Anal. Chem. 1990, 62,1428-1431
1428
inductively coupled plasma atomic emission spectrometry (13), atomic absorption spectrometry (AAS) (14,15),and catalytic spectrophotometry (16-18). The first two methods are disadvantages in terms of cost and instruments in the routine analysis. AAS is often lacking in sensitivity and affected by a matrix of samples such as salinity. Catalytic methods are highly sensitive but are generally lacking in simplicity. The proposed method using DMQAP not only is one of the most sensitive methods for the determination of vanadium but also is excellent in terms of selectivity and simplicity. Therefore, this method will be successfully applied to the monitoring of vanadium in small amounts of environmental samples, for example, 1 m3 of air and 10 cm3 of natural waters. This technique with TAAS further permits an improvement of selectivity for RP-HPLC separation in addition to the choice of a column used and of an organic modifier and its concentration in a mobile phase.
ACKNOWLEDGMENT I thank Professor Hiroto Watanabe of Hokkaido University for helpful discussions and for criticizing the manuscript. I also thank Professor Hidehiko Kitajima and Toshio Morita of Fukui University for suggestions for the synthesis of DMQAP and its elemental analysis. LITERATURE CITED (1) Yotsuyanagi, T.; Hoshino, H. Bunseki 1983, 566-572, and references
therein.
Shibata, S. Chelates in Analyricel Chemistty; Barnard, A. J. Jr.; Flaschka. H. A., Ed.; Marcel Dekker: New York, 1972; Vol. 4. Shibata, S.;Furukawa, M.; Toei, K. Anal. Chim. Acta 1973, 66, 397-409.
Hoshino, H.; Yotsuyanagi, T. Bunseki Kagaku 1982, 37,E435-E438. Miura, J. Analyst 1989, 174, 1323-1329. Sato, K.; Sakata, M. Bunseki Kagaku 1985, 34, 271-275. Kiel, J. S.;Morgan, S.L.; Abramson, R . K. J. Chromatogr. 1985, 320, 313-323.
Hansen, S.H.; Helboe, P.;Thomsen, M. Trends Anal. Chem. 1985, 4 , 233-237.
Eartha, A.; Vigh, G. J. Chromatogr. 1983, 260, 337-345. Division of Chemistry and Chemical Technology, Environmental Studies Board, National Research Council, Medical and Biologic Effects of Environmental Pollutants, Vanadium; National Academy of Sciences: Washington, DC, 1974; Japanese Edition, Tokyo Kagakudojin: Tokyo, 1977.
Miura, J.; Hoshino, H.; Yotsuyanagi, T. International Symposium on New Sensors and Methods for Environmental Characterization, Kyoto, Japan, November, 1986; S1-03. Greenberg, R. R.; Kingston, H. M. Anal. Chem. 1983, 55, 1160-1165. Wang, C.-F.; Miau, T. T.; Perng, J. Y.; Yeh, S. J.; Chiang, P. C.; Tsai, H. T.; Yang, M. H. Analyst 1989, 174, 1067-1070. Torninaga, M.; Bansho, K.; Umezaki, Y. Anal. Chim. Acta 1985, 169, 171-177.
Yamashige. T.; Yamamoto, M.; Sunahara, H. Analyst 1989, 774, 1071-1077.
Hirayama, K.; Unohara, N. Bunseki Kagaku 1980, 29, 733-737. Fukasawa, T.; Kawakubo, S.;Okabe, T.; Mlzuike, A. Bunseki Kagaku 1984, 33, 609-614. Nakano, S.;Yamada. C.; Sakai, M.; Kawashima, T. Anal. Sci. 1988, 2 , 61-65.
RECEIVED for review December 28, 1989. Accepted April 2, 1990. This work was supported by a grant from Saneyoshi Scholarship Foundation.
Ion-Selective Electrodes Using an Ionophore Covalently Attached to Carboxylated Poly(viny1 chloride) Sylvia Daunert and Leonidas G. Bachas*
Department of Chemistry and Center of Membrane Sciences, University of Kentucky, Lexington, Kentucky 40506-0055
An Ion-selective electrode was prepared by using a polymer membrane on which an Ionophore had been covalently attached. Specifically, benro-15-crown4 was modified chemically to yield 4'-aminobenzo-l5-crownb, whlch was then attached covalently to carboxylated poly(viny1 chloride). The prepared electrode responded selectlvely to potassium wlth a Nernstian slope. This polymer-bound ionophore had an Increased llfetlme when compared to the one obtalned by using a poly( vinyl chloride)-based matrlx impregnated wlth benzo-15-crown-5.
Ion-selective electrodes (ISEs) based on ionophore-impregnated polymer membranes (typically, plasticized poly(vinyl chloride) (PVC)) are now commonly employed in a variety of analyses (1-4). The operational lifetime of these ISEs is affected by the solubility of the ionophore in the plasticizer and can be limited by the leaching of the ionophore and the plasticizer from the polymer matrix (2,5,6). Usually, this leaching also worsens the detection limits of the electrode and results in gradual deterioration of the response (7). Covalently grafting the ionophore to a polymeric backbone has been suggested as a possible solution to the problem of leaching ( 2 ) . The first report of chemical immobilization of
an ionophore on a polymer matrix involved the use of phosphorylated VAGH to develop an electrode for Caz+ (8, 9); VAGH is a copolymer composed of vinyl chloride, vinyl acetate, and vinyl alcohol. An electrode for calcium was also prepared by cross-linking a styrene-n-butadiene-styrene triblock elastomer (SBS) with triallyl phosphate (10). A graft copolymer of cellulose and poly(acrylonitri1e) containing hydroxamic acid was used by Volovik et al. in a Ni2+-selective electrode ( I I ) . Further, anion-selective electrodes have resulted by covalently attaching quaternary ammonium moieties to polymer matrices such as poly(vinylbenzy1 chloride) (71, SBS (12), chloromethylated cross-linked styrene (13), and sulfonated PVC (14). Finally, polymers with functional groups such as Nafion (a perfluorosulfonate polymer) (5),poly( 1,2diaminobenzene) (15), and polypyrrole (16) have been used without plasticizers in the development of ISEs for tributylammonium ions, protons, and chloride, respectively. Typically, the above electrodes were reported to have extended operation lifetimes when compared to PVC-based ISEs prepared with the same ionophore. Some of the reports also indicated that covalent immobilization of the ionophore may result in electrodes that have better detection limits and a wider pH working range (7, IO). In this study, the ionophore was covalently attached to carboxylated poly(viny1 chloride) (PVC-COOH),a polymeric
0003-2700/90/0362-1428$02.50/00 1990 American Chemical Society
ANALYTICAL CHEMISTRY, VOL. 62, NO. 14, JULY 15, 1990
Table I. Composition of Membranes” ionophore electrode 1 2
(mg)
polymer matrix (mg)
B15C5 (2.3) PVC (31.2) B15C5 (2.3) PVC-COOH
slope, mV/decade
kRh8
52 51
6.6 X 1.0 X
56
2.6 X
(31.2)
3
PVC-CONHB15C5 (34)
“11 membranes contained 66 mg of NPOE and 0.4 mg of KTClPB.
matrix containing free carboxyl groups. Recently, there has been an increased interest in using this polymer for the development of ISEs. Indeed, it has been demonstrated that, in plasticized membranes, the majority of the carboxyl groups of this polymer are not ionized (17-19). In addition, electrodes based on PVC-COOH have similar selectivity patterns to those prepared with regular PVC (17, 18). Thus far, membranes based on PVC-COOH and the ionophores tridodecylamine (18, 20-22), valinomycin (17, 18),E T H 1001 (17),nonactin (18), and a tin(1V) porphyrin (18)have been studied. The above studies demonstrate that PVC-COOH is a viable alternative to PVC in the development of ISEs. Surprisingly, to the best our knowledge, there has been no report on covalently immobilizing ionophores to this matrix for the purpose of developing ISEs. This report describes the immobilization of a relatively hydrophilic crown ether, 4’-aminobenzo-15-crown-5,to PVC-COOH. Membranes containing the covalently attached ionophore were prepared, and an ion-selective electrode was developed. This polymer-bound ionophore is expected to be resistant to leaching from the membrane and should give ISEs with increased lifetime when compared to conventional polymer-membrane-based electrodes.
EXPERIMENTAL SECTION Reagents. The following reagents were used as received: benzo-15-crown-5 (B15C5) (Parish Chemical Co., Orem, UT), chromatographic-grade poly(viny1chloride) (Polyscience, Warrington, PA), and carboxylated poly(viny1 chloride), (Aldrich Chemical Co., Milwaukee, WI). According to the manufacturer, PVC-COOH has a carboxyl content of 1.8% (w/w). Potassium tetrakis(chloropheny1)borate (KTClPB) and o-nitrophenyl octyl ether (NPOE) were purchased from Fluka (Ronkokoma, NY). Tetrahydrofuran (THF) was obtained from Fisher Scientific (Cincinnati,OH). Tris(hydroxymethy1)aminomethane (Tris) and all the inorganic salts were from Sigma Chemical Co. (St. Louis, MO). All standard solutions and buffers were prepared with deionized (Milli-Q,Millipore Corp., Bedford, MA) distilled water. Apparatus. Voltages were monitored with a Fisher Accumet (Model 810) digital pH/mV meter and recorded on a Linear (Model 1200) strip-chart recorder. Preparation of Immobilized Benzo-15-crown-5.B15C5 was derivatized first by nitration and then by hydrogenation to form 4’-aminobenzo-15-crown-5(23). The latter was coupled to PVC-COOH by means of a carbodiimide-mediated reaction. (29 mg, 0.10 mmol) was Specifically, 4’-aminobenzo-15-crown-5 dissolved in 1.0 mL of 0.100 M NaHCO3 PVC-COOH (250 mg, equivalent to 0.10 mmol of carboxyl groups) and l-ethyl-3-[3(dimethylamino)propyl]carbodiimidehydrochloride (38 mg, 0.20 mmol) were added to the solution. The reaction mixture was stirred for 18 h at room temperature. The insoluble polymer was isolated by filtration and was dried thoroughly under vacuum. Elemental analysis (Atlantic Microlab Inc., Norcross, GA) gave the following composition for the modified polymer, PVCCONH-B15C5: C, 39.69%; H, 5.01%; N, 0.33%; C1, 51.40%. These data indicate that 6.4% (w/w) of the modified polymer was B15C5. Membranes and Cell Assembly. The membranes used in these studies had the composition shown in Table I. The solvent
1429
polymeric membranes were prepared by dissolving all membrane components in 1 mL of THF. The solution was poured into a 16-mm-i.d. glass ring, and the solvent was allowed to evaporate at room temperature overnight (24). The membranes formed were cut into small-diameter disks and were positioned on an IS-561 Philips electrode body. The resulting electrodes were conditioned M KCl. All potentiometric measurements for 48 h in 1.00 X were performed with the following cell assembly: Ag-AgCll0.100 M M Tris-HC1, pH 7.511sample solutionlmembranell.00 X KClIAg-AgC1. Procedure. The calibration of the electrodes was carried out by adding, while stirring, aliquots of known concentrations of the different electrolyte standard solutions to a beaker containing 3.0 mL of a 0.100 M Tris-HC1, pH 7.5 buffer (Tris did not present any interference in the measurement). The response of the electrodes was measured by the pH/mV meter and was registered by the strip-chart recorder. All experiments were performed at room temperature. When not in use, the electrodes were stored M KCl at room temperature. in 1.00 X To obtain the calibration curve of the ISEs, the data were plotted as AE (i.e., the increase in potential with respect to the base line) vs the logarithm of the concentration of cation present in the buffered solution. Selectivity coefficients were calculated by using the fixed interference method (25). The sample solution was either 0.100 M NaCl or 0,0100 M NaCl in 0.100 M Tris-HC1, pH 7.5, to which M KC1 solution (containing 0.100 M additions of a 1.00 X M KCl solution (containing 0.0100 M NaC1) NaC1) or a 1.00 X were made, respectively.
RESULTS AND DISCUSSION Besides few exceptions (Z), plasticized PVC has been the matrix of choice in the development of polymer membrane ISEs. High solubility in the plasticized polymer is one of the desired characteristics of ionophores used in this type of electrodes. Indeed, such solubility properties reduce leaching of the ionophore from the membrane and prolong the lifetime of ISEs (6). B15C5, a relatively hydrophilic ionophore that has been employed previously in the preparation of electrodes selective for K+ (26,27), was chosen as a model ionophore to perform the studies described in this paper. To demonstrate the compatibility of the PVC-COOH matrix with the selected crown ether, electrodes were prepared containing 2.3% B15C5 in PVC or PVC-COOH (electrodes 1 and 2 in Table I). When freshly prepared, both electrodes were selective for K+ and had slopes in the range of 51-52 mV/decade. Further, the values of the selectivity coefficient, &ha, for the two electrodes were fairly similar. The above data indicate that there is little difference in the characteristics of PVC-COOH- and PVC-based polymer membrane electrodes prepared with this crown ether. This is consistent with similar observations involving these two polymers and different ion carriers (17,18). Consequently, PVC-COOH can be used as a suitable matrix for ISEs, and at the same time it can provide functional groups for covalent attachment of ionophores. Amination of B15C5 yields 4’-aminobenzo-15-crown-5 (23), which was reacted with PVC-COOH in the presence of a carbodiimide to form PVC-CONH-Bl5C5. A membrane prepared with this polymer was compared to PVC and PVC-COOH membranes impregnated with B15C5 (all membranes contained an equivalent amount of 2.3% (w/w) of the ionophore, B15C5). Figure 1 shows that the response of electrode 3 toward potassium was near-Nernstian and presented a selectivity pattern similar to the one of traditional electrodes prepared by using 15-crown-5 derivatives (28). Indeed, electrode 3 responded in decreasing order to the following cations: K+ > Rb+ > Cs+ > NH4+ > Na+ > Li+. As shown in Figure 1, electrode 3 responds preferentially to potassium over sodium. Its selectivity is slightly better than that observed with PVC electrodes based on 4’-methylbenzo-15-crown-5 and B15C5. These ISEs were shown to respond 15 times (29) and 150 times (electrode 1, Table I)
1430
ANALYTICAL CHEMISTRY, VOL. 62, NO. 14, JULY 15, 1990 I 1
T a b l e 11. L i f e t i m e S t u d i e s o f a Typical E l e c t r o d e B a s e d on I m m o b i l i z e d Benzo-15-crown-5 day
1 3 17 50 I
I
-50
-30
-40
-20
139 [ C a t i m ]
Figure 1. Selectivity pattern of the K+-selective electrode based on PVC-CONH-B15C5. Membrane composition: 34 % PVC-CONHB15C5,66% NPOE, and 0.04% KTCIPB. The membrane contains an equivalent amount of 2.3% B15C5. The electrode was exposed to the chlorkle salts of potassium (1). rubidium (2), cesium (3), ammonium (4), sodium (5), and lithium (6).
-50
-40
-30
-2c
-I0
og [Potassium] -
Figure 2. Comparison of the response of electrodes 1-3 toward potassium: (A)electrode 1, (m)electrode 2, and (0)electrode 3.
better to K+ than to Na+ (i.e., Izj& = 6.7 x and kcha = 6.6 X respectively. Electrode 3 also exhibits substantial near-Nernstian responses to rubidium and cesium, which should prohibit use of this electrode in samples containing high concentrations of these two cations. The response of electrodes 1-3 toward K+ is shown in Figure 2. This figure illustrates that immobilization of the ionophore results in electrodes with better detection limits (the limit of detection is defined according to the recommendations by IUPAC (25)). Consequently, electrode 3 has an extended linear response for K+ that ranges over almost 4 decades of concentration. Such improvement in the detection limits is not surprising in view of previous observations with a Ca2+-selectiveelectrode that employs a membrane composed of dialkyl phosphate groups covalently attached to a SBS polymer (10). It should be noted that Donnan failure appears to occur at the same concentration for all three electrodes (i.e., it is not affected by the presence of carboxyl groups on the polymer) (Figure 2). This supports literature reports that suggest that the majority of the carboxyl groups of PVCCOOH are not ionized in the bulk polymer membrane (17-19). Indeed, if negatively charged carboxyl groups were present, they should have improved the permselectivity of the membrane, and, as a result, Donnan failure should have been observed a t higher K+ concentrations. The lifetime of the electrode based on the covalently immobilized crown ether was studied (Table 11), and it was compared to that of electrode 1. The detection limits of electrode 3 remained almost constant during a 50-day period of time (only a 2.7-fold change was observed). These relatively small changes in the detection limits were also observed with other electrodes prepared with the same composition as electrode 3. The deterioration of the slopes of the electrodes with time was also studied. While electrode 3 maintained near-Nernstian slopes for an extended period of time, the slopes of electrode 1 decreased more rapidly; by using data
detection limits,
9.0 6.8 2.0 2.4
X X
M
lo4
lo*
x 10-5 x 10-5
slope, mV/decade
56 54 53 49
from three electrodes, an average change in the slope of 11% was observed over 2.5 weeks. These results suggest that, by covalently attaching the crown ether to PVC-COOH leaching of the ionophore could be reduced, as evidenced by the increased lifetime of electrode 3. It should be noted that covalent immobilization of the ionophore does not preclude leaching of the plasticizer from the polymer matrix. This may account for the small but significant change in the detection limits and slopes of electrode 3 with time. Therefore, in order to further improve the lifetime of ISEs, it may be necessary to covalently attach both the ionophore and the plasticizer to a suitable polymer. Indeed, Hobby et al. (8)and Ebdon et al. (10) came to similar conclusions with their calcium-selective electrodes. An additional explanation for the reduction of the slopes of electrode 3 may be a slow hydrolysis of the amide linkage between the carboxyl group of the polymer and 4'-aminobenzo-l5-crown-5. In conclusion, although 4'-aminobenzo-15-crown-5is water-soluble and therefore unsuitable for the preparation of liquid polymeric membranes, by covalently attaching it to PVC-COOH, functional electrodes could be prepared. This approach could be extended to other hydrophilic ionophores that previously could not be used in the development of ISEs, as long as they could be covalently immobilized on appropriate polymeric matrices. Further, because of the observed improvement in the detection limits, there is an indication that ISEs may be developed for ions unable to be determined previously with sufficient detection limits and selectivity. Finally, the development of electrodes based on ionophores covalently attached to PVC-COOH may also find application in the field of ion-selective field effect transistors (ISFETs) and enzyme sensors. Indeed, Satchwill and Harrison (22)have pointed out that PVC-COOH adheres more strongly to silica than does PVC, and therefore, it may be advantageous for the development of ISFETs. In addition, membranes based on PVC-COOH may be used in the preparation of enzyme electrodes in which the enzyme is held electrostatically by the carboxyl groups of the polymer (21,30). Alternatively, enzyme sensors may be obtained by covalently immobilizing proteins (through the free amino groups of lysine residues) to carboxyl groups at the membrane-solution interface of PVC-COOHbased electrodes (18, 20).
ACKNOWLEDGMENT We thank Allan Witkowski for technical assistance, LITERATURE CITED Arnold, M. A,; Meyerhoff. M. E. Crit. Rev. Anal. Chem. 1988, 20, 149-196. Moody, G.J.; Saad, E. E.; Thomas, J. D. R. Sel. Electrode Rev. 1988, 70,71-106. Meyerhoff, M. E.; Opdycke, W. N. A&. Clin. Chem. 1988, 2 5 , 1-47. Oggenfuss, P.; Morf, W. E.; Oesch, U.; Ammann, D.; Pretsch, E.; Simon, W. Anal. Chlm. Acta 1988, 180. 299-311. Martin, C.; Freiser, H. Anal. Chem. 1981, 53, 904-905. Oesch, U.; Simon, W. Anal. Chem. 1980, 52, 692-700. Lawton, R . S.;Yacynych. A. M. Anal. Chim. Acta 1984, 160, 149- 158. Hobby, P. C.; Moody, G. J.; Thomas, J. D. R. Ana/yst(London) 1983, 708,-581-590. Keil, L.; Moody, G. J.; Thomas, J. D. R. Analyst(London) 1977, 102, 274-280. Ebdon. L.: Eiiis. A. T.: Corfieid. G. C. Analyst (London) 1982, 107, 288-294. Voiovik, A. M.; Miroshnik, L. V.; Tolmachev, V. N. Elektrokhimlya 1985, 21. 402-404.
Anal. Chem. 1990, 62, 1431-1438 (12) Ebdon, L.; King, B. A.; Corfleld, 0.C. Anal. Proc. 1985, 22, 354-356. (13) Oka, S.; Sibazaki, Y.; Tahara, S. Anal. Chem. 1981, 53,588-593. (14) Cutler, S.0.;Meares, P.; Hall, D. G. J. Electroanal. Chem. 1977, 85, 145-16I. (15) Heineman, W. R.; Wieck, H. J.; Yacynych, A. M. Anal. Chem. 1980, 52,345-346. (16)Dong, S.;Sun, 2.; Lu, 2 . Analyst (London) 1988, 773. 1525-1528. (17) Lindner, E.; &if, E.; Zsuzsa, N.; TBth, K.; Pungor, E.; Buck, R. Anal. Chem. 1988, 60, 295-301. (18) Ma, S. C.; Chaniotakis, N. A.; Meyerhoff, M. E. Anal. Chem. 1988, 60,2293-2299. (19)Van den Berg, A,; Van der Wai, P.; SkowrBnska-Ptasinska, M.;
1431
gy; Covington, A. K.; Ed.; CRC Press: Boca Raton, FL., 1980; pp 111-130. (25) Commission on Analytical Nomenclature. plae Appl. Chem. 1975, 48, 129-132. (26) Keatlng, M. Y.; Rechnitz, G. A. Anel. Chem. 1984, 56, 801-806. (27) Suzuki, K.; Aruga, H.; Shiral, T. Anal. Chem. 1983, 55, 2011-2013. (28) Blair, T. L.; Daunert, S.; Bachas, L. G. Anal. Chlm. Acta 1989, 222, 253-26 1. (29) Petranek, J.; Ryba, 0. Anal. Chlm. Acta 1974, 72, 375-380. (30) Chen, C.-W.; Anzai, J.4.; Osa, T. Chem. phsrm. Bull. 1988, 36, 3671-3674.
Sudhoker, E. J. R.; Relnhoudt, D. N. Anal. Chem. 1987, 59,
2827-2829. (20) Taguchi, H.; Ishihara, N.;Okumura. K.; Shimabayashi, Y. Anal. Chim. Acta 1990, 228, 159-162. (21)Anzai, J.: Shimada, M.; Osa, T.; Chen, C.-W. Bull. Chem. Soc. Jpn. 1987, 60, 4133-4137. (22)Satchwill, T.; Harrison, D. J. J. €/ectroanal. Chem. 1988, 202, 75-81. (23)Ungaro, R.; El Hal. B.; Smid, J. J. Am. Chem. SOC. 1978, 98, 5198-5202. (24) Moody, G.J.; Thomas, J. D. R. In Ion-Selective Electrode Methodolo-
RECEIVED for review December 20,1989. Accepted March 30, 1990. Acknowledgment is made to the donors of the Petroleum Research Fund, administered by the American Chemical Society, for support of this research. Partial support by the Research Committee of the University of Kentucky is also acknowledged.
Synthetic Chemoreceptive Membranes. Sensing Bitter or Odorous Substances on a Synthetic Lipid Multibilayer Film by Using Quartz-Crystal Microbalances and Electric Responses Yoshio Okahata,* Gen-ichiro En-na, and Hiroshi Ebato
Department of Polymer Chemistry, Tokyo Institute of Technology, Ookayama, Meguro-ku, Tokyo 152, Japan
Speclflc adsorptions of bitter or odorous substances on a synthetlc llpld multlbllayer matrlx (2Cl,N+2Cl/PSS-) were detected by observlng frequency changes of a multlbllayercoated quartztrystal microbalance (QCM). Partitlon coefflclent (P) and diffusion constants ( D ) of these substances In the llpld matrlx could be obtained quantltatlvely by using the QCM method. There were good correlations between partltlon coefflclents of various bitter or odor substances to the synthetlc multlbllayer fllm on the QCM and the Intensity of bltter tastes or olfactory receptlons In humans: the stronger the lntenslty of a bltter substance or odorant, the greater the adsorptlon on the llpld matrlx. This Indicates that the Ilpldcoated QCM acts as a sensitive and selective sensor for M e r taste and odor. Electrlc responses (changes of membrane potential and membrane resistance) of the 2Cl,N+2Cl/PSSfllm occurred consecutively by the adsorption of these substances. The bltter or odor substance showlng the stronger lntenslty Induced membrane potential change In lower concentratlons. It was found that bitter substances having sterkally bulky molecular structures adsorb on the surface of the llpld matrlx, and the phase-boundary potentlal of the membrane is thereby changed. On the contrary, odor substances wlth relatlvely small or slender structures can penetrate Into the llpld matrlx and cause reductlon of the membrane resistance (the Increase of Ion permeablllty). The selectlve adsorptlon behavlor of bitter and odor substances by molecular shapes was conflrmed by adsorption studies of simple Co-lo hydrophoblc alcohols havlng various molecular structures.
The gustatory or olfactory cells in our bodies can detect multifarious substances in external environments. Although *To w h o m a l l correspondence should be addressed.
the molecular mechanism of the receptor of chemical substances and the transaction process in chemoreceptors are not well understood at present, the following mechanism is proposed (1-4). When chemical substances such as sugars and amino acids interact with receptor membranes, the membrane potential is changed (depolarized),and the depolarization is propagated electrically to the synaptic region or impulsegenerating area of taste cells. Recently, the reception of bitter taste in gustatory cells or the olfactory reception is proposed to be quite different from other specific receptor-induced mechanisms. For example, bitter substances and odorants are relatively hydrophobic, and the threshold concentration of general odorants is roughly determined by the partition coefficient between water and oil (5). Kurihara and co-workers reported that the surface pressures of the lipid monolayer (6) from bovine olfactory epithelium or the membrane potential of liposomes (7,8)and planar bilayer membranes (8)from a simple phospholipid are changed selectively by the addition of various bitter substances and odorants. Their magnitude of response has a good correlation with gustatory or olfactory reception in humans (68). The responses to bitter substances are seen not only in gustatory cells but also in nongustatory systems such as the Helix giant neuron, the turtle trigeminal nerve, and frog taste cells (4). The structures of bitter and odor substances are extremely diverse, and it is difficult to find a common chemical structure to these substances. In other words, it is difficult to prepare each specific receptor protein for each bitter or odor substance. From these findings the following mechanism is proposed: bitter or odor substances may be detected by direct adsorption to the lipid bilayer matrix without specific receptor proteins, which cause the membrane potential in biological cells (2,6-8). These biological findings prompted us to study chemical receptions of bitter or odor substances on the simple synthetic multibilayer-immobilized lipid f i (2Cla+2Cl/PSS-), as well as the phospholipid cast f i i(9). The adsorption step of these substances to the multibilayer film was followed by frequency
0003-2700/90/0362-1431$02.50/00 1990 American Chemical Society
