release of nitric oxide) and they could also be involved in vivo in some of the remarkable physiological processes brought about and controlled by nitric oxide^6–8 such as inhibition of platelet aggression,⁹ vasodilation¹⁰ and the inhibition of neutrophil functions.¹¹ There is a very real case for the generation of alternative drugs given the tolerance problem associated in many cases with the widespread use of glyceryl trinitrate for the treatment of angina and other circulation problems which relies on their ability to generate NO in vivo. The medical importance of S-nitrosothiols has been highlighted recently in two reports,^12,13 which describe the clinical use of S-nitrosoglutatione (GSNO) to inhibit platelet aggregation during coronary angioplasty and also to treat a form of pre-eclampsia, a high blood pressure condition suffered by some pregnant women.

These findings could well have implications for the reaction of S-nitrosothiols in vivo. Under appropriate conditions these compounds decompose to liberate nitric oxide and the corresponding disulfide. Intense interest in the release of nitric oxide from S-nitrosothiols has been generated recently, not only in view of their potential use therapeutically as alternative NO-releasing drugs, but also with regard to their possible involvement in vivo as potential NO-storage and transport vehicles.

Owing to the great importance of the research mentioned above, there has been an explosive growth in the measurement of NO in biological systems. However, this is extremely difficult because of its low concentration, high activity and fleeting presence. Most of the techniques for NO measurement are indirect, based on chemical detection of the oxidation products removed from biological systems.¹⁴ Other methods, such as chemiluminescence and electron paramagnetic resonance (EPR), can determine NO directly, but they cannot be applied to monitor NO in vivo. Electrochemical techniques are promising because of their inherent sensitivity and speed, and they are especially suitable for the development of analytical approaches for NO in real time. Several types of effective electrochemical sensors have been reported. Shibuki¹⁵ reported a sensor based on chloraprene membranes. Malinski and co-workers¹⁶ demonstrated electrodes modified with an electropolymerized film of metalloporphyrins. Some chemically modified electrodes were based on the electrocatalytic reduction of NO.¹⁷ The aim of our work was to provide a new NO microsensor with experimental calibration data for concentrations below 1 μmol L⁻¹ and corresponding to the expected concentration of NO in the biological media of interest.

NO, like other simple diatomic molecules, such as O₂ and CO, has been reported not only to react with hemoproteins and metalloenzymes, but also to interact with non-heme metal complexes to form its adducts.^18–20 Moreover, NO is an extremely powerful ligand, giving binding constants to metal ions free or in complex form often greatly in excess of those of CO and almost always much higher than those of O₂. This indicates that some interaction occurs between NO and certain metal complexes which is similar to that between simple diatomic molecules such as O₂, CO and certain metal complexes to form their adducts.^21,22

Bioinorganic structures have been demonstrated to be able potentially to mimic metalloenzyme structure and functions. Comparative studies of the functions in both synthetic models and natural systems led to the conclusions that the synthetic models can mimic the activity of cytochrome P-450 families in the case of biomimetic oxidation of hydrocarbons and transportation of oxygen,^23–25 and that metalloporphyrins, metallophthalocyanines and metal Schiff bases can be used as biomimetic oxidation catalysts and oxygen carriers such as so-called inorganic enzymes.²⁶ This suggests an effective approach in developing sensors for biological compounds.

Metalloporphyrins have been reported as the best candidates for the construction of NO sensors^27–30 Recently we have developed several microsensors for the direct determination of NO based on an electropolymerized film of metal Schiff base.^31–33 It appeared that metal Schiff-base complexes could be exploited as a new kind of material for the design of NO electrochemical sensors. Metallophthalocyanines as macrocyclic complexes have been exploited extensively in electrocatalysis. In this work we attempted to prepare a microsensor modified with an electropolymerized film of tetraaminophthalocyaninecopper(II) [Cu(TAPc)]. The microsensor was found to display good electrocatalytic activity toward NO oxidation with high sensitivity and selectivity and a low detection limit of 4.0 ×10⁻⁹ mol L⁻¹. Using this microsensor, we successfully determined in situ NO release from S-nitrosoglutathione (GSNO). All of this work should be promising in applications to the in vivo measurement of NO and the study of the pharmacology and the biological effects in vivo of S-nitrosothiols.

Experimental

Chemicals

NO saturated solution was obtained by bubbling NO gas through deoxygenated distilled water for 30 min, using a value of 1.9 mmol L⁻¹ for its concentration at saturation.³⁴ NO standard solutions were prepared by making serial dilutions of the NO saturated solution. NO standard solutions could also be obtained with the NO producing system developed in our laboratory.³⁵ The prepared NO solutions were kept in a glass flask with a rubber septum and stored in the dark and were stable for 3 h.

Tetrabutylammonium perchlorate (TBAP) was prepared by the reaction of tetrabutylammonium bromide with sodium perchlorate. It was recrystallized with acetyl acetate (m.p. 212.5–213.5 °C). Nafion (5% solution in methanol) was purchased from Aldrich (Milwaukee, WI, USA). Phosphate-buffered saline (PBS) containing 137 mmol L⁻¹ NaCl, 2.7 mmol L⁻¹ KCl, 8.0 mmol L⁻¹ Na₂HPO₄ and 1.5 mmol L⁻¹ KH₂PO₄ was prepared and adjusted to pH 7.4. Other chemicals were of at least reagent grade quality and used as received. Aqueous solutions were prepared with doubly distilled water.

Synthesis of tetraaminophthalocyaninecopper(II)

A 3.0 g amount of copper chloride pentahydrate, 9.5 g of 4-nitrophthalic acid, 1.1 g of ammonium chloride, 0.2 g of ammonium molybdate and excess urea (15 g) were finely ground and placed in a 500 mL three-necked flask containing 25 mL of nitrobenzene. The reaction mixture was stirred at 185 °C for 4.5 h. The solid product was washed with alcohol until free from nitrobenzene. The product was treated twice with 100 mL of 1.0 mol L⁻¹ hydrochloric acid and 1.0 mol L⁻¹ sodium hydroxide. The chloride free of 4,9,16,23-tetranitrophthalocyaninecopper(II) was dried at 125 °C.

About 5 g of finely ground 4,9,16,23-tetranitrophthalocyaninecopper(II) were placed in 100 mL of water. To this slurry 100 g of Na₂S·9H₂O were added and stirred at 50 °C for 5 h. The separated product was treated with 750 mL of 1.0 mol L⁻¹ hydrochloric acid and then treated with 500 mL of 1.0 mol L⁻¹ sodium hydroxide solution, stirred for 1 h and centrifuged to separate the dark green, solid complex. The product was repeatedly treated with water, stirred and centrifuged until the material was free from sodium hydroxide and sodium chloride. The pure copper complex was dried in a vacuum. The structure was characterized by its FTIR spectra data (KBr, cm⁻¹): 3281, 3183(γ_NH2), 1345, 1258, 1060, 1090 ( γ_C–N), 826, 868 (δ_Ar), 735, 752, 950, 1607 (δ_N–H).

Synthesis of S-nitrosoglutathione (GNSO)

To a stirred, ice-cold solution of glutathione (0.76 g, 2.5 mmol) in water (4 mL) containing 2 mol L⁻¹ HCl (1.5 mL), sodium nitrite (0.18 g, 2.5 mmol) was added in a single portion. After 40 min at 5 °C, the red solution was treated with acetone (5 mL) and stirred for a further 10 min. The resulting fine, pale red precipitate was filtered off and then washed successively with ice-cold water, acetone and diethyl ether to give S-nitrosoglutathione.

Apparatus

Electrochemical experiments were performed with a three-electrode system. A platinum disc microelectrode prepared by sealing a platinum wire (diameter 200 μm) into a glass capillary served as the working electrode. A saturated calomel electrode (SCE) was used as the reference electrode and a gold electrode as the auxiliary electrode. Electrochemical experiments were performed with a CHI832 Electrochemical Analyzer (Austin, TX, USA) in conjunction with an IBM-compatible Pentium-133 computer. Data and voltammograms were recorded and stored in the computer. During NO determination, a hermetic electrochemical cell was employed to ensure no NO leakage and to avoid O₂ permeation because NO is very active and could be oxidized in the presence of O₂.

Preparation of the microsensor

The microelectrode was thoroughly sonicated with acetone, NaOH (1 + 1), HNO₃ (1 + 1) and distilled water. Its electrochemical pre-treatment was performed by scanning the potential ranging from −0.30 to +1.20 V for 10 cycles in 0.50 mol L⁻¹ sulfuric acid. The microelectrode was allowed to air-dry and placed in dimethyl sulfoxide (DMSO) solution containing 5.0 × 10⁻³ mol L⁻¹ monomer of Cu(TAPc) and 0.1 mol L⁻¹ TBAP as supporting electrolyte. Cu(TAPc) was deposited on the microelectrode surface by scanning the potential between −0.20 and +0.90 V at a scan rate of 100 mV s⁻¹. After consecutively cycling for 70 cycles, the electrode was removed from the electroploymerization solution, rinsed with acetone and distilled water and allowed to air-dry. The modified microelectrode was further coated with Nafion twice by depositing 1 μl of 1% w/v Nafion solution and then placing the electrode under an infrared lamp to allow the methanol to evaporate.

Electrochemical determination of NO

Prior to NO determination, the microsensor was placed in PBS solution, and cyclic voltammetry was performed in the potential range between 0.00 and +1.00 V until steady cyclic voltammetric (typically after about 10 cycles) and differential pulse voltammetric (DPV) responses were obtained. Electrochemical responses of the microsensor to NO and selectivity tests were evaluated by employing the DPV method. DPV was performed at a sweep rate of 5 mV s⁻¹ and five pulses s⁻¹ (pulse height 50 mV, pulse width 60 ms). Calibration of the microsensor and determination of NO released from GSNO were performed using the differential pulse amperometric (DPA) technique. The electrode was cleaned at 0.00 V for 1 s, pulsed to +0.70 V for 50 ms and then pulsed to +0.80 V for 50 ms from +0.70 V. The current was measured as the current change between the values at +0.70 and +0.80 V. Aliquots of NO solution were subsequently added with a gas-tight syringe, and the current response due to NO oxidation was recorded for each addition.

Results and discussion

Electropolymerization of Cu(TAPc) at microelelctrode

Unlike those macrocyclic complexes reported as effective molecular material for the design of NO sensors^23–26 electropolymerization of Cu(TAPc) was carried out in DMSO solution owing to its insolubility in aqueous solution. The Cu(TAPc) was polymerized by the consecutive cyclic voltammetry in DMSO solution containing 5.0 × 10⁻³ mol L⁻¹ Cu(TAPc) and 0.1 mol L⁻¹ TBAP at a platinum microelectrode. In the potential window between −0.20 and +0.90 V vs. SCE, an ill-defined anodic peak appeared during the first scan which was related to the oxidation of the tetraaminophthalocyanine ring. As a consequence of the repeated potential scans, two new pairs of well-defined peaks appeared at +0.76 and +0.50 V. The pair of peaks at +0.76 V was assigned to the redox activity of the tetraaminophthalocyanine which resulted in the electropolymerization of Cu(TAPc) at the microelectrode. The other anodic peak observed at a formal potential of +0.50 V could be assigned to the redox process of the central ion Cu⁺ to Cu²⁺.

It was observed that if the potential was more negative than +0.70 V during continuous cycling, there was no indication of electropolymerization. With continuous scan cyclic voltammetry, gradual increases in the amplitude of these peaks were observed. After 70 cycles of scanning, the microelectrode was transferred with thorough rinsing to DMSO solution only containing 0.1 mol L⁻¹ TBAP. The cyclic voltammagram with this microelectrode exhibited the same electrochemical behavior as the electropolymerization of Cu(TAPc). This indicated the electropolymerization of Cu(TAPc) on the microelectrode surface.

Catalytic oxidation of NO

Fig. 1 demonstrates the sensitivity of the developed microelectrode studied by comparing the responses of NO with the same concentration at microelectrodes modified with Nafion and poly[Cu(TAPc)]–Nafion. As can be seen, the DPV response of NO at the microelectrode modified with Nafion exhibited a small and broad peak. However, a sharp peak at a minor negative-shifted potential could be observed at the microsensor modified with poly[Cu(TAPc)]–Nafion. This indicated that the microsensor based on a poly[Cu(TAPc)] film showed good catalytic activity for NO oxidation.

Fig. 1 Differential pulse voltammograms of microelectrodes modified with (a) Nafion and (b) poly[Cu(TAPc)]–Nafion in deoxygenated PBS solution containing NO at a concentration of 1.5 × 10⁻⁵ mol L⁻¹.

Selectivity of the microsensor for NO

The purpose of the NO microsensor is to carry out in vivo measurements of NO. Therefore, the constructed microsensor should be free from interference from other electroactive substances co-existing in biological fluids, such as ascorbate, uric acid, some neurotransmitters and their metabolites. Nitrite, besides the interference mentioned above, can cause a serious interference in NO measurement because it is one of the main metabolites of NO and can be oxidized at nearly the same potential as NO at the microsensor. In addition, for in vivo analysis, the electrode should have a high resistance to fouling caused by non-specific absorption of macromolecules. Our previous work has shown that Nafion, a cation-exchange film, can not only prevent the diffusion of anions, such as ascorbate, nitrite and metabolites of some neurotransmitters, to the electrode, but also minimize the electrode fouling. Furthermore, a Nafion film could also stabilize NO⁺ formed upon oxidation of NO, and prevent a complicated pattern of reactions that could possibly lead to the formation of NO₂⁻ and NO₃⁻. We therefore further coated the poly[Cu(TAPc)]-modified microelectrode with Nafion to enhance the selectivity and avoid fouling.

Fig. 2 shows a typical DPV of the microsensor based on poly[Cu(TAPc)]–Nafion in the presence of NO at various concentrations and 1.0 × 10⁻⁵ mol L⁻¹ of nitrite. As can be seen the microsensor could effectively prevent the interference from nitrite. We also found that ascorbate and uric acid at a concentration of 1.0 × 10⁻⁴ mol L⁻¹, and 5-hydroxyindole-3-acetic acid and 3,4-dihydroxyphenylacetic acid at a concentration of 1.0 × 10⁻⁵ mol L⁻¹ showed no responses at the microsensors, suggesting that these co-existing substances did not interfere with NO determination using DPV or DPA. However, dopamine, epinephrine and 5-hydroxytryptamine had voltammetric peaks at a typical potential of about +0.30 V at concentrations higher than 1.0 × 10⁻⁶ mol L⁻¹, suggesting that the microsensor also showed responses to these substances, but they did not interfere with the NO determination at the microsensor using DPV and especially DPA.

Fig. 2 Differential pulse voltammograms of the microsensor based on poly[Cu(TAPc)]–Nafion in deoxygenated PBS solution containing 1.0 × 10⁻⁵ mol L⁻¹ NO₂⁻ and NO at concentrations of (a) 0.0, (b) 2.0 × 10⁻⁶, (c) 4.0 × 10⁻⁶ and (d) 6.0 × 10⁻⁶ mol L⁻¹.

Linear range, detection limit and lifetime of the microsensor

The NO concentration varies between 10⁻⁶ and 10⁻⁹ mol L⁻¹ in biological systems. We found that it was difficult to calibrate the microsensor when the NO concentration was below 0.1 μmol L⁻¹ using a DPV. Furthermore, it is more important to measure the relative change in chemical substances while studying their physiological roles. Therefore, when the NO concentration ranges from 1 nmol L⁻¹ to 0.1 μmol L⁻¹, the microsensor should be calibrated using DPA, an active method. Fig. 3 shows the DPA of the poly[Cu(TAPc)]–Nafion modified microelectrode in deoxygenated PBS solution with successive additions of NO. The microsensor clearly showed substantial increase in the oxidation current upon successive additions of NO. The steady-state oxidation currents were linear with NO concentrations in the range 6.2 × 10⁻⁹–3.0 × 10⁻⁵ mol L⁻¹ with a calculated detection limit of 4.0 × 10⁻⁹ mol L⁻¹ (based on a signal-to-noise ratio of 3) and a linear coefficient of 0.9984 (I = −0.27[NO] −2.232 × 10⁻⁸). Fig. 3 also illustrates the high reproducibility of the microsensor. The response time of the microsensor was about 300 ms (n = 14 measurements). The sensitivity of the microsensor showed no observable change after 1 month of storage in PBS solution at 4 °C or successive potential cycling from 0.00 to 1.00 V for 5 h, indicating that the film of poly[Cu(TAPc)]–Nafion was stable and that the microsensor developed in this work could have a long shelf-life.

Fig. 3 Differential pulse amperograms of the microsensor based on poly[Cu(TAPc)]–Nafion in deoxygenated PBS solution with successive additions of NO at concentrations of (a) 1.0 × 10⁻⁸ and (b) 2.0 × 10⁻⁸ mol L⁻¹.

NO release from NO donor GSNO

It is known that GSNO can decompose photochemically and thermally to give nitric oxide and the corresponding disulfide:In this reaction, it is likely that homolysis of the S–N bond is the primary process. In aqueous solution and in the absence of heat and light, many investigations have indicated that the same overall reaction occurs. In the presence of oxygen the final product is nitrate, but in the absence of oxygen, nitric oxide can be detected.

In this work, we used the poly[Cu(TAPc)]–Nafion microsensor to detect nitric oxide release from GSNO. Fig. 4 demonstrates the DPA responses to NO release from GSNO in deoxygenated and incubated PBS in the presence of 1.0 × 10⁻⁴ mol L⁻¹ Cu⁺ with successive addition of 5.0 × 10⁻⁵ mol L⁻¹ GSNO at 37 °C. This amperogram clearly shows an increase in the measured current with successive additions of GSNO. Fig. 5 shows that the addition of Cu⁺ to GSNO in deoxygenated PBS at 37 °C resulted in the rapid release of NO. We found that the highest concentration of NO release from GSNO observed within about 100 s was 7.0 × 10⁻⁶ mol L⁻¹ at a GSNO concentration of 1.0 × 10⁻³ mol L⁻¹. We also found that under the same experimental conditions without the effect of Cu⁺, GSNO did not release NO for more than 2 h. These results indicate that the decomposition of GSNO occurs by a Cu⁺-catalyzed reaction pathway. The precise mechanism of Cu⁺-induced decomposition of S-nitrosothiols is unknown. It is likely that the chelation of Cu⁺ by thiol disulfides may complicate the kinetics of the processes.

Fig. 4 Differential pulse amperograms of the microsensor based on poly[Cu(TAPc)]–Nafion in deoxygenated PBS solution incubated at 37 °C with successive additions of 5.0 × 10⁻⁵ mol L⁻¹ GSNO in the presence of 1 × 10⁻⁴ mol L⁻¹ Cu⁺.

Fig. 5 Release of NO during Cu⁺-catalyzed decomposition of 1.0 × 10⁻³ mol L⁻¹ GSNO in deoxygenated PBS at 37 °C in the presence of 1.0 × 10⁻⁴ mol L⁻¹ Cu⁺.

Conclusion

A novel electrochemical microsensor for NO determination has been developed. The microsensor displayed high sensitivity to NO with a low detection limit of 4.0 × 10⁻⁹ mol L⁻¹ and was free from interference from some endogenous electroactive substances such as ascorbate and nitrite co-existing in biological systems. The results indicate the possible applicability of the microsensor to transient and localized NO assay in biological systems. Using this microsensor we monitored NO release from the NO donor GSNO. GSNO could decompose by a Cu⁺-catalyzed reaction pathway and had a high release efficiency. This work sets a foundation for the study of the pharmacology and the biological effects of S-nitrosothiols in vivo.

Acknowledgements

This work was supported by the Deptartment of Science and Technology of China and the Shanghai Science and Technology Committee.

References

1. P. M. J. Palmer, A. G. Ferrige and S. Moncada, Nature (London), 1987, 327, 524.
2. E. Southam and J. Garthwaite, Neurochemistry, 1991, 2, 658.
3. J. B. Hibbs, Z. Wavrin and R. R. Tainror, J. Immunol., 1987, 138, 550.
4. T. Malinski, M. W. Radomski, Z. Taha and S. Moncada, Biochem. Biophys. Res. Commun., 1993, 194, 960.
5. M. Hoshino, R. Konishi, N. Tezuka, I. Ueno and H. Seki, J. Phys. Chem., 1996, 100, 13569.
6. S. Moncada, R. M. J. Palmer and E. A. Higgs, Pharmacol. Rev., 1991, 43, 109.
7. A. R. Butler and D. L. H. Williams, Chem. Soc. Rev., 1993, 22, 233.
8. P. L. Feldman, O. W. Griffith and D. J. Stuehr, Chem. Eng. News, 1993, December 20, 26..
9. L. J. Ignarro, H. Lipton, J. C. Edwards, W. H. Baricos, A. L. Hyman, P. J. Kadowiyz and C. A. Gruetter, J. Pharmacol. Exp. Ther., 1981, 218, 739.
10. M. W. Radomski, D. D. Rees, A. Dutra and S. Moncada, Br. J. Pharmacol., 1992, 107, 745.
11. R. M. Clancey and S. B. Abramson, Anal. Biochem., 1992, 204, 365.
12. E. J. Langford, A. S. Brown, R. J. Wainwright, A. J. de Belder, M. R. Thomas, R. E. A. Smith, M. W. Radomski, J. F. Martin and S. Moncada, Lancet, 1994, 344, 1458.
13. A. de Belder, C. Lees, J. Martin, S. Moncada and S. Campbell, Lancet, 1995, 345, 124.
14. L. C. Green, D. A. Wanger, J. Glogowski, D. L. Skipper, J. S. Wishnok and S. R. Tannenbaum, Anal. Biochem., 1982, 126, 131.
15. K. Shibuki, Neurosci. Res., 1990, 9, 69.
16. (a) T. Malinski and Z. Taha, Nature (London), 1992, 358, 676; (b) T. Malinski, Z. Taha, S. Grunfeld, A. Burewicz, P. Tomboulian and F. Kiechele, Anal. Chim. Acta, 1993, 279, 135.
17. M. Maskus, F. Pariente, Q. Wu, A. Toffanin, J. P. Shapleigh and H. D. Abruna, Anal. Chem., 1996, 68, 3128.
18. H. Sugimori, H. Uchida, T. Akiyama, M. Mukaida and K. Shimizu, Chem. Lett., 1982, 28, 1135.
19. M. Kubota, M. K. Chen, C. D. Boyd and K. R. Mann, Inorg. Chem., 1987, 26, 3261.
20. H. Ikezawa, E. Miki, K. Mizumachi, T. Ishimori, T. Nagai and M. Tanaka, Bull. Chem. Soc. Jpn., 1993, 66, 89.
21. D. C. Detschman and S. G. Utterback, J. Am. Chem. Soc., 1981, 103, 2847.
22. W. S. Bringar and C. K. Chang, J. Am. Chem. Soc., 1974, 96, 5595.
23. R. E. White and M. J. Coon, Annu. Rev. Biochem., 1980, 49, 315.
24. I. C. Gunsalus and S. C. Sligar, Adv. Enzymol., 1978, 47, 1.
25. D. Mansay and P. Battioni, in Bioinorganic Catalysis, ed. J. Reedijk, Marcel Dekker, New York, 1993, p. 395..
26. F. Bedioui, Coord. Chem. Rev., 1995, 144, 39.
27. T. Malinskied and Z. Taha, Nature (London), 1992, 358, 676.
28. T. Malinski, S. Grunfeld, A. Burewicz and P. Tomboulian, Anal. Chim. Acta., 1993, 279, 135.
29. S. Trevin, F. Bedioui and J. Devynck, Talanta, 1996, 43, 303.
30. S. Trevin, F. Bedioui and J. Devynck, J. Electroanal. Chem., 1996, 408, 261.
31. L. Mao, Y. Tian, H. Lian and L. Jin, Chin. J. Anal. Chem., 1998, 6, 689.
32. L. Mao, Y. Tian, G. Shi, H. Liu, L. Jin, K. Yamamoto, S. Tao and J. Jin, Anal. Lett., 1998, 31, 1991.
33. H. P. Tu, L. Q. Mao, X. N. Cao and L. T. Jin, Electroanalysis, 1999, 11, 69.
34. W. Gerrard, Gas Solubilities Widespread Applications, Pergamon Press, Oxford, 1980..
35. L. Mao, Y. Tian, Y. Xian and L. Jin, China Pat., 97 206334, 1997..

---