- Email: [email protected]
Lanthanum-selective membrane electrode based on 2,2′-dithiodipyridine
Analytica Chimica Acta 531 (2005) 179–184
Lanthanum-selective membrane electrode based on 2,2-dithiodipyridine Morteza Akhond∗ , Mohammad Bagher Najafi, Javad Tashkhourian Department of Chemistry, Shiraz University, Shiraz 71454, Islamic Republic of Iran Received 21 January 2003; received in revised form 6 August 2004; accepted 24 September 2004 Available online 2 December 2004
Abstract In this study, a new poly(vinyl chloride) (PVC) membrane sensor for La3+ ion based on 2,2 -dithiodipyridine as an ion carrier was prepared. This electrode revealed good selectivity for La3+ over a wide variety of other metal ions. Effects of experimental parameters such as membrane composition, nature and amount of plasticizer, the amount of additive and concentration of internal solution on the potential response of La3+ sensor were investigated. The electrode exhibited a Nernstian slope of 20.0 ± 1.0 mV per decade of La3+ over a concentration range of 7.1 × 10−6 to 2.2 × 10−2 M of La3+ in the pH range 3.3–8.0. The response time was about 7 s and the detection limit was 3.1 × 10−6 M. The electrode can be used for at least 2 months without a considerable divergence in potential. The proposed electrode was used as an indicator electrode in potentiometric titration of oxalate and fluoride ions and was applied for determination of F− ion in mouthwash solution. © 2004 Elsevier B.V. All rights reserved. Keywords: Lanthanum ion-selective electrode; 2,2 -Dithiodipyridine; PVC membrane; Potentiometry; Fluoride determination
1. Introduction The utility of ion-selective electrodes (ISEs) is being increasingly realized by analytical chemists as they represent a rapid, accurate and low-cost method of analysis. Moreover, analysis by ISEs is non-destructive and adaptable to small sample volumes. Also, they have an extremely wide range of applications and are one of the few techniques, which can be used for determination of both cations and anions in aqueous solutions [1]. Lanthanides are widely distributed in low concentrations throughout the earth’s crust [2]. Because of the increasing interest in bioinorganic and coordination chemistry and increased industrial use of lanthanum compounds as well as their enhanced discharge and toxic properties, the determination of these species has been an increasing concern. Many methods described for assay of lanthanum and related metals in mixtures, require separation steps, which are time consuming and introduce lower accuracy and higher costs. ∗
Corresponding author. Tel.: +98 711 222 7601; fax: +98 711 222 0027. E-mail address: [email protected] (M. Akhond).
0003-2670/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2004.09.089
However, some of these methods are not sufficiently selective and sensitive for trace determination of lanthanum [3,4]. A few ion-selective electrodes have also been developed for the potentiometric determination of La3+ [5–10]. However, these efforts have not been very fruitful as the developed electrodes possess narrow working concentration ranges and suffer serious interference from various cations including Cu2+ , Ni2+ and Ce3+ . It is well known that sulfur-containing ligands coordinate selectively with transition and heavy metal ions [11]. In this respect, macrocyclic polythiaethers [12–17] and dithiocarbamate derivatives [18,19] have received considerable attention, due to their unique properties as carriers for some transition and heavy metal ions. In this paper, 2,2 -dithiodipyridine was used as a neutral ion carrier in the construction of a lanthanum–poly(vinyl chloride) (PVC) membrane electrode. This sensor has a simple design, a fast response time and a Nernstian slope and shows fairly good discriminating ability towards La3+ ion in comparison with some alkali, alkaline earth and heavy metal ions.
180
M. Akhond et al. / Analytica Chimica Acta 531 (2005) 179–184
Fig. 1. Chemical structure of 2, 2 -dithiodipyridine.
2. Experimental 2.1. Reagents 2,2 -Dithiodipyridine (Fig. 1), sodium tetraphenylborate (NaTPB), high relative molecular weight poly(vinyl chloride), tetrahydrofuran (THF), o-nitrophenyloctyl ether (NPOE), dibuthyl phthalate (DBP), benzyl acetate (BA), dimethyl sebacate (DMS) and oleic acid (OA) were purchased from Merck and used as received. Nitrate salts of all cations were used (all from Merck). Sodium fluoride mouthwash solution was obtained from Shahre Daru Co. (Tehran, Iran). Triply distilled deionized water was used throughout. 2.2. Preparation of PVC membrane The general procedure for preparing the PVC membrane was to mix 2.4 mg 2,2 -dithiodipyridine, 30.2 mg powdered PVC, 6.8 mg OA and 60.6 mg o-NPOE as a solvent mediator. The mixture was then thoroughly dissolved in 2 ml of THF. The resulting mixture was transferred into a glass dish of 2 cm diameter. The solvent was slowly evaporated until a concentrated oily mixture was obtained. A Pyrex tube of 5 mm internal diameter was dipped into the mixture for about 5 s. After the tube was removed from the mixture, it was kept at room temperature for about 3 h so that a non-transparent membrane of about 0.3 mm thickness was formed. Then, it was filled with 1.0 × 10−3 M La(NO3 )3 as an internal filling solution. The electrode was finally conditioned for 20 h by soaking in 1.0 × 10−3 M lanthanum nitrate. A silver/silver chloride-coated wire was used as an internal reference electrode. 2.3. Electrode potential measurement All emf measurements were carried out with the following cell assembly: Ag–AgCl|internal solution (1.0 × 10−3 M La (NO3 )2 )|PVC membrane|test solution|Hg–Hg2 Cl2 , KCl (saturated). A HIOKI Digital Hitester (model 3256-01) as a potentiometer was used for potential measurement at 25 ◦ C. The emf observations were made versus a double-junction saturated calomel electrode (SCE, Philips) with the chamber filled with a potassium nitrate solution. A Corning 130 pH meter was used for pH measurement at 25 ◦ C.
Fig. 2. The potential responses of various ISEs based on 2,2 dithiodipyridine.
3. Results and discussion In preliminary experiments, 2,2 -dithiodipyridine was used as a neutral carrier to prepare PVC membrane ionselective electrode for a wide variety of metal ions, including alkali, alkaline earth, transition and heavy metal ions. The potential responses of various ISEs based on 2,2 dithiodipyridine were obtained separately for each ion and the results are shown in Fig. 2. As it is seen in this figure, the lanthanum ISE has shown the most sensitive response, which indicates that the membrane electrode based on 2,2 dithiodipyridine could be suitable for determination of La3+ . 3.1. Electrode optimization It is well known that the sensitivity, linearity and selectivity obtained for a given ionophore depends significantly on the membrane composition [20–25]. Since the nature of the plasticizer influences the dielectric constant of the membrane phase as well as the mobility of the ionophore molecule and its complexes [26], it is expected to play a fundamental role in specifying the selective electrode characteristics. So, several solvent mediators such as o-NPOE, DMS, DBP and BA, which are often used in PVC-membrane electrodes, were evaluated and the results are summarized in the Table 1. Among these plasticizers, o-NPOE provided faster, more stable and sensitive response in the concentration range of 7.1 × 10−6 to 2.2 × 10−2 M of La3+ ion. In general, the thickness and hardness of the membrane depend upon the amount of PVC used. At higher PVC content, the membrane becomes too dense and this makes the transport of cations into the membrane more difficult. At lower PVC content, the membrane becomes mechanically weak and swells up easily in aqueous solution. The plasticizers/PVC ratio of 2 is most frequently used in ISEs for suitable
M. Akhond et al. / Analytica Chimica Acta 531 (2005) 179–184 Table 1 Optimization of membrane ingredients No.
PVC (mg)
o-NPOE Ionophore (mg) (mg)
Additive (mg)
Slope (mV/decade)
1 2 3 4 5 6 7 8 9 10 11 12
33.0 32.6 32.5 32.4 31.7 31.0 30.2 29.2 30.2 30.0 30.2 30.2
66.0 65.9 65.1 64.6 63.4 62.2 60.6 58.5 60.6 59.6 60.6 60.6
– – – – 2.5 OA 4.4 OA 6.8 OA 10.0 OA 6.8 NaTPB 8.0 NaTPB 4.0 OA + 2.8 NaTPB 6.8 OA
9.0 11.0 14.0 12.0 16.0 18.0 19.7 Unstable 16.0 13.0 10.0 ∼0
1.0 1.5 2.4 3.0 2.4 2.4 2.4 2.4 2.4 2.4 2.4 –
application; therefore this ratio was applied in construction of electrodes. It is well known that the presence of lipophilic anionic sites in a cation-selective membrane electrode not only improves the response behavior and selectivity of electrode [27–30], but also increases the sensitivity of the membrane [31], especially where the extraction capability is poor. The use of ionic additives such as different tetraphenylborate salts and its more lipophilic derivative, tetrakis (p-chlorophenyl) borate (K-TCPB) also fatty acids such as oleic acid [15,32] as lipophilic additives is widely reported in the preparation of different ion-selective electrodes. In this study the effect of NaTPB and oleic acid as an additive on the response of membrane were investigated. From the data given in Table 1, it is immediately obvious that the nature and amount of additive influences the performance characteristics of the membrane sensor significantly. As shown in this table, OA is a more suitable additive than NaTPB. The electrode based on this additive, the membrane no. 7 with PVC:o-NPOE:ionophore:OA percent ratio of 30.2:60.6:2.4:6.8, has been shown a Nernstian slope and better concentration range. Oleic acid is not only primarily a phase-transfer catalyst but also contributes to the complexation mechanism, as described by Eugster et al. [33]. Furthermore, it has no disadvantages cited concern with NaTPB [29]. Moreover, oleic acid is expected to contribute significantly to the dielectric constant of the membrane in addition to the plasticizer. The effect of relative amount of 2,2 -dithiodipyridine on the response function of membrane was investigated (Table 1). 2.4 mg of ionophore was chosen as the optimum amount of ionophore in construction of the PVC membrane electrode. Further addition of ionophore, however, resulted in some decreases in the response of the electrode, most probably due to some inhomogeneities and possible saturation of the membrane [13] and may also induce strong interactions between polymeric chains and ionophore-preventing mobility of the segments as explained by Hall considering experimental observations of Reinhoudt and co-workers [34]. The concentration of the internal solution in the electrode was changed from 1.0 × 10−1 to 1.0 × 10−3 M and the
181
potential response of the La3+ -selective membrane electrode was obtained. It was found that variation of the concentration of the internal solution does not cause any significant difference in the potential response, except for an expected change in the intercept of the resulting Nernstian plots. A 1.0 × 10−3 M concentration of the reference solution is quite appropriate for smooth functioning of the electrode system. Optimum conditioning time for the membrane sensor in a 1.0 × 10−3 M La3+ solution was obtained to be 20 h. Then, the electrode generates stable potentials when placed in contact with La3+ solution. The response time of the electrode was tested by measuring the time required to achieve a 90% of the steady potential when the concentration of La(NO3 )3 solution was rapidly increased by one decade from 1.0 × 10−4 to 1.0 × 10−3 M. The static response time thus obtained was less than 7 s over the entire concentration range, and potentials stayed constant for about 5 min, after which only a very slow divergence was recorded. The sensing behavior of the membrane remained unchanged when the potentials were recorded either from high to low concentrations or vice versa. The reproducibility of the electrode was examined using six similarly constructed electrodes under the optimum conditions. The results showed good reproducibility for the proposed electrode. The long-term stability of the electrode was studied by periodically re-calibrating in standard solutions and calculating the response slope. The slope of the electrode response was reproducible over a period of at least 2 months. Therefore, the proposed electrode could be used for 2 months without any considerable change in its response characteristics towards La3+ . The pH dependence of the electrode potential was investigated over the pH range of 2.0–10.0 in a 1.0 × 10−4 M solution of La3+ ion. As shown in Fig. 3, the potential was independent of pH in the range of 3.3–8.0 pH unit. At higher pH values, the potential decreased due to the formation of lanthanum hydroxide in solution; and at lower pH values, the potential increased, indicating that the electrode also responds to hydrogen ion.
Fig. 3. The effect of pH on the response of La3+ membrane electrode.
182
M. Akhond et al. / Analytica Chimica Acta 531 (2005) 179–184
Table 2 Selectivity coefficients of various interfering ions for the La3+ ISE Cations
Present work 1.4 × 10−3
Cu2+
1.8 × 10−3 7.3 × 10−3 <1.0 × 10−6 <1.0 × 10−6 6.2 × 10−4 3.1 × 10−4 <1.0 × 10−6 <1.0 × 10−6 <1.0 × 10−6 <1.0 × 10−6 <1.0 × 10−6 1.6 × 10−3 1.1 × 10−3 <1.0 × 10−6 <1.0 × 10−6 <1.0 × 10−6 2.0 × 10−3 3.1 × 10−4
Ce3+ Pb2+ Mg2+ Mn2+ Cd2+ Co2+ Ni2+ Sr2+ Na+ Zn2+ Rb+ Al3+ Fe3+ NH4+ Ba2+ Cs+ Ag+ Cr3+
Ref. [9]
Ref. [8]
Ref. [6]
Ref. [5]
7.1 × 10−3
8.30 × 10−2
– – – – – – 2.5 × 10−2 2.8 × 10−2 – – – – 6.4 × 10−5 – – – – – –
– – – 0.96 – – – – – 30.0 – – – – 75.6 – – – –
3.5 × 10−2 2.7 × 10−4 1.7 × 10−4 – 6.7 × 10−4 5.1 × 10−3 5.6 × 10−3 1.3 × 10−4 2.9 × 10−4 3.5 × 10−3 – – – – 1.3 × 10−4 – 7.7 × 10−3 –
3.2. Analytical performance The potential response of the optimized electrode, to varying concentration of La3+ ions, was examined. The calibration plot indicated a linear range of 7.1 × 10−6 to 2.2 × 10−2 M with a Nernstian slope of 20.0 ± 1.0 mV per decade of La3+ ion activity. The practical limit of detection was 3.1 × 10−6 M as determined from the intersection of the two extrapolated segments of the calibration graph based on recommended procedure by IUPAC [35]. The selectivity is obviously one of the most important characteristics of an ion-selective electrode, determining whether a reliable measurement in the target sample is possible. To investigate the selectivity of the proposed membrane electrode, its potential responses were investigated in the presence of various interfering foreign cations using the matched potential method (MPM) [36]. This is a recently recommended procedure by IUPAC [36], to overcome the limitation of Nicolski–Eisenmann equation for the determination of potentiometric selectivity coefficients. These limitations include non-Nernstian behavior of interfering ions, inequality of charges of primary and pot interfering ions, and activity dependence of KA,B According to the MPM, the selectivity coefficient is defined as activity ratio of primary ion and the interfering ion that gives the same potential changes in a reference solution. Thus, one should measure the changes in potentials upon changing the primary ion activities (aA ). Then, the interfering ion would be added to an identical reference solution until the same potential change upon changing the interfering ion activity is MPM determined obtained (aB ). The selectivity coefficients KA,B as: MPM KA,B =
aA aB
– 5.90 × 10−2 5.90 × 10−2 – – 4.00 × 10−2 5.90 × 10−2 – – 4.20 × 10−2 – – – – 1.27 × 10−1 – – –
where aA and aB are the change in the primary ion activity and the selected interfering ion activity, respectively. pot The resulting KA,B , values thus obtained for the proposed La3+ -selective electrode are summarized in Table 2. As seen, the alkali, alkaline earth and transition metal ions do not significantly disturb the functioning of the proposed La3+ ionselective membrane electrode. The selectivity coefficients for previously reported lanthanum-selective electrodes are also included in Table 2 for comparison.
4. Analytical applications 4.1. Determination of lanthanum in different samples The sensor was examined for the determination of La3+ in the presence of various cations. The results in Table 3 show Table 3 Determination of lanthanum ions in ternary mixtures La3+ (mol/L) Added cations (mol/L)
Found (mol/L)
Recovery (%)
1 2.0 × 10−4
3.0 × 10−3 Ca2+ 2.0 × 10−3 Cd2+
(1.97 ± 0.10) × 10−4
98.5
2 2.0 × 10−4
2.0 × 10−3 Zn2+ 3.0 × 10−3 Na+
(1.90 ± 0.06) × 10−4
95.0
3 2.0 × 10−4
3.0 × 10−3 Cs+ 2.0 × 10−3 Hg2+
(2.10 ± 0.05) × 10−4 105.0
4 2.0 × 10−4
2.0 × 10−3 Ce3+ (2.10 ± 0.10) × 10−4 105.0 3.0 × 10−3 NH4 +
5 2.0 × 10−4
1.0 × 10−3 Ag+ 1.0 × 10−3 Cu2+
(2.20 ± 0.04) × 10−4 110.0
6 2.0 × 10−4
2.0 × 10−3 Sr2+ 1.0 × 10−4 Fe3+
(2.10 ± 0.06) × 10−4 105.0
M. Akhond et al. / Analytica Chimica Acta 531 (2005) 179–184 Table 4 Results of determination of lanthanum spiked in tap and river water samples Sample
Added (mol/L)
Found (mol/L)
Recovery (%)
1
Tap water
2.0 × 10−3 4.0 × 10−4
(2.10 ± 0.05) × 10−3 (4.15 ± 0.10) × 10−4
105.0 103.7
2
River water
2.0 × 10−3 4.0 × 10−4
(2.18 ± 0.04) × 10−3 (4.27 ± 0.12) × 10−4
109.0 106.8
that, the recovery of lanthanum ion (n = 4) is quantitative in the presence of foreign ions in the ternary mixture of cations. The proposed sensor was also successfully applied to the direct determination of La(III) ions in tap water and river water samples. Different spiked samples were prepared by adding aliquots of La3+ solution into river water and tap water and the amount of spiked La3+ in samples was directly determined as is shown in Table 4. As seen the recovery (n = 4) of La3+ at various lanthanum concentrations is quantitative. 4.2. Determination of fluoride and oxalate ions in synthetic and pharmaceutical samples The proposed La3+ -selective electrode was found to work well under laboratory conditions. It was used as an indicator electrode in potentiometric titration of 20.0 ml oxalate ion solution (1.0 × 10−4 M) and also 20 ml fluoride ion solution (1.0 × 10−3 M) with a 0.01 and 0.1 M La3+ solution, respectively. The resulting titration curve for fluoride, as example, is shown in Fig. 4. The exact amount of oxalate and fluoride ions were then evaluated from the sharp inflection points of their resulting titration curves and the amount of each anion in solution was accurately determined with the proposed electrode. The electrode was also successfully applied to the determination of F− ion in a pharmaceutical sample. Sodium fluoride mouthwash solution Shahre Daru Co. (Tehran, Iran) was chosen as a pharmaceutical sample. An appropriate volume of the sodium fluoride mouthwash solution was cho-
Fig. 4. Potentiometric titration curve of 20 ml 1.0 × 10−3 M fluoride with 0.1 M La3+ solution.
183
sen (20.0 ml of the 0.2% solution). The solution was then titrated (n = 3) with a 0.1 M of lanthanum nitrate. The results showed that there was a satisfactory agreement between the declared fluoride content (0.20% (labeled)) and the obtained value (0.2.0 ± 0.01%).
5. Conclusions Based on the results discussed in this paper, 2,2 dithiodipyridine could be considered as a suitable neutral ionophore for construction of a PVC-based membraneselective electrode for the direct determination of La(III) ion in solution. The results showed that oleic acid is a suitable lipophilic additive for the electrode construction. The proposed electrode showed high selectivity and sensitivity to La3+ ion, wide dynamic range, low detection limit and fast response time. The proposed electrode reveals excellent selectivity for La3+ over a wide variety of alkali, alkaline earth, transitions and heavy metal ions. The lanthanumselective electrode was found to work well under laboratory conditions. It was used in direct determination of La3+ in ternary mixture of cations, tap water and river water and also as an indicator electrode in potentiometric titration of oxalate and fluoride ions with La3+ solution. This electrode could be also successfully applied to the determination of F− ion in mouthwash solution.
Acknowledgements We would like to acknowledge the support of this work by the Shiraz University Research Council. We would also like to acknowledge Dr. G. Absalan for carefully reading the manuscript and for his helpful suggestions.
References [1] C.C. Rundle, Beginners Guide to ISE Measurement, Nico 2000 Ltd., London, UK, 2000. [2] R.E. Krik, D.F. Othmer, Encyclopedia of Chemical Technology, vol. 19, Wiley, New York, 1982, p. 836. [3] P. Voldet, W. Haerdi, Anal. Chim. Acta 72 (1974) 111. [4] J.G. Crock, F.E. Lichte, Anal. Chem. 54 (1982) 1329. [5] J.B. Harrell, A.D. Jones, G.R. Choppin, Anal. Chem. 41 (1969) 1459. [6] W. Szczepaniak, M. Ren, Talanta 41 (1994) 1393. [7] E. Pan, Z. Huang, Z. Xuemei, Yingyong HuaXie, 11 (1994) 20. [8] S. Amarchand, S.K. Menon, Y.K. Agrawal, Electroanalysis 12 (2000) 522. [9] M. Shamsipur, M. Yousefi, M. Hosseini, M.R. Ganjali, Anal. Chem. 74 (2002) 5538. [10] V.K. Gupta, S. Jain, S. Chandra, Anal. Chim. Acta 486 (2003) 199. [11] R.M. Izatt, K. Pawlak, J.S. Bradshaw, R.L. Bruening, Chem. Rev. 91 (1991) 1721. [12] D. Siswanta, K. Nagatsuka, H. Yamada, K. Kumakura, H. Hisamoto, Y. Shichi, K. Toshima, K. Suzuki, Anal. Chem. 68 (1996) 4166.
184
M. Akhond et al. / Analytica Chimica Acta 531 (2005) 179–184
[13] X. Yang, N. Kumar, H. Chi, D.D. Hibbert, P.N.W. Alexander, Electroanalysis 9 (1997) 549. [14] V.K. Gupta, S. Jain, U. Khrana, Electroanalysis 9 (1997) 478. [15] A.R. Fakhari, M.R. Ganjali, M. Shamsipur, Anal. Chem. 99 (1997) 3693. [16] M. Javanbakht, M.R. Ganjali, H. Eshghi, H. Sharghi, M. Shamsipur, Electroanalysis 11 (1999) 81. [17] M.H. Mashhadizadeh, M. Shamsipur, Anal. Chim. Acta 381 (1999) 111. [18] S. Kamata, H. Murata, Y. Kubo, A. Bhale, Analyst 114 (1989) 1029. [19] S. Kamata, K. Onoyama, Anal. Chem. 63 (1991) 1295. [20] S.K. Srivastava, V.K. Gupta, S. Jain, Electroanalysis 8 (1996) 938. [21] R.J.W. Lugtenberg, R.J.M. Egberink, J.F.J. Engbersen, D.N. Reinhoudt, J. Chem. Soc. Perkin Trans. 2 (1997) 1353. [22] V.K. Gupta, P. Kumar, Anal. Chim. Acta 389 (1999) 205. [23] M. Javanbakht, A. Shabani-Kia, M.R. Darvich, M.R. Ganjali, M. Shamsipur, Anal. Chim. Acta 408 (2000) 75. [24] V.K. Gupta, P. Kumar, R. Mangla, Electroanalysis 12 (2000) 752. [25] M. Javanbakht, M.R. Ganjali, H. Eshghi, H. Sharghi, M. Shamsipur, Electroanalysis 11 (1999) 81.
[26] H. Parham, M. Shamsipur, J. Membr. Sci. 86 (1994) 29. [27] E. Bakker, P. B¨uhlmann, E. Pretsch, Chem. Rev. 97 (1997) 3083. [28] R. Eugster, P.M. Gehrig, W.E. Morf, U.E. Spichiger, W. Simon, Anal. Chem. 63 (1991) 2285. [29] T. Rosatzin, E. Bakker, K. Suzuki, W. Simon, Anal. Chim. Acta 280 (1993) 197. [30] U. Schaller, E. Bakker, U.E. Spichiger, E. Pretsch, Anal. Chem. 66 (1994) 391. [31] D. Ammann, W.E. Morf, P. Anken, P.C. Meier, E. Pretsch, W. Simon, Ion-Sel. Electrode Rev. 5 (1983) 3. [32] M. Shamsipur, M.R. Ganjali, A. Moghimi, Anal. Chem. 70 (402) (1998) 5259. [33] R. Eugster, U.E. Spichiger, W. Simon, Anal. Chem. 65 (1993) 689. [34] P.L.H.M. Cobben, R.J.M. Egberink, J.B. Bomer, P. Bergveld, W. Verboom, D.N. Reinhoudt, J. Am. Chem. Soc. 114 (1992) 10573. [35] G.G. Guilbault, R.A. Durst, M.S. Frant, H. Freiser, E.H. Hansen, T.S. Light, E. Pungor, G. Rechnitz, N.M. Rice, T.J. Rohm, W. Simon, J.D.R. Thomas, Pure Appl. Chem. 48 (1976) 127. [36] Y. Umezawa, K. Umezawa, H. Sato, Pure Appl. Chem. 67 (1995) 507.
Lanthanum-selective membrane electrode based on 2,2-dithiodipyridine Morteza Akhond∗ , Mohammad Bagher Najafi, Javad Tashkhourian Department of Chemistry, Shiraz University, Shiraz 71454, Islamic Republic of Iran Received 21 January 2003; received in revised form 6 August 2004; accepted 24 September 2004 Available online 2 December 2004
Abstract In this study, a new poly(vinyl chloride) (PVC) membrane sensor for La3+ ion based on 2,2 -dithiodipyridine as an ion carrier was prepared. This electrode revealed good selectivity for La3+ over a wide variety of other metal ions. Effects of experimental parameters such as membrane composition, nature and amount of plasticizer, the amount of additive and concentration of internal solution on the potential response of La3+ sensor were investigated. The electrode exhibited a Nernstian slope of 20.0 ± 1.0 mV per decade of La3+ over a concentration range of 7.1 × 10−6 to 2.2 × 10−2 M of La3+ in the pH range 3.3–8.0. The response time was about 7 s and the detection limit was 3.1 × 10−6 M. The electrode can be used for at least 2 months without a considerable divergence in potential. The proposed electrode was used as an indicator electrode in potentiometric titration of oxalate and fluoride ions and was applied for determination of F− ion in mouthwash solution. © 2004 Elsevier B.V. All rights reserved. Keywords: Lanthanum ion-selective electrode; 2,2 -Dithiodipyridine; PVC membrane; Potentiometry; Fluoride determination
1. Introduction The utility of ion-selective electrodes (ISEs) is being increasingly realized by analytical chemists as they represent a rapid, accurate and low-cost method of analysis. Moreover, analysis by ISEs is non-destructive and adaptable to small sample volumes. Also, they have an extremely wide range of applications and are one of the few techniques, which can be used for determination of both cations and anions in aqueous solutions [1]. Lanthanides are widely distributed in low concentrations throughout the earth’s crust [2]. Because of the increasing interest in bioinorganic and coordination chemistry and increased industrial use of lanthanum compounds as well as their enhanced discharge and toxic properties, the determination of these species has been an increasing concern. Many methods described for assay of lanthanum and related metals in mixtures, require separation steps, which are time consuming and introduce lower accuracy and higher costs. ∗
Corresponding author. Tel.: +98 711 222 7601; fax: +98 711 222 0027. E-mail address: [email protected] (M. Akhond).
0003-2670/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2004.09.089
However, some of these methods are not sufficiently selective and sensitive for trace determination of lanthanum [3,4]. A few ion-selective electrodes have also been developed for the potentiometric determination of La3+ [5–10]. However, these efforts have not been very fruitful as the developed electrodes possess narrow working concentration ranges and suffer serious interference from various cations including Cu2+ , Ni2+ and Ce3+ . It is well known that sulfur-containing ligands coordinate selectively with transition and heavy metal ions [11]. In this respect, macrocyclic polythiaethers [12–17] and dithiocarbamate derivatives [18,19] have received considerable attention, due to their unique properties as carriers for some transition and heavy metal ions. In this paper, 2,2 -dithiodipyridine was used as a neutral ion carrier in the construction of a lanthanum–poly(vinyl chloride) (PVC) membrane electrode. This sensor has a simple design, a fast response time and a Nernstian slope and shows fairly good discriminating ability towards La3+ ion in comparison with some alkali, alkaline earth and heavy metal ions.
180
M. Akhond et al. / Analytica Chimica Acta 531 (2005) 179–184
Fig. 1. Chemical structure of 2, 2 -dithiodipyridine.
2. Experimental 2.1. Reagents 2,2 -Dithiodipyridine (Fig. 1), sodium tetraphenylborate (NaTPB), high relative molecular weight poly(vinyl chloride), tetrahydrofuran (THF), o-nitrophenyloctyl ether (NPOE), dibuthyl phthalate (DBP), benzyl acetate (BA), dimethyl sebacate (DMS) and oleic acid (OA) were purchased from Merck and used as received. Nitrate salts of all cations were used (all from Merck). Sodium fluoride mouthwash solution was obtained from Shahre Daru Co. (Tehran, Iran). Triply distilled deionized water was used throughout. 2.2. Preparation of PVC membrane The general procedure for preparing the PVC membrane was to mix 2.4 mg 2,2 -dithiodipyridine, 30.2 mg powdered PVC, 6.8 mg OA and 60.6 mg o-NPOE as a solvent mediator. The mixture was then thoroughly dissolved in 2 ml of THF. The resulting mixture was transferred into a glass dish of 2 cm diameter. The solvent was slowly evaporated until a concentrated oily mixture was obtained. A Pyrex tube of 5 mm internal diameter was dipped into the mixture for about 5 s. After the tube was removed from the mixture, it was kept at room temperature for about 3 h so that a non-transparent membrane of about 0.3 mm thickness was formed. Then, it was filled with 1.0 × 10−3 M La(NO3 )3 as an internal filling solution. The electrode was finally conditioned for 20 h by soaking in 1.0 × 10−3 M lanthanum nitrate. A silver/silver chloride-coated wire was used as an internal reference electrode. 2.3. Electrode potential measurement All emf measurements were carried out with the following cell assembly: Ag–AgCl|internal solution (1.0 × 10−3 M La (NO3 )2 )|PVC membrane|test solution|Hg–Hg2 Cl2 , KCl (saturated). A HIOKI Digital Hitester (model 3256-01) as a potentiometer was used for potential measurement at 25 ◦ C. The emf observations were made versus a double-junction saturated calomel electrode (SCE, Philips) with the chamber filled with a potassium nitrate solution. A Corning 130 pH meter was used for pH measurement at 25 ◦ C.
Fig. 2. The potential responses of various ISEs based on 2,2 dithiodipyridine.
3. Results and discussion In preliminary experiments, 2,2 -dithiodipyridine was used as a neutral carrier to prepare PVC membrane ionselective electrode for a wide variety of metal ions, including alkali, alkaline earth, transition and heavy metal ions. The potential responses of various ISEs based on 2,2 dithiodipyridine were obtained separately for each ion and the results are shown in Fig. 2. As it is seen in this figure, the lanthanum ISE has shown the most sensitive response, which indicates that the membrane electrode based on 2,2 dithiodipyridine could be suitable for determination of La3+ . 3.1. Electrode optimization It is well known that the sensitivity, linearity and selectivity obtained for a given ionophore depends significantly on the membrane composition [20–25]. Since the nature of the plasticizer influences the dielectric constant of the membrane phase as well as the mobility of the ionophore molecule and its complexes [26], it is expected to play a fundamental role in specifying the selective electrode characteristics. So, several solvent mediators such as o-NPOE, DMS, DBP and BA, which are often used in PVC-membrane electrodes, were evaluated and the results are summarized in the Table 1. Among these plasticizers, o-NPOE provided faster, more stable and sensitive response in the concentration range of 7.1 × 10−6 to 2.2 × 10−2 M of La3+ ion. In general, the thickness and hardness of the membrane depend upon the amount of PVC used. At higher PVC content, the membrane becomes too dense and this makes the transport of cations into the membrane more difficult. At lower PVC content, the membrane becomes mechanically weak and swells up easily in aqueous solution. The plasticizers/PVC ratio of 2 is most frequently used in ISEs for suitable
M. Akhond et al. / Analytica Chimica Acta 531 (2005) 179–184 Table 1 Optimization of membrane ingredients No.
PVC (mg)
o-NPOE Ionophore (mg) (mg)
Additive (mg)
Slope (mV/decade)
1 2 3 4 5 6 7 8 9 10 11 12
33.0 32.6 32.5 32.4 31.7 31.0 30.2 29.2 30.2 30.0 30.2 30.2
66.0 65.9 65.1 64.6 63.4 62.2 60.6 58.5 60.6 59.6 60.6 60.6
– – – – 2.5 OA 4.4 OA 6.8 OA 10.0 OA 6.8 NaTPB 8.0 NaTPB 4.0 OA + 2.8 NaTPB 6.8 OA
9.0 11.0 14.0 12.0 16.0 18.0 19.7 Unstable 16.0 13.0 10.0 ∼0
1.0 1.5 2.4 3.0 2.4 2.4 2.4 2.4 2.4 2.4 2.4 –
application; therefore this ratio was applied in construction of electrodes. It is well known that the presence of lipophilic anionic sites in a cation-selective membrane electrode not only improves the response behavior and selectivity of electrode [27–30], but also increases the sensitivity of the membrane [31], especially where the extraction capability is poor. The use of ionic additives such as different tetraphenylborate salts and its more lipophilic derivative, tetrakis (p-chlorophenyl) borate (K-TCPB) also fatty acids such as oleic acid [15,32] as lipophilic additives is widely reported in the preparation of different ion-selective electrodes. In this study the effect of NaTPB and oleic acid as an additive on the response of membrane were investigated. From the data given in Table 1, it is immediately obvious that the nature and amount of additive influences the performance characteristics of the membrane sensor significantly. As shown in this table, OA is a more suitable additive than NaTPB. The electrode based on this additive, the membrane no. 7 with PVC:o-NPOE:ionophore:OA percent ratio of 30.2:60.6:2.4:6.8, has been shown a Nernstian slope and better concentration range. Oleic acid is not only primarily a phase-transfer catalyst but also contributes to the complexation mechanism, as described by Eugster et al. [33]. Furthermore, it has no disadvantages cited concern with NaTPB [29]. Moreover, oleic acid is expected to contribute significantly to the dielectric constant of the membrane in addition to the plasticizer. The effect of relative amount of 2,2 -dithiodipyridine on the response function of membrane was investigated (Table 1). 2.4 mg of ionophore was chosen as the optimum amount of ionophore in construction of the PVC membrane electrode. Further addition of ionophore, however, resulted in some decreases in the response of the electrode, most probably due to some inhomogeneities and possible saturation of the membrane [13] and may also induce strong interactions between polymeric chains and ionophore-preventing mobility of the segments as explained by Hall considering experimental observations of Reinhoudt and co-workers [34]. The concentration of the internal solution in the electrode was changed from 1.0 × 10−1 to 1.0 × 10−3 M and the
181
potential response of the La3+ -selective membrane electrode was obtained. It was found that variation of the concentration of the internal solution does not cause any significant difference in the potential response, except for an expected change in the intercept of the resulting Nernstian plots. A 1.0 × 10−3 M concentration of the reference solution is quite appropriate for smooth functioning of the electrode system. Optimum conditioning time for the membrane sensor in a 1.0 × 10−3 M La3+ solution was obtained to be 20 h. Then, the electrode generates stable potentials when placed in contact with La3+ solution. The response time of the electrode was tested by measuring the time required to achieve a 90% of the steady potential when the concentration of La(NO3 )3 solution was rapidly increased by one decade from 1.0 × 10−4 to 1.0 × 10−3 M. The static response time thus obtained was less than 7 s over the entire concentration range, and potentials stayed constant for about 5 min, after which only a very slow divergence was recorded. The sensing behavior of the membrane remained unchanged when the potentials were recorded either from high to low concentrations or vice versa. The reproducibility of the electrode was examined using six similarly constructed electrodes under the optimum conditions. The results showed good reproducibility for the proposed electrode. The long-term stability of the electrode was studied by periodically re-calibrating in standard solutions and calculating the response slope. The slope of the electrode response was reproducible over a period of at least 2 months. Therefore, the proposed electrode could be used for 2 months without any considerable change in its response characteristics towards La3+ . The pH dependence of the electrode potential was investigated over the pH range of 2.0–10.0 in a 1.0 × 10−4 M solution of La3+ ion. As shown in Fig. 3, the potential was independent of pH in the range of 3.3–8.0 pH unit. At higher pH values, the potential decreased due to the formation of lanthanum hydroxide in solution; and at lower pH values, the potential increased, indicating that the electrode also responds to hydrogen ion.
Fig. 3. The effect of pH on the response of La3+ membrane electrode.
182
M. Akhond et al. / Analytica Chimica Acta 531 (2005) 179–184
Table 2 Selectivity coefficients of various interfering ions for the La3+ ISE Cations
Present work 1.4 × 10−3
Cu2+
1.8 × 10−3 7.3 × 10−3 <1.0 × 10−6 <1.0 × 10−6 6.2 × 10−4 3.1 × 10−4 <1.0 × 10−6 <1.0 × 10−6 <1.0 × 10−6 <1.0 × 10−6 <1.0 × 10−6 1.6 × 10−3 1.1 × 10−3 <1.0 × 10−6 <1.0 × 10−6 <1.0 × 10−6 2.0 × 10−3 3.1 × 10−4
Ce3+ Pb2+ Mg2+ Mn2+ Cd2+ Co2+ Ni2+ Sr2+ Na+ Zn2+ Rb+ Al3+ Fe3+ NH4+ Ba2+ Cs+ Ag+ Cr3+
Ref. [9]
Ref. [8]
Ref. [6]
Ref. [5]
7.1 × 10−3
8.30 × 10−2
– – – – – – 2.5 × 10−2 2.8 × 10−2 – – – – 6.4 × 10−5 – – – – – –
– – – 0.96 – – – – – 30.0 – – – – 75.6 – – – –
3.5 × 10−2 2.7 × 10−4 1.7 × 10−4 – 6.7 × 10−4 5.1 × 10−3 5.6 × 10−3 1.3 × 10−4 2.9 × 10−4 3.5 × 10−3 – – – – 1.3 × 10−4 – 7.7 × 10−3 –
3.2. Analytical performance The potential response of the optimized electrode, to varying concentration of La3+ ions, was examined. The calibration plot indicated a linear range of 7.1 × 10−6 to 2.2 × 10−2 M with a Nernstian slope of 20.0 ± 1.0 mV per decade of La3+ ion activity. The practical limit of detection was 3.1 × 10−6 M as determined from the intersection of the two extrapolated segments of the calibration graph based on recommended procedure by IUPAC [35]. The selectivity is obviously one of the most important characteristics of an ion-selective electrode, determining whether a reliable measurement in the target sample is possible. To investigate the selectivity of the proposed membrane electrode, its potential responses were investigated in the presence of various interfering foreign cations using the matched potential method (MPM) [36]. This is a recently recommended procedure by IUPAC [36], to overcome the limitation of Nicolski–Eisenmann equation for the determination of potentiometric selectivity coefficients. These limitations include non-Nernstian behavior of interfering ions, inequality of charges of primary and pot interfering ions, and activity dependence of KA,B According to the MPM, the selectivity coefficient is defined as activity ratio of primary ion and the interfering ion that gives the same potential changes in a reference solution. Thus, one should measure the changes in potentials upon changing the primary ion activities (aA ). Then, the interfering ion would be added to an identical reference solution until the same potential change upon changing the interfering ion activity is MPM determined obtained (aB ). The selectivity coefficients KA,B as: MPM KA,B =
aA aB
– 5.90 × 10−2 5.90 × 10−2 – – 4.00 × 10−2 5.90 × 10−2 – – 4.20 × 10−2 – – – – 1.27 × 10−1 – – –
where aA and aB are the change in the primary ion activity and the selected interfering ion activity, respectively. pot The resulting KA,B , values thus obtained for the proposed La3+ -selective electrode are summarized in Table 2. As seen, the alkali, alkaline earth and transition metal ions do not significantly disturb the functioning of the proposed La3+ ionselective membrane electrode. The selectivity coefficients for previously reported lanthanum-selective electrodes are also included in Table 2 for comparison.
4. Analytical applications 4.1. Determination of lanthanum in different samples The sensor was examined for the determination of La3+ in the presence of various cations. The results in Table 3 show Table 3 Determination of lanthanum ions in ternary mixtures La3+ (mol/L) Added cations (mol/L)
Found (mol/L)
Recovery (%)
1 2.0 × 10−4
3.0 × 10−3 Ca2+ 2.0 × 10−3 Cd2+
(1.97 ± 0.10) × 10−4
98.5
2 2.0 × 10−4
2.0 × 10−3 Zn2+ 3.0 × 10−3 Na+
(1.90 ± 0.06) × 10−4
95.0
3 2.0 × 10−4
3.0 × 10−3 Cs+ 2.0 × 10−3 Hg2+
(2.10 ± 0.05) × 10−4 105.0
4 2.0 × 10−4
2.0 × 10−3 Ce3+ (2.10 ± 0.10) × 10−4 105.0 3.0 × 10−3 NH4 +
5 2.0 × 10−4
1.0 × 10−3 Ag+ 1.0 × 10−3 Cu2+
(2.20 ± 0.04) × 10−4 110.0
6 2.0 × 10−4
2.0 × 10−3 Sr2+ 1.0 × 10−4 Fe3+
(2.10 ± 0.06) × 10−4 105.0
M. Akhond et al. / Analytica Chimica Acta 531 (2005) 179–184 Table 4 Results of determination of lanthanum spiked in tap and river water samples Sample
Added (mol/L)
Found (mol/L)
Recovery (%)
1
Tap water
2.0 × 10−3 4.0 × 10−4
(2.10 ± 0.05) × 10−3 (4.15 ± 0.10) × 10−4
105.0 103.7
2
River water
2.0 × 10−3 4.0 × 10−4
(2.18 ± 0.04) × 10−3 (4.27 ± 0.12) × 10−4
109.0 106.8
that, the recovery of lanthanum ion (n = 4) is quantitative in the presence of foreign ions in the ternary mixture of cations. The proposed sensor was also successfully applied to the direct determination of La(III) ions in tap water and river water samples. Different spiked samples were prepared by adding aliquots of La3+ solution into river water and tap water and the amount of spiked La3+ in samples was directly determined as is shown in Table 4. As seen the recovery (n = 4) of La3+ at various lanthanum concentrations is quantitative. 4.2. Determination of fluoride and oxalate ions in synthetic and pharmaceutical samples The proposed La3+ -selective electrode was found to work well under laboratory conditions. It was used as an indicator electrode in potentiometric titration of 20.0 ml oxalate ion solution (1.0 × 10−4 M) and also 20 ml fluoride ion solution (1.0 × 10−3 M) with a 0.01 and 0.1 M La3+ solution, respectively. The resulting titration curve for fluoride, as example, is shown in Fig. 4. The exact amount of oxalate and fluoride ions were then evaluated from the sharp inflection points of their resulting titration curves and the amount of each anion in solution was accurately determined with the proposed electrode. The electrode was also successfully applied to the determination of F− ion in a pharmaceutical sample. Sodium fluoride mouthwash solution Shahre Daru Co. (Tehran, Iran) was chosen as a pharmaceutical sample. An appropriate volume of the sodium fluoride mouthwash solution was cho-
Fig. 4. Potentiometric titration curve of 20 ml 1.0 × 10−3 M fluoride with 0.1 M La3+ solution.
183
sen (20.0 ml of the 0.2% solution). The solution was then titrated (n = 3) with a 0.1 M of lanthanum nitrate. The results showed that there was a satisfactory agreement between the declared fluoride content (0.20% (labeled)) and the obtained value (0.2.0 ± 0.01%).
5. Conclusions Based on the results discussed in this paper, 2,2 dithiodipyridine could be considered as a suitable neutral ionophore for construction of a PVC-based membraneselective electrode for the direct determination of La(III) ion in solution. The results showed that oleic acid is a suitable lipophilic additive for the electrode construction. The proposed electrode showed high selectivity and sensitivity to La3+ ion, wide dynamic range, low detection limit and fast response time. The proposed electrode reveals excellent selectivity for La3+ over a wide variety of alkali, alkaline earth, transitions and heavy metal ions. The lanthanumselective electrode was found to work well under laboratory conditions. It was used in direct determination of La3+ in ternary mixture of cations, tap water and river water and also as an indicator electrode in potentiometric titration of oxalate and fluoride ions with La3+ solution. This electrode could be also successfully applied to the determination of F− ion in mouthwash solution.
Acknowledgements We would like to acknowledge the support of this work by the Shiraz University Research Council. We would also like to acknowledge Dr. G. Absalan for carefully reading the manuscript and for his helpful suggestions.
References [1] C.C. Rundle, Beginners Guide to ISE Measurement, Nico 2000 Ltd., London, UK, 2000. [2] R.E. Krik, D.F. Othmer, Encyclopedia of Chemical Technology, vol. 19, Wiley, New York, 1982, p. 836. [3] P. Voldet, W. Haerdi, Anal. Chim. Acta 72 (1974) 111. [4] J.G. Crock, F.E. Lichte, Anal. Chem. 54 (1982) 1329. [5] J.B. Harrell, A.D. Jones, G.R. Choppin, Anal. Chem. 41 (1969) 1459. [6] W. Szczepaniak, M. Ren, Talanta 41 (1994) 1393. [7] E. Pan, Z. Huang, Z. Xuemei, Yingyong HuaXie, 11 (1994) 20. [8] S. Amarchand, S.K. Menon, Y.K. Agrawal, Electroanalysis 12 (2000) 522. [9] M. Shamsipur, M. Yousefi, M. Hosseini, M.R. Ganjali, Anal. Chem. 74 (2002) 5538. [10] V.K. Gupta, S. Jain, S. Chandra, Anal. Chim. Acta 486 (2003) 199. [11] R.M. Izatt, K. Pawlak, J.S. Bradshaw, R.L. Bruening, Chem. Rev. 91 (1991) 1721. [12] D. Siswanta, K. Nagatsuka, H. Yamada, K. Kumakura, H. Hisamoto, Y. Shichi, K. Toshima, K. Suzuki, Anal. Chem. 68 (1996) 4166.
184
M. Akhond et al. / Analytica Chimica Acta 531 (2005) 179–184
[13] X. Yang, N. Kumar, H. Chi, D.D. Hibbert, P.N.W. Alexander, Electroanalysis 9 (1997) 549. [14] V.K. Gupta, S. Jain, U. Khrana, Electroanalysis 9 (1997) 478. [15] A.R. Fakhari, M.R. Ganjali, M. Shamsipur, Anal. Chem. 99 (1997) 3693. [16] M. Javanbakht, M.R. Ganjali, H. Eshghi, H. Sharghi, M. Shamsipur, Electroanalysis 11 (1999) 81. [17] M.H. Mashhadizadeh, M. Shamsipur, Anal. Chim. Acta 381 (1999) 111. [18] S. Kamata, H. Murata, Y. Kubo, A. Bhale, Analyst 114 (1989) 1029. [19] S. Kamata, K. Onoyama, Anal. Chem. 63 (1991) 1295. [20] S.K. Srivastava, V.K. Gupta, S. Jain, Electroanalysis 8 (1996) 938. [21] R.J.W. Lugtenberg, R.J.M. Egberink, J.F.J. Engbersen, D.N. Reinhoudt, J. Chem. Soc. Perkin Trans. 2 (1997) 1353. [22] V.K. Gupta, P. Kumar, Anal. Chim. Acta 389 (1999) 205. [23] M. Javanbakht, A. Shabani-Kia, M.R. Darvich, M.R. Ganjali, M. Shamsipur, Anal. Chim. Acta 408 (2000) 75. [24] V.K. Gupta, P. Kumar, R. Mangla, Electroanalysis 12 (2000) 752. [25] M. Javanbakht, M.R. Ganjali, H. Eshghi, H. Sharghi, M. Shamsipur, Electroanalysis 11 (1999) 81.
[26] H. Parham, M. Shamsipur, J. Membr. Sci. 86 (1994) 29. [27] E. Bakker, P. B¨uhlmann, E. Pretsch, Chem. Rev. 97 (1997) 3083. [28] R. Eugster, P.M. Gehrig, W.E. Morf, U.E. Spichiger, W. Simon, Anal. Chem. 63 (1991) 2285. [29] T. Rosatzin, E. Bakker, K. Suzuki, W. Simon, Anal. Chim. Acta 280 (1993) 197. [30] U. Schaller, E. Bakker, U.E. Spichiger, E. Pretsch, Anal. Chem. 66 (1994) 391. [31] D. Ammann, W.E. Morf, P. Anken, P.C. Meier, E. Pretsch, W. Simon, Ion-Sel. Electrode Rev. 5 (1983) 3. [32] M. Shamsipur, M.R. Ganjali, A. Moghimi, Anal. Chem. 70 (402) (1998) 5259. [33] R. Eugster, U.E. Spichiger, W. Simon, Anal. Chem. 65 (1993) 689. [34] P.L.H.M. Cobben, R.J.M. Egberink, J.B. Bomer, P. Bergveld, W. Verboom, D.N. Reinhoudt, J. Am. Chem. Soc. 114 (1992) 10573. [35] G.G. Guilbault, R.A. Durst, M.S. Frant, H. Freiser, E.H. Hansen, T.S. Light, E. Pungor, G. Rechnitz, N.M. Rice, T.J. Rohm, W. Simon, J.D.R. Thomas, Pure Appl. Chem. 48 (1976) 127. [36] Y. Umezawa, K. Umezawa, H. Sato, Pure Appl. Chem. 67 (1995) 507.