- Email: [email protected]
Highly selective methylammonium-selective membrane electrode made by the modification of the upper rim of calix[6]arene derivatives
Sensors and Actuators B 106 (2005) 772–778
Highly selective methylammonium-selective membrane electrode made by the modification of the upper rim of calix[6]arene derivatives Keisuke Uedaa , Hideaki Hiokib,∗∗ , Miwa Kubob , Mitsuaki Kodamab , Kiyoka Takaishic , Hirochika Yokosuc , Takashi Katsua,∗ b
a Faculty of Pharmaceutical Sciences, Okayama University, Tsushima, Okayama 700-8530, Japan Faculty of Pharmaceutical Sciences, Tokushima Bunri University, Yamashiro-cho, Tokushima 770-8514, Japan c Tokyo Kasei Kogyo Co. Ltd., 6-15-9, Toshima, Kita-ku, Tokyo 114-0003, Japan
Received 21 July 2004; received in revised form 9 September 2004; accepted 28 September 2004 Available online 5 November 2004
Abstract The upper rim of calix[6]arene-hexaacetic acid hexaethyl esters was modified with various alkyl substituents to develop ionophores for constructing organic ammonium ion-selective membrane electrodes. We found that the substitution of bulkier alkyl groups enabled a more effective discrimination of simple aliphatic ammonium ions such as methylammonium and ethylammonium. In particular, an electrode based on p-1,1,3,3-tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester showed the highest response to methylammonium among various Pot organic ammonium ions and inorganic cations, giving the following selectivity coefficients; log kCH : C2 H5 NH3 + , −0.2; C3 H7 NH3 + , + 3 NH3 ,X −1.4; C4 H9 NH3 + , −1.9; C6 H13 NH3 + , −1.6; phenethylammonium, −1.9; Na+ , −2.8; K+ , −1.8; Rb+ , −1.6; Cs+ , −0.2; NH4 + , −2.8. The electrode exhibited a near Nernstian response to methylammonium in the concentration range of 1 × 10−5 to 1 × 10−2 M with a slope of 58.5 mV per concentration decade in 0.1 M MgCl2 . The limit of detection was 3 × 10−6 M. The electrode could be used over a pH range of 2–10. Formation of a 1:1 complex between p-1,1,3,3-tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester and methylammonium was confirmed by electrospray ionization mass spectrometry. The response characteristics of the methylammonium-selective electrode were compared with those of NH4 + -selective electrodes so far developed. © 2004 Elsevier B.V. All rights reserved. Keywords: Ion-selective electrode; Organic ammonium ion; Methylammonium ionophore; Ammonium ionophore; Neutral carrier; Molecular receptor; Calix[6]arene derivative
1. Introduction One of the most pressing challenges in sensor research is the construction of highly selective receptors for molecular recognition. Calixarene derivatives are one candidate for molecular receptors for the selective recognition of primary organic ammonium ions and have been used to construct various types of molecular sensors including ∗
Corresponding author. Tel.: +81 86 251 7955; fax: +81 86 251 7926. Co-corresponding author. Tel.: +81 88 622 9611x5413; fax: +81 88 655 3051. E-mail addresses: [email protected] (H. Hioki), [email protected] (T. Katsu). ∗∗
0925-4005/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2004.09.029
ion-selective electrodes [1,2]. Among various calixarene derivatives, calix[6]arene-hexaacetic acid hexaethyl ester (1) (see the structure in Fig. 1) was an excellent ionophore (or more generally a molecular receptor) for making a hexylammonium-selective electrode [3]; this ionophore, named amine ionophore I, soon became commercially available from Fluka [4]. Later, it was found to induce a strong response to methylammonium [5,6]. However, there was extensive interference by Cs+ , consistent with the fact that this ionophore also acts as a cesium ionophore [4,7,8]. Recently, the corresponding p-tert-butylcarix[6]arene derivative (4), shown in Fig. 1, was found to respond to ethylammonium with remarkably less interference from Cs+ [9]. This ionophore also became commercially available
K. Ueda et al. / Sensors and Actuators B 106 (2005) 772–778
773
(p-chlorophenyl)borate (KTpClPB) and 2,6,13,16,23,26tetraoxaheptacyclo[25.4.4.47,12 .417,22 .01,17 .07,12 .017,22 ]tritetracontane (TD19C6) were from Dojindo Laboratories (Kumamoto, Japan); nonactin was from Sigma (St. Louis, MO, USA); and poly(vinyl chloride) (PVC) (degree of polymerization, 1020) was from Nacalai Tesque (Kyoto, Japan). All other chemicals were of analytical reagent grade. 2.2. Electrode system
Fig. 1. Chemical structures of calix[6]arene derivatives tested.
from Fluka; however, the name given to it was cesium ionophore II, although the response to ethylammonium was much stronger than that to Cs+ as mentioned above. We were particularly interested in the marked decrease in interference by Cs+ when a tert-butyl group was introduced at the upper rim of calix[6]arene. It is tempting to speculate that the introduction of an appropriate alkyl substituent covering the calix[6]arene cavity would lead to a stricter inclusion of the hydrophobic part of organic ammonium ions, thus making it possible to construct more specific organic ammonium ion-selective electrodes. To test this speculation, we investigated the effect of the substitution in calix[6]arene on the response to organic ammonium ions systematically in the present study. It was found that a much bulkier substituent covering the calix[6]arene cavity enhanced the recognition of methylammonium or ethylammonium with remarkable suppression from interference by inorganic cations. Notably, the introduction of a 1,1,3,3-tetramethylbutyl group (5) gave the highest selectivity toward methylammonium among various organic ammonium ions and inorganic cations.
A methylammonium-selective electrode was constructed using PVC-based membranes, as reported previously [9,15,16]. The PVC membranes had the following composition: 1 mg of ionophore, 20 mol% of KTpClPB relative to the ionophore, 60 l (55 mg) of bis(2-ethylhexyl) sebacate and 30 mg of PVC. The materials were dissolved in tetrahydrofuran (about 1 ml) and poured into a flat Petri dish (28 mm in diameter). The solvent was then evaporated off at room temperature. The resulting membrane was excised and attached to a PVC tube (4 mm o.d., 3 mm i.d.) with tetrahydrofuran adhesive. The sensor membranes were conditioned overnight in a solution of 10 mM methylamine hydrochloride. The electrochemical cell arrangement was Ag, AgCl/internal solution/sensor membrane/sample solution/1 M NH4 NO3 (salt bridge)/10 mM KCl/Ag, AgCl. The internal solution was the same as that used to condition the membrane. For comparison, NH4 + -selective electrodes were similarly prepared using nonactin or TD19C6 as an ionophore. In this case, the internal solution was 10 mM NH4 Cl. Potential measurements were made with a voltmeter produced by a field-effect transistor operational amplifier (LF356; National Semiconductor, Sunnyvale, CA, USA; input resistance >1012 ) connected to a recorder. To examine the pHdependence of the electrode, a miniature pH glass electrode (1826A-06T; Horiba, Kyoto, Japan), together with test and reference electrodes, was immersed in each sample solution to simultaneously measure the solution pH. The volume of the sample solution was 1 ml, because our electrode system was compact, as described previously [16,17]. 2.3. Evaluation of the electrode performance
2. Experimental 2.1. Reagents p-1,1,3,3-Tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester (5), p-tert-butylcalix[6]arene-hexaacetic acid hexaethyl ester (4), p-isopropylcalix[6]arene-hexaacetic acid hexaethyl ester (3) and p-methylcalix[6]arenehexaacetic acid hexaethyl ester (2) were synthesized according to procedures similar to ones described previously [10–13]. Methylammonium trifluoromethanesulfonate was synthesized as reported previously [14]. Other chemicals were obtained from commercial sources: amine ionophore I (1) and bis(2-ethylhexyl) sebacate were from Fluka (Buchs, Switzerland); potassium tetrakis
The detection limit was defined as the intersection of the extrapolated linear regions of the calibration graph [9,15,16,18]. The selectivity coefficients of the electrode Pot ) were determined by a separate solution method (ki,j [9,15,16,18,19] using respective chloride salts. The values were calculated from the equation, (Ej − Ei ) 1/z +log ci − log cj j S where Ei and Ej represent the emf readings measured for methylammonium and the interfering ion, respectively, S the theoretical slope of the electrode for methylammonium (59.2 mV at 25 ◦ C), ci and cj the concentrations of methylammonium and the interfering ion, respectively, and zj the charge
Pot = log ki,j
774
K. Ueda et al. / Sensors and Actuators B 106 (2005) 772–778
of the interfering ion. In some instances, a matched potential method [9,15,16,20,21] was also applied to evaluate the selectivity coefficients. In this case, we used a fixed concentration (1.0 × 10−4 M) of methylammonium as the background. The selectivity coefficients were calculated from the concentration of the interfering ion that induced the same amount of potential change as that induced by increasing the concentration of methylammonium to 2.0 × 10−4 M. This measurement was performed in the presence of 0.1 M MgCl2 to keep the ionic strength of the solution constant; MgCl2 was chosen, because the interference by Mg2+ was very slight [9,15,16]. All measurements were performed at room temperature (about 25 ◦ C). 2.4. Electrospray ionization (ESI) mass spectral measurements ESI mass spectra in the positive ion mode were obtained with a JEOL MS-BU30 instrument (Akishima, Tokyo, Japan). Samples dissolved in acetonitrile were infused directly into the mass spectrometer with a spray voltage of 3.7 kV. The resulting small particles were introduced into a heated capillary for desolvation, the temperature of which was 230 ◦ C.
3. Results and discussion 3.1. Selectivity of the electrodes Fig. 2 shows the potentiometric ion-selectivity coefficients of the electrodes using the ionophores shown in Fig. 1. The se-
lectivity coefficients against various organic ammonium ions and inorganic cations were compared. These were determined by a separate solution method [9,15,16,18,19]. An electrode made from amine ionophore (1) responded more strongly to methylammonium than to other primary alkylammonium ions, except for hexylammonium and phenethylammonium, and remarkably suppressed the response to lipophilic quaternary ammonium ions such as (CH3 )4 N+ and (C2 H5 )4 N+ . However, there was significant interference from Cs+ , consistent with the fact that this ionophore also acts as a cesium ionophore [4,7,8]. In the case of the p-methylcalix[6]arene derivative (2), the response to methylammonium was significantly enhanced and the response to hexylammonium and phenethylammonium was suppressed, suggesting that methyl groups attached to the p-position of calix[6]arene blocked the inclusion of bulkier primary organic ammonium ions and produced a more suitable structure that included organic ammonium ions of smaller size. Further investigation of the substituent’s effect revealed that there was a clear tendency for ionophores having bulkier alkyl substituents (3–5) to respond more strongly to methylammonium and ethylammonium, suppressing the response to Cs+ . A characteristic feature is that ionophore 4 afforded the strongest response to ethylammonium, while ionophore 5 afforded the strongest response to methylammonium among various organic ammonium ions and inorganic cations. The molecular basis for the recognition of calix[6]arene derivatives by primary organic ammonium ions is that receptors bearing a C3 -symmetrical proton acceptor site (negatively polarized oxygen atoms of carbonyl groups) of the calix[6]arene derivatives can strongly interact with primary ammonium ions, which act as C3 -symmetrical proton donors, while the hydrophobic cavities generated by
Fig. 2. Comparison of the selectivity coefficients of electrodes. Each number (1–5) corresponds to the ionophores shown in Fig. 1. PA, phenethylammonium.
K. Ueda et al. / Sensors and Actuators B 106 (2005) 772–778
775
Table 1 Comparison of electrode performance in response to methylammoniuma
Fig. 3. ESI mass spectrum of p-1,1,3,3-tetramethylbutylcalix[6]arenehexaacetic acid hexaethyl ester (4 × 10−5 M) in the presence of methylammonium trifluoromethanesulfonate (4 × 10−3 M).
the aromatic walls of phenol residues of the derivatives are potentially useful for the inclusion of non-polar moieties of organic ammonium ions [15,22,23]. Thus, enhanced responses to specific organic ammonium ions observed in ionophores 4 and 5 can be reasonably explained by the inclusion of a nonpolar methyl or ethyl group of organic ammonium ions in the hydrophobic cavities of the calix[6]arenes with bulky alkyl substituents. Indeed, ionophore 5 was found to form a 1:1 complex with methylammonium by ESI mass spectrometry, which will be discussed later. Among the ionophores tested in the present study, ionophore 5 was an excellent artificial receptor for methylammonium, suppressing interference from various organic and inorganic ions. To cross check the values of the selectivity coefficients, we measured the selectivity coefficients against organic and alkali metal ions for ionophore 5 using a matched potenPot tial method (log kCH : C2 H5 NH3 + , −0.3; C3 H7 NH3 + , + 3 NH3 ,X −1.0; C4 H9 NH3 + , −1.7; C6 H13 NH3 + , −1.4; K+ , −1.6; Rb+ , −1.5; Cs+ , −0.1). These values were similar to those obtained by a separate solution method.
Ionophore
Slope (mV per decade)
Detection limit (M)
Amine ionophore I (1) p-Methylcalix[6]arene-hexaacetic acid hexaethyl ester (2) p-Isopropylcalix[6]arene-hexaacetic acid hexaethyl ester (3) p-tert-Butylcalix[6]arene-hexaacetic acid hexaethyl ester (4) p-1,1,3,3-Tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester (5)
57.3 53.3
4 10
57.9
2
59.0
7
58.5
3
a All measurements were performed in the concentration range of 1 × 10−8 to 1 × 10−2 M methylamine hydrochloride in the presence of 0.1 M MgCl2 .
against the corresponding potential readings. A high concentration of MgCl2 was added to adjust the ionic strength of the solution [9,15,16]; MgCl2 was chosen because the interferPot ence from Mg2+ was slight (log kCH + 2+ < −4). The 3 NH3 ,Mg measurements were performed in the concentration range of 1 × 10−8 to 1 × 10−2 M methylamine hydrochloride. As indicated in Table 1, all of the ionophores tested in this study gave near-Nernstian responses to methylammonium, and 3 gave the best detection limit, though the difference among ionophores was small. The slope within the linear range and the detection limit for 5, which afforded the highest selectivity for methylammonium, were 58.5 mV per concentration decade and 3 × 10−6 M, respectively. The response time (90% of the final signal) of the electrode using 5 was below 10 s when the concentration of methylamine hydrochloride was changed from 10 to 100 M.
3.2. ESI mass spectral measurements ESI mass spectrometry is now a widely used technique for the detection of host–guest complexes [24]. We applied this technique to obtain evidence for the formation of a 1:1 complex between ionophore 5 and methylammonium. As shown in Fig. 3, the admixture of an acetonitrile solution of 5 (4 × 10−5 M) and methylammonium trifluoromethanesulfonate (4 × 10−3 M) gave a mass spectrum showing only one predominant peak (m/z 1857) attributed to the formation of a 1:1 complex. Because no peaks corresponding to other complexes were observed, this result substantiated the formation of a 1:1 complex suggested by ion-selective electrodes. 3.3. Sensitivity of the electrodes Calibration graphs were then obtained for the electrodes by measuring known amounts of methylamine hydrochloride added to 0.1 M MgCl2 , and plotting the concentrations
Fig. 4. Effects of pH on the response to methylammonium of the electrode based on p-1,1,3,3-tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester.
776
K. Ueda et al. / Sensors and Actuators B 106 (2005) 772–778
Fig. 5. Chemical structures of nonactin (6) and TD19C5 (7).
3.4. pH-dependence of the electrode We examined the pH-dependence to determine the effective pH range for the electrode using ionophore 5. The pH of the solution was adjusted by adding an appropriate amount of dilute hydrochloric acid or sodium hydroxide solution. The ionic strength of the solution was adjusted by adding 0.1 M CaCl2 instead of 0.1 M MgCl2 , because Mg(OH)2 is deposited at pH values above 9 [9,15,16]. As shown in Fig. 4, the electrode response was independent of the pH over the pH range of 2–10. The decrease in potential above pH 10 was attributable to an increase in the concentration of unprotonated amine, as the pKa of methylammonium has been reported to be 10.6 [25]. The decrease in potential below pH 2 may be due to the protonation of 5 in the membrane phase, resulting in a loss of the ability to form a complex with methylammonium. A similar observation was made with a methylammonium electrode using calix[4]arene-crown-6 conjugate [26], a serotoninselective electrode using homooxacalix[3]arene derivatives [16], and an ethylammonium electrode using ionophore 4 [9].
NH4 + coordinated with the four ether oxygen atoms of nonactin, thus increasing the ability to recognize NH4 + . In contrast, TD19C6 was designed to form a hydrogen bond with NH4 + through “three” sites [28]. Hence, TD19C6 recognized simultaneously the NH3 + group of organic ammonium ions, resulting in extensive interference by these ions. Thus, the present study indicated that inorganic and organic ammonium ions can be determined separately with the use of nonactin and ionophore 5. This was verified by comparing the calibration graphs of the electrodes for methylammonium and NH4 + . As shown in Fig. 7, the ionophore 5-based electrode responded more strongly to methylammonium than NH4 + , while the nonactin-based electrode showed the opposite response characteristics. The slope and detection limit of the electrode using ionophore 5 for CH3 NH3 + were 58.5 mV/decade and 3 × 10−6 M, respectively, as indicated in Table 1, while those using nonactin (6) for NH4 + were 56.9 mV/decade and 5 × 10−6 M, respectively.
3.5. Comparison of the response characteristics of the present electrode with those of NH4 + electrodes The present methylammonium-selective electrode based on ionophore 5 can sufficiently discriminate inorganic ammonium ions as is clear from the selectivity sequence shown in Fig. 2. Then, we were particularly interested in whether the inorganic ammonium ion-selective electrodes developed to date can discriminate between inorganic and organic ammonium ions. Nonactin [27,28] and TD19C6 [28] (see the structure in Fig. 5) were chosen as typical NH4 + ionophores. The selectivity coefficients of the electrodes based on these ionophores for organic ammonium ions as well as alkali metal cations are shown in Fig. 6. Although the electrode using nonactin exhibited higher selectivity against NH4 + than organic ammonium ions including methylammonium, the TD19C5based electrode showed lower selectivity. An X-ray crystallographic study showed that nonactin interacted with NH4 + by forming a hydrogen bond from four directions [29]; that is,
Fig. 6. Selectivity coefficients of electrodes based on nonactin (6) and TD19C5 (7). PA, phenethylammonium.
K. Ueda et al. / Sensors and Actuators B 106 (2005) 772–778
777
Fig. 7. Comparison of the calibration graphs of electrodes based on p-1,1,3,3-tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester (a) and nonactin (b) against CH3 NH3 + (䊉) and NH4 + (). Measurements were performed in the presence of 0.1 M MgCl2 .
4. Conclusion We demonstrated that calix[6]arene derivatives whose upper rim is modified with bulky alkyl substituents are quite suitable for the recognition of organic ammonium ions of small size, and p-1,1,3,3-tetramethylbutylcalix[6]arenehexaacetic acid hexaethyl ester is especially suitable for constructing a methylammonium-selective membrane electrode, being the best in several methylammonium ionophores so far developed [5,6,26]. Formation of a 1:1 complex between p-1,1,3,3-tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester and methylammonium was confirmed by electrospray ionization mass spectrometry.
Acknowledgements This work was supported by a Grant-in-Aid for Scientific Research (KAKENHI 16590027) from the Japan Society for the promotion of Science.
References [1] R. Ludwig, Calixarenes in analytical and separation chemistry, Fresenius J. Anal. Chem. 367 (2000) 103–128. [2] D. Diamond, K. Nolan, Calixarenes: designer ligands for chemical sensors, Anal. Chem. 73 (2001) 22A–29A. [3] W.H. Chan, K.K. Shiu, X.H. Gu, Ion-selective electrodes in organic analysis: primary amine selective polymeric membrane electrodes based on a calix[6]arene ionophore, Analyst 118 (1993) 863–867. [4] Fluka Chemika Selectophore® , Fluka, Buchs, Switzerland, 1996. [5] T. Katsu, D. Xu, K. Tsuji, T. Nagamatsu, Tris(2-ethylhexyl) phosphate and tricresyl phosphate as new neutral carriers for organic ammonium ion-selective membrane electrodes, Anal. Chim. Acta 354 (1997) 301–305. [6] T. Katsu, M. Akagi, T. Hiramatsu, T. Tsuchiya, Determination of the pH difference across a cell membrane using a methylammoniumselective membrane electrode, Analyst 123 (1998) 1369–1372.
[7] A. Cadogan, D. Diamond, M.R. Smyth, G. Svehla, M.A. McKervey, E.M. Seward, S.J. Harris, Caesium-selective poly(vinyl chloride) membrane electrodes based on calix[6]arene esters, Analyst 115 (1990) 1207–1210. [8] T. Katsu, K. Ido, K. Takaishi, H. Yokosu, Thallium(I)-selective membrane electrodes based on calix[6]arene or calix[5]arene derivatives, Sens. Actuat. B 87 (2002) 331–335. [9] T. Katsu, K. Ido, Ethylammonium-selective membrane electrode using p-tert-butylcalix[6]arene derivatives, Anal. Sci. 18 (2002) 473–476. [10] M.A. McKervey, E.M. Seward, G. Ferguson, B. Ruhl, S.J. Harris, Synthesis, X-ray crystal structures, and cation transfer properties of alkyl calixaryl acetates, a new series of molecular receptors, J. Chem. Soc., Chem. Commun. (1985) 388–390. [11] F. Arnaud-Neu, E.M. Collins, M. Deasy, G. Ferguson, S.J. Harris, B. Kaitner, A.J. Lough, M.A. McKervey, E. Marques, B.L. Ruhl, M.J. Schwing-Weill, E.M. Seward, Synthesis, X-ray crystal structures, and cation-binding properties of alkyl calixaryl esters and ketones, a new family of macrocyclic molecular receptors, J. Am. Chem. Soc. 111 (1989) 8681–8691. [12] K. Ohto, M. Yano, K. Inoue, T. Yamamoto, M. Goto, F. Nakashio, S. Shinkai, T. Nagasaki, Solvent extraction of trivalent rare earth metal ions with carboxylate derivatives of calixarenes, Anal. Sci. 11 (1995) 893–902. [13] R. Lamartine, H. Meziani, Synthesis, conformational analysis and extraction studies of p-isopropylcalix[n]arene derivatives (n = 4, 6, 8). A new family of molecular receptors, Supramol. Chem. 13 (2001) 143–162. [14] J. Vicente, M.T. Chicote, R. Guerrero, I.M. Saura-Llamas, P.G. Jones, M.C. Ram´ırez de Arellano, The first acetimine gold(I) and gold(III) complexes and the first acetonine complexes, Chem. Eur. J. 7 (2001) 638–646. [15] T. Katsu, N. Okaki, K. Watanabe, K. Takaishi, H. Yokosu, Comparative study of the response of membrane electrodes based on calix[6]arene and calix[8]arene derivatives to organic ammonium ions, Anal. Sci. 19 (2003) 771–774. [16] T. Katsu, K. Ido, S. Sagara, K. Tsubaki, K. Fuji, Homooxacalix[3]arene derivatives as ionophores for serotonin-selective membrane electrodes, Electroanalysis 15 (2003) 287–293. [17] T. Katsu, H. Kobayashi, Y. Fujita, Mode of action of gramicidin S on Escherichia coli membrane, Biochim. Biophys. Acta 860 (1986) 608–619. [18] R.P. Buck, E. Lindner, Recommendations for nomenclature of ionselective electrodes, Pure Appl. Chem. 66 (1994) 2527–2536.
778
K. Ueda et al. / Sensors and Actuators B 106 (2005) 772–778
[19] E. Bakker, E. Pretsch, P. B¨uhlmann, Selectivity of potentiometric ion sensors, Anal. Chem. 72 (2000) 1127–1133. [20] Y. Umezawa, K. Umezawa, H. Sato, Selectivity coefficients for ionpot selective electrodes: recommended methods for reporting KA,B values, Pure Appl. Chem. 67 (1995) 507–518. [21] K. Tohda, D. Dragoe, M. Shibata, Y. Umezawa, Studies on the matched potential method for determining the selectivity coefficients of ion-selective electrodes based on neutral ionophores: experimental and theoretical verifications, Anal. Sci. 17 (2001) 733– 743. [22] S.-K. Chang, M. Jang, S.Y. Han, J.H. Lee, M.H. Kang, K.T. No, Molecular recognition of butylamines by calixarene-based ester ligands, Chem. Lett. (1992) 1937–1940. [23] K. Odashima, K. Yagi, K. Tohda, Y. Umezawa, Potentiometric discrimination of organic amines by a liquid membrane electrode based on a lipophilic hexaesters of calix[6]arene, Anal. Chem. 65 (1993) 1074–1083.
[24] B. Ganem, J.D. Henion, Going gently into flight: analyzing noncovalent interactions by mass spectrometry, Bioorg. Med. Chem. 11 (2003) 311–314. [25] J.A. Dean (Ed.), Lange’s Handbook of Chemistry, 11th ed., McGrawHill, New York, 1973. [26] T. Katsu, M. Matsumoto, Discrimination of methylammonium from organic ammonium ions using ion-selective electrodes based on calix[4]arene-crown-6 conjugates, Anal. Sci. 17 (2001) 721–725. [27] P. B¨uhlmann, E. Pretsch, E. Bakker, Carrier-based ion-selective electrodes and bulk optodes. 2. Ionophores for potentiometric and optical sensors, Chem. Rev. 98 (1998) 1593–1687. [28] S. Sasaki, T. Amano, G. Monma, T. Otsuka, N. Iwasawa, D. Citterio, H. Hisamoto, K. Suzuki, Comparison of two molecular design strategies for the development of an ammonium ionophore more highly selective than nonactin, Anal. Chem. 74 (2002) 4845–4848. [29] K. Neupert-Laves, M. Dobler, The crystal structure of the NH4 NCS complex of nonactin, Helv. Chim. Acta 59 (1976) 614–623.
Highly selective methylammonium-selective membrane electrode made by the modification of the upper rim of calix[6]arene derivatives Keisuke Uedaa , Hideaki Hiokib,∗∗ , Miwa Kubob , Mitsuaki Kodamab , Kiyoka Takaishic , Hirochika Yokosuc , Takashi Katsua,∗ b
a Faculty of Pharmaceutical Sciences, Okayama University, Tsushima, Okayama 700-8530, Japan Faculty of Pharmaceutical Sciences, Tokushima Bunri University, Yamashiro-cho, Tokushima 770-8514, Japan c Tokyo Kasei Kogyo Co. Ltd., 6-15-9, Toshima, Kita-ku, Tokyo 114-0003, Japan
Received 21 July 2004; received in revised form 9 September 2004; accepted 28 September 2004 Available online 5 November 2004
Abstract The upper rim of calix[6]arene-hexaacetic acid hexaethyl esters was modified with various alkyl substituents to develop ionophores for constructing organic ammonium ion-selective membrane electrodes. We found that the substitution of bulkier alkyl groups enabled a more effective discrimination of simple aliphatic ammonium ions such as methylammonium and ethylammonium. In particular, an electrode based on p-1,1,3,3-tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester showed the highest response to methylammonium among various Pot organic ammonium ions and inorganic cations, giving the following selectivity coefficients; log kCH : C2 H5 NH3 + , −0.2; C3 H7 NH3 + , + 3 NH3 ,X −1.4; C4 H9 NH3 + , −1.9; C6 H13 NH3 + , −1.6; phenethylammonium, −1.9; Na+ , −2.8; K+ , −1.8; Rb+ , −1.6; Cs+ , −0.2; NH4 + , −2.8. The electrode exhibited a near Nernstian response to methylammonium in the concentration range of 1 × 10−5 to 1 × 10−2 M with a slope of 58.5 mV per concentration decade in 0.1 M MgCl2 . The limit of detection was 3 × 10−6 M. The electrode could be used over a pH range of 2–10. Formation of a 1:1 complex between p-1,1,3,3-tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester and methylammonium was confirmed by electrospray ionization mass spectrometry. The response characteristics of the methylammonium-selective electrode were compared with those of NH4 + -selective electrodes so far developed. © 2004 Elsevier B.V. All rights reserved. Keywords: Ion-selective electrode; Organic ammonium ion; Methylammonium ionophore; Ammonium ionophore; Neutral carrier; Molecular receptor; Calix[6]arene derivative
1. Introduction One of the most pressing challenges in sensor research is the construction of highly selective receptors for molecular recognition. Calixarene derivatives are one candidate for molecular receptors for the selective recognition of primary organic ammonium ions and have been used to construct various types of molecular sensors including ∗
Corresponding author. Tel.: +81 86 251 7955; fax: +81 86 251 7926. Co-corresponding author. Tel.: +81 88 622 9611x5413; fax: +81 88 655 3051. E-mail addresses: [email protected] (H. Hioki), [email protected] (T. Katsu). ∗∗
0925-4005/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2004.09.029
ion-selective electrodes [1,2]. Among various calixarene derivatives, calix[6]arene-hexaacetic acid hexaethyl ester (1) (see the structure in Fig. 1) was an excellent ionophore (or more generally a molecular receptor) for making a hexylammonium-selective electrode [3]; this ionophore, named amine ionophore I, soon became commercially available from Fluka [4]. Later, it was found to induce a strong response to methylammonium [5,6]. However, there was extensive interference by Cs+ , consistent with the fact that this ionophore also acts as a cesium ionophore [4,7,8]. Recently, the corresponding p-tert-butylcarix[6]arene derivative (4), shown in Fig. 1, was found to respond to ethylammonium with remarkably less interference from Cs+ [9]. This ionophore also became commercially available
K. Ueda et al. / Sensors and Actuators B 106 (2005) 772–778
773
(p-chlorophenyl)borate (KTpClPB) and 2,6,13,16,23,26tetraoxaheptacyclo[25.4.4.47,12 .417,22 .01,17 .07,12 .017,22 ]tritetracontane (TD19C6) were from Dojindo Laboratories (Kumamoto, Japan); nonactin was from Sigma (St. Louis, MO, USA); and poly(vinyl chloride) (PVC) (degree of polymerization, 1020) was from Nacalai Tesque (Kyoto, Japan). All other chemicals were of analytical reagent grade. 2.2. Electrode system
Fig. 1. Chemical structures of calix[6]arene derivatives tested.
from Fluka; however, the name given to it was cesium ionophore II, although the response to ethylammonium was much stronger than that to Cs+ as mentioned above. We were particularly interested in the marked decrease in interference by Cs+ when a tert-butyl group was introduced at the upper rim of calix[6]arene. It is tempting to speculate that the introduction of an appropriate alkyl substituent covering the calix[6]arene cavity would lead to a stricter inclusion of the hydrophobic part of organic ammonium ions, thus making it possible to construct more specific organic ammonium ion-selective electrodes. To test this speculation, we investigated the effect of the substitution in calix[6]arene on the response to organic ammonium ions systematically in the present study. It was found that a much bulkier substituent covering the calix[6]arene cavity enhanced the recognition of methylammonium or ethylammonium with remarkable suppression from interference by inorganic cations. Notably, the introduction of a 1,1,3,3-tetramethylbutyl group (5) gave the highest selectivity toward methylammonium among various organic ammonium ions and inorganic cations.
A methylammonium-selective electrode was constructed using PVC-based membranes, as reported previously [9,15,16]. The PVC membranes had the following composition: 1 mg of ionophore, 20 mol% of KTpClPB relative to the ionophore, 60 l (55 mg) of bis(2-ethylhexyl) sebacate and 30 mg of PVC. The materials were dissolved in tetrahydrofuran (about 1 ml) and poured into a flat Petri dish (28 mm in diameter). The solvent was then evaporated off at room temperature. The resulting membrane was excised and attached to a PVC tube (4 mm o.d., 3 mm i.d.) with tetrahydrofuran adhesive. The sensor membranes were conditioned overnight in a solution of 10 mM methylamine hydrochloride. The electrochemical cell arrangement was Ag, AgCl/internal solution/sensor membrane/sample solution/1 M NH4 NO3 (salt bridge)/10 mM KCl/Ag, AgCl. The internal solution was the same as that used to condition the membrane. For comparison, NH4 + -selective electrodes were similarly prepared using nonactin or TD19C6 as an ionophore. In this case, the internal solution was 10 mM NH4 Cl. Potential measurements were made with a voltmeter produced by a field-effect transistor operational amplifier (LF356; National Semiconductor, Sunnyvale, CA, USA; input resistance >1012 ) connected to a recorder. To examine the pHdependence of the electrode, a miniature pH glass electrode (1826A-06T; Horiba, Kyoto, Japan), together with test and reference electrodes, was immersed in each sample solution to simultaneously measure the solution pH. The volume of the sample solution was 1 ml, because our electrode system was compact, as described previously [16,17]. 2.3. Evaluation of the electrode performance
2. Experimental 2.1. Reagents p-1,1,3,3-Tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester (5), p-tert-butylcalix[6]arene-hexaacetic acid hexaethyl ester (4), p-isopropylcalix[6]arene-hexaacetic acid hexaethyl ester (3) and p-methylcalix[6]arenehexaacetic acid hexaethyl ester (2) were synthesized according to procedures similar to ones described previously [10–13]. Methylammonium trifluoromethanesulfonate was synthesized as reported previously [14]. Other chemicals were obtained from commercial sources: amine ionophore I (1) and bis(2-ethylhexyl) sebacate were from Fluka (Buchs, Switzerland); potassium tetrakis
The detection limit was defined as the intersection of the extrapolated linear regions of the calibration graph [9,15,16,18]. The selectivity coefficients of the electrode Pot ) were determined by a separate solution method (ki,j [9,15,16,18,19] using respective chloride salts. The values were calculated from the equation, (Ej − Ei ) 1/z +log ci − log cj j S where Ei and Ej represent the emf readings measured for methylammonium and the interfering ion, respectively, S the theoretical slope of the electrode for methylammonium (59.2 mV at 25 ◦ C), ci and cj the concentrations of methylammonium and the interfering ion, respectively, and zj the charge
Pot = log ki,j
774
K. Ueda et al. / Sensors and Actuators B 106 (2005) 772–778
of the interfering ion. In some instances, a matched potential method [9,15,16,20,21] was also applied to evaluate the selectivity coefficients. In this case, we used a fixed concentration (1.0 × 10−4 M) of methylammonium as the background. The selectivity coefficients were calculated from the concentration of the interfering ion that induced the same amount of potential change as that induced by increasing the concentration of methylammonium to 2.0 × 10−4 M. This measurement was performed in the presence of 0.1 M MgCl2 to keep the ionic strength of the solution constant; MgCl2 was chosen, because the interference by Mg2+ was very slight [9,15,16]. All measurements were performed at room temperature (about 25 ◦ C). 2.4. Electrospray ionization (ESI) mass spectral measurements ESI mass spectra in the positive ion mode were obtained with a JEOL MS-BU30 instrument (Akishima, Tokyo, Japan). Samples dissolved in acetonitrile were infused directly into the mass spectrometer with a spray voltage of 3.7 kV. The resulting small particles were introduced into a heated capillary for desolvation, the temperature of which was 230 ◦ C.
3. Results and discussion 3.1. Selectivity of the electrodes Fig. 2 shows the potentiometric ion-selectivity coefficients of the electrodes using the ionophores shown in Fig. 1. The se-
lectivity coefficients against various organic ammonium ions and inorganic cations were compared. These were determined by a separate solution method [9,15,16,18,19]. An electrode made from amine ionophore (1) responded more strongly to methylammonium than to other primary alkylammonium ions, except for hexylammonium and phenethylammonium, and remarkably suppressed the response to lipophilic quaternary ammonium ions such as (CH3 )4 N+ and (C2 H5 )4 N+ . However, there was significant interference from Cs+ , consistent with the fact that this ionophore also acts as a cesium ionophore [4,7,8]. In the case of the p-methylcalix[6]arene derivative (2), the response to methylammonium was significantly enhanced and the response to hexylammonium and phenethylammonium was suppressed, suggesting that methyl groups attached to the p-position of calix[6]arene blocked the inclusion of bulkier primary organic ammonium ions and produced a more suitable structure that included organic ammonium ions of smaller size. Further investigation of the substituent’s effect revealed that there was a clear tendency for ionophores having bulkier alkyl substituents (3–5) to respond more strongly to methylammonium and ethylammonium, suppressing the response to Cs+ . A characteristic feature is that ionophore 4 afforded the strongest response to ethylammonium, while ionophore 5 afforded the strongest response to methylammonium among various organic ammonium ions and inorganic cations. The molecular basis for the recognition of calix[6]arene derivatives by primary organic ammonium ions is that receptors bearing a C3 -symmetrical proton acceptor site (negatively polarized oxygen atoms of carbonyl groups) of the calix[6]arene derivatives can strongly interact with primary ammonium ions, which act as C3 -symmetrical proton donors, while the hydrophobic cavities generated by
Fig. 2. Comparison of the selectivity coefficients of electrodes. Each number (1–5) corresponds to the ionophores shown in Fig. 1. PA, phenethylammonium.
K. Ueda et al. / Sensors and Actuators B 106 (2005) 772–778
775
Table 1 Comparison of electrode performance in response to methylammoniuma
Fig. 3. ESI mass spectrum of p-1,1,3,3-tetramethylbutylcalix[6]arenehexaacetic acid hexaethyl ester (4 × 10−5 M) in the presence of methylammonium trifluoromethanesulfonate (4 × 10−3 M).
the aromatic walls of phenol residues of the derivatives are potentially useful for the inclusion of non-polar moieties of organic ammonium ions [15,22,23]. Thus, enhanced responses to specific organic ammonium ions observed in ionophores 4 and 5 can be reasonably explained by the inclusion of a nonpolar methyl or ethyl group of organic ammonium ions in the hydrophobic cavities of the calix[6]arenes with bulky alkyl substituents. Indeed, ionophore 5 was found to form a 1:1 complex with methylammonium by ESI mass spectrometry, which will be discussed later. Among the ionophores tested in the present study, ionophore 5 was an excellent artificial receptor for methylammonium, suppressing interference from various organic and inorganic ions. To cross check the values of the selectivity coefficients, we measured the selectivity coefficients against organic and alkali metal ions for ionophore 5 using a matched potenPot tial method (log kCH : C2 H5 NH3 + , −0.3; C3 H7 NH3 + , + 3 NH3 ,X −1.0; C4 H9 NH3 + , −1.7; C6 H13 NH3 + , −1.4; K+ , −1.6; Rb+ , −1.5; Cs+ , −0.1). These values were similar to those obtained by a separate solution method.
Ionophore
Slope (mV per decade)
Detection limit (M)
Amine ionophore I (1) p-Methylcalix[6]arene-hexaacetic acid hexaethyl ester (2) p-Isopropylcalix[6]arene-hexaacetic acid hexaethyl ester (3) p-tert-Butylcalix[6]arene-hexaacetic acid hexaethyl ester (4) p-1,1,3,3-Tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester (5)
57.3 53.3
4 10
57.9
2
59.0
7
58.5
3
a All measurements were performed in the concentration range of 1 × 10−8 to 1 × 10−2 M methylamine hydrochloride in the presence of 0.1 M MgCl2 .
against the corresponding potential readings. A high concentration of MgCl2 was added to adjust the ionic strength of the solution [9,15,16]; MgCl2 was chosen because the interferPot ence from Mg2+ was slight (log kCH + 2+ < −4). The 3 NH3 ,Mg measurements were performed in the concentration range of 1 × 10−8 to 1 × 10−2 M methylamine hydrochloride. As indicated in Table 1, all of the ionophores tested in this study gave near-Nernstian responses to methylammonium, and 3 gave the best detection limit, though the difference among ionophores was small. The slope within the linear range and the detection limit for 5, which afforded the highest selectivity for methylammonium, were 58.5 mV per concentration decade and 3 × 10−6 M, respectively. The response time (90% of the final signal) of the electrode using 5 was below 10 s when the concentration of methylamine hydrochloride was changed from 10 to 100 M.
3.2. ESI mass spectral measurements ESI mass spectrometry is now a widely used technique for the detection of host–guest complexes [24]. We applied this technique to obtain evidence for the formation of a 1:1 complex between ionophore 5 and methylammonium. As shown in Fig. 3, the admixture of an acetonitrile solution of 5 (4 × 10−5 M) and methylammonium trifluoromethanesulfonate (4 × 10−3 M) gave a mass spectrum showing only one predominant peak (m/z 1857) attributed to the formation of a 1:1 complex. Because no peaks corresponding to other complexes were observed, this result substantiated the formation of a 1:1 complex suggested by ion-selective electrodes. 3.3. Sensitivity of the electrodes Calibration graphs were then obtained for the electrodes by measuring known amounts of methylamine hydrochloride added to 0.1 M MgCl2 , and plotting the concentrations
Fig. 4. Effects of pH on the response to methylammonium of the electrode based on p-1,1,3,3-tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester.
776
K. Ueda et al. / Sensors and Actuators B 106 (2005) 772–778
Fig. 5. Chemical structures of nonactin (6) and TD19C5 (7).
3.4. pH-dependence of the electrode We examined the pH-dependence to determine the effective pH range for the electrode using ionophore 5. The pH of the solution was adjusted by adding an appropriate amount of dilute hydrochloric acid or sodium hydroxide solution. The ionic strength of the solution was adjusted by adding 0.1 M CaCl2 instead of 0.1 M MgCl2 , because Mg(OH)2 is deposited at pH values above 9 [9,15,16]. As shown in Fig. 4, the electrode response was independent of the pH over the pH range of 2–10. The decrease in potential above pH 10 was attributable to an increase in the concentration of unprotonated amine, as the pKa of methylammonium has been reported to be 10.6 [25]. The decrease in potential below pH 2 may be due to the protonation of 5 in the membrane phase, resulting in a loss of the ability to form a complex with methylammonium. A similar observation was made with a methylammonium electrode using calix[4]arene-crown-6 conjugate [26], a serotoninselective electrode using homooxacalix[3]arene derivatives [16], and an ethylammonium electrode using ionophore 4 [9].
NH4 + coordinated with the four ether oxygen atoms of nonactin, thus increasing the ability to recognize NH4 + . In contrast, TD19C6 was designed to form a hydrogen bond with NH4 + through “three” sites [28]. Hence, TD19C6 recognized simultaneously the NH3 + group of organic ammonium ions, resulting in extensive interference by these ions. Thus, the present study indicated that inorganic and organic ammonium ions can be determined separately with the use of nonactin and ionophore 5. This was verified by comparing the calibration graphs of the electrodes for methylammonium and NH4 + . As shown in Fig. 7, the ionophore 5-based electrode responded more strongly to methylammonium than NH4 + , while the nonactin-based electrode showed the opposite response characteristics. The slope and detection limit of the electrode using ionophore 5 for CH3 NH3 + were 58.5 mV/decade and 3 × 10−6 M, respectively, as indicated in Table 1, while those using nonactin (6) for NH4 + were 56.9 mV/decade and 5 × 10−6 M, respectively.
3.5. Comparison of the response characteristics of the present electrode with those of NH4 + electrodes The present methylammonium-selective electrode based on ionophore 5 can sufficiently discriminate inorganic ammonium ions as is clear from the selectivity sequence shown in Fig. 2. Then, we were particularly interested in whether the inorganic ammonium ion-selective electrodes developed to date can discriminate between inorganic and organic ammonium ions. Nonactin [27,28] and TD19C6 [28] (see the structure in Fig. 5) were chosen as typical NH4 + ionophores. The selectivity coefficients of the electrodes based on these ionophores for organic ammonium ions as well as alkali metal cations are shown in Fig. 6. Although the electrode using nonactin exhibited higher selectivity against NH4 + than organic ammonium ions including methylammonium, the TD19C5based electrode showed lower selectivity. An X-ray crystallographic study showed that nonactin interacted with NH4 + by forming a hydrogen bond from four directions [29]; that is,
Fig. 6. Selectivity coefficients of electrodes based on nonactin (6) and TD19C5 (7). PA, phenethylammonium.
K. Ueda et al. / Sensors and Actuators B 106 (2005) 772–778
777
Fig. 7. Comparison of the calibration graphs of electrodes based on p-1,1,3,3-tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester (a) and nonactin (b) against CH3 NH3 + (䊉) and NH4 + (). Measurements were performed in the presence of 0.1 M MgCl2 .
4. Conclusion We demonstrated that calix[6]arene derivatives whose upper rim is modified with bulky alkyl substituents are quite suitable for the recognition of organic ammonium ions of small size, and p-1,1,3,3-tetramethylbutylcalix[6]arenehexaacetic acid hexaethyl ester is especially suitable for constructing a methylammonium-selective membrane electrode, being the best in several methylammonium ionophores so far developed [5,6,26]. Formation of a 1:1 complex between p-1,1,3,3-tetramethylbutylcalix[6]arene-hexaacetic acid hexaethyl ester and methylammonium was confirmed by electrospray ionization mass spectrometry.
Acknowledgements This work was supported by a Grant-in-Aid for Scientific Research (KAKENHI 16590027) from the Japan Society for the promotion of Science.
References [1] R. Ludwig, Calixarenes in analytical and separation chemistry, Fresenius J. Anal. Chem. 367 (2000) 103–128. [2] D. Diamond, K. Nolan, Calixarenes: designer ligands for chemical sensors, Anal. Chem. 73 (2001) 22A–29A. [3] W.H. Chan, K.K. Shiu, X.H. Gu, Ion-selective electrodes in organic analysis: primary amine selective polymeric membrane electrodes based on a calix[6]arene ionophore, Analyst 118 (1993) 863–867. [4] Fluka Chemika Selectophore® , Fluka, Buchs, Switzerland, 1996. [5] T. Katsu, D. Xu, K. Tsuji, T. Nagamatsu, Tris(2-ethylhexyl) phosphate and tricresyl phosphate as new neutral carriers for organic ammonium ion-selective membrane electrodes, Anal. Chim. Acta 354 (1997) 301–305. [6] T. Katsu, M. Akagi, T. Hiramatsu, T. Tsuchiya, Determination of the pH difference across a cell membrane using a methylammoniumselective membrane electrode, Analyst 123 (1998) 1369–1372.
[7] A. Cadogan, D. Diamond, M.R. Smyth, G. Svehla, M.A. McKervey, E.M. Seward, S.J. Harris, Caesium-selective poly(vinyl chloride) membrane electrodes based on calix[6]arene esters, Analyst 115 (1990) 1207–1210. [8] T. Katsu, K. Ido, K. Takaishi, H. Yokosu, Thallium(I)-selective membrane electrodes based on calix[6]arene or calix[5]arene derivatives, Sens. Actuat. B 87 (2002) 331–335. [9] T. Katsu, K. Ido, Ethylammonium-selective membrane electrode using p-tert-butylcalix[6]arene derivatives, Anal. Sci. 18 (2002) 473–476. [10] M.A. McKervey, E.M. Seward, G. Ferguson, B. Ruhl, S.J. Harris, Synthesis, X-ray crystal structures, and cation transfer properties of alkyl calixaryl acetates, a new series of molecular receptors, J. Chem. Soc., Chem. Commun. (1985) 388–390. [11] F. Arnaud-Neu, E.M. Collins, M. Deasy, G. Ferguson, S.J. Harris, B. Kaitner, A.J. Lough, M.A. McKervey, E. Marques, B.L. Ruhl, M.J. Schwing-Weill, E.M. Seward, Synthesis, X-ray crystal structures, and cation-binding properties of alkyl calixaryl esters and ketones, a new family of macrocyclic molecular receptors, J. Am. Chem. Soc. 111 (1989) 8681–8691. [12] K. Ohto, M. Yano, K. Inoue, T. Yamamoto, M. Goto, F. Nakashio, S. Shinkai, T. Nagasaki, Solvent extraction of trivalent rare earth metal ions with carboxylate derivatives of calixarenes, Anal. Sci. 11 (1995) 893–902. [13] R. Lamartine, H. Meziani, Synthesis, conformational analysis and extraction studies of p-isopropylcalix[n]arene derivatives (n = 4, 6, 8). A new family of molecular receptors, Supramol. Chem. 13 (2001) 143–162. [14] J. Vicente, M.T. Chicote, R. Guerrero, I.M. Saura-Llamas, P.G. Jones, M.C. Ram´ırez de Arellano, The first acetimine gold(I) and gold(III) complexes and the first acetonine complexes, Chem. Eur. J. 7 (2001) 638–646. [15] T. Katsu, N. Okaki, K. Watanabe, K. Takaishi, H. Yokosu, Comparative study of the response of membrane electrodes based on calix[6]arene and calix[8]arene derivatives to organic ammonium ions, Anal. Sci. 19 (2003) 771–774. [16] T. Katsu, K. Ido, S. Sagara, K. Tsubaki, K. Fuji, Homooxacalix[3]arene derivatives as ionophores for serotonin-selective membrane electrodes, Electroanalysis 15 (2003) 287–293. [17] T. Katsu, H. Kobayashi, Y. Fujita, Mode of action of gramicidin S on Escherichia coli membrane, Biochim. Biophys. Acta 860 (1986) 608–619. [18] R.P. Buck, E. Lindner, Recommendations for nomenclature of ionselective electrodes, Pure Appl. Chem. 66 (1994) 2527–2536.
778
K. Ueda et al. / Sensors and Actuators B 106 (2005) 772–778
[19] E. Bakker, E. Pretsch, P. B¨uhlmann, Selectivity of potentiometric ion sensors, Anal. Chem. 72 (2000) 1127–1133. [20] Y. Umezawa, K. Umezawa, H. Sato, Selectivity coefficients for ionpot selective electrodes: recommended methods for reporting KA,B values, Pure Appl. Chem. 67 (1995) 507–518. [21] K. Tohda, D. Dragoe, M. Shibata, Y. Umezawa, Studies on the matched potential method for determining the selectivity coefficients of ion-selective electrodes based on neutral ionophores: experimental and theoretical verifications, Anal. Sci. 17 (2001) 733– 743. [22] S.-K. Chang, M. Jang, S.Y. Han, J.H. Lee, M.H. Kang, K.T. No, Molecular recognition of butylamines by calixarene-based ester ligands, Chem. Lett. (1992) 1937–1940. [23] K. Odashima, K. Yagi, K. Tohda, Y. Umezawa, Potentiometric discrimination of organic amines by a liquid membrane electrode based on a lipophilic hexaesters of calix[6]arene, Anal. Chem. 65 (1993) 1074–1083.
[24] B. Ganem, J.D. Henion, Going gently into flight: analyzing noncovalent interactions by mass spectrometry, Bioorg. Med. Chem. 11 (2003) 311–314. [25] J.A. Dean (Ed.), Lange’s Handbook of Chemistry, 11th ed., McGrawHill, New York, 1973. [26] T. Katsu, M. Matsumoto, Discrimination of methylammonium from organic ammonium ions using ion-selective electrodes based on calix[4]arene-crown-6 conjugates, Anal. Sci. 17 (2001) 721–725. [27] P. B¨uhlmann, E. Pretsch, E. Bakker, Carrier-based ion-selective electrodes and bulk optodes. 2. Ionophores for potentiometric and optical sensors, Chem. Rev. 98 (1998) 1593–1687. [28] S. Sasaki, T. Amano, G. Monma, T. Otsuka, N. Iwasawa, D. Citterio, H. Hisamoto, K. Suzuki, Comparison of two molecular design strategies for the development of an ammonium ionophore more highly selective than nonactin, Anal. Chem. 74 (2002) 4845–4848. [29] K. Neupert-Laves, M. Dobler, The crystal structure of the NH4 NCS complex of nonactin, Helv. Chim. Acta 59 (1976) 614–623.