Search

Taste is of five basic types, namely, sourness, saltiness, sweetness, bitterness and umami. In this review, we focus on a potentiometric taste sensor that we developed and fabricated using lipid polymer membranes. The taste sensor can measure the taste perceived by humans and is called an electronic tongue with global selectivity, which is the property to discriminate taste qualities and quantify them without discriminating each chemical substance. This property is similar to the gustatory system; hence, the taste sensor is a type of biomimetic device. In this paper, we first explain the sensing mechanism of the taste sensor, its application to beer evaluation and the measurement mechanism. Second, results recently obtained are introduced; i.e., the application of the senor to high-potency sweeteners and the improvement of the bitterness sensor are explained. Last, quantification of the bitterness-masking effect of high-potency sweeteners is explained using a regression analysis based on both the outputs of bitterness and sweetness sensors. The taste sensor provides a biomimetic method different from conventional analytical methods.

Technologies for quantifying bitterness are essential for classifying medicines. As previously reported, taste sensors with lipid polymer membranes can respond to bitter hydrochloride substances in pharmaceuticals. However, the acid hydrolysis reaction between the lipid phosphoric acid di-n-decyl ester (PADE) and the plasticizer tributyl o-acetylcitrate (TDAB) led to a deterioration in sensor responses during storage. Given the cost of transportation and preservation for commercialization, membrane components that maintain physical and chemical stability during long-term storage are needed. Here we present a membrane electrode based on hydrophobic tetrakis [3,5-bis (trifluoromethyl) phenyl] borate (TFPB) and a plasticizer 2-nitrophenyl octyl ether (NPOE) for the quantification of pharmaceutical bitterness, they maintain a stable response before and after accelerated deterioration, as well as high selectivity and sensitivity. It is a first attempt to use a completely dissociative substance to replace non-completely dissociative lipids. Our work offsets the long-term stability issue of a bitterness sensor with a negatively charged hydrophobic membrane. Meanwhile, we provide the opportunity to select surface charge modifiers for a membrane surface using ester plasticizers containing oppositely charged impurities.

A taste sensor with lipid/polymer membranes is one of the devices that can evaluate taste objectively. However, the conventional taste sensor cannot measure non-charged bitter substances, such as caffeine contained in coffee, because the taste sensor uses the potentiometric measurement based mainly on change in surface electric charge density of the membrane. In this study, we aimed at the detection of typical non-charged bitter substances such as caffeine, theophylline and theobromine included in beverages and pharmaceutical products. The developed sensor is designed to detect the change in the membrane potential by using a kind of allosteric mechanism of breaking an intramolecular hydrogen bond between the carboxy group and hydroxy group of aromatic carboxylic acid (i.e., hydroxy-, dihydroxy-, and trihydroxybenzoic acids) when non-charged bitter substances are bound to the hydroxy group. As a result of surface modification by immersing the sensor electrode in a modification solution in which 2,6-dihydroxybenzoic acid was dissolved, it was confirmed that the sensor response increased with the concentration of caffeine as well as allied substances. The threshold and increase tendency were consistent with those of human senses. The detection mechanism is discussed by taking into account intramolecular and intermolecular hydrogen bonds, which cause allostery. These findings suggest that it is possible to evaluate bitterness caused by non-charged bitter substances objectively by using the taste sensor with allosteric mechanism.

The bitterness sensor with negatively charged lipid polymer membrane has been reported to perform high sensitivity and selectivity to bitterness of medicines. However, the conventional cleaning solution cannot completely remove residual substances after measuring high concentration bitter samples. Surfactant is an important cleaning agent used for membrane materials. This paper reported the effects of four kinds of surfactants in cleaning the lipid polymer membrane of the bitterness sensor. Among these surfactants, anionic surfactant linear alkylbenzene sulfonate (LAS) showed no detectable impact on membrane integrity and a good cleaning effect for the bitterness sensor.

The saltiness enhancement effect is the effect whereby saltiness is enhanced by adding specific substances to salt (sodium chloride). Since this effect can be used in the development of salt-reduced foods, a method to objectively evaluate the saltiness with this effect is required. A taste sensor with lipid/polymer membranes has been used to quantify the taste of food and beverages in recent years. The sensor electrodes of this taste sensor have the feature of selectively responding to each of the five basic tastes, which is realized by the lipid/polymer membranes. In this study, we developed a new saltiness sensor based on the lipid/polymer membrane with the aim of quantifying the saltiness enhancement effect. In addition to the conventional components of a lipid, plasticizer, and polymer supporting reagent, the membrane we developed comprises ionophores, which selectively capture sodium ions. As a result, the response of the sensor increased logarithmically with the activity of NaCl in measured samples, similarly to the taste response of humans. In addition, all of the sensor responses increased upon adding saltiness-enhancing substances, such as citric acid, tartaric acid and branched-chain amino acids (BCAAs), to NaCl samples. These findings suggest that it is possible to quantify the saltiness enhancement effect using a taste sensor with lipid/polymer membranes.

This paper describes a preliminary test for sum-frequency generation (SFG) spectroscopy analysis of lipid polymer membranes. The SFG technique is used to analyze surface and structural information of molecules at gas-solid, gas-liquid, liquid-solid interfaces. A specially designed SFG system, which allows to analyze orientational polarization of polymer molecules at interfaces between a lipid polymer membrane and liquid/gas/solid, is being developed in our laboratory. A laser-induced damage threshold test of lipid polymer membranes was conducted in an early stage of this work.

This paper reports the improvement of a bitterness sensor based on a lipid polymer membrane consisting of phosphoric acid di-n-decyl ester (PADE) as a lipid and bis(1-butylpentyl) adipate (BBPA) and tributyl o-acetylcitrate (TBAC) as plasticizers. Although the commercialized bitterness sensor (BT0) has high sensitivity and selectivity to the bitterness of medicines, the sensor response gradually decreases to almost zero after two years at room temperature and humidity in a laboratory. To reveal the reason for the deterioration of the response, we investigated sensor membranes by measuring the membrane potential, contact angle, and adsorption amount, as well as by performing gas chromatography-mass spectrometry (GC-MS), liquid chromatography-tandem mass spectrometry (LC-MS/MS). We found that the change in the surface charge density caused by the hydrolysis of TBAC led to the deterioration of the response. The acidic environment generated by PADE promoted TBAC hydrolysis. Finally, we succeeded in fabricating a new membrane for sensing the bitterness of medicines with higher durability and sensitivity by adjusting the proportions of the lipid and plasticizers.

High-potency sweeteners are applied to low-calorie diets and bitterness-masking ingredients in pharmaceutical products. We have studied taste sensors with lipid polymer membranes based on potentiometric measurement system for high-potency sweeteners. However, the sensor also responds to astringency substances because of hydrophobic characteristics of the lipid polymer membrane. In this study, we developed a new taste sensor using a lipid polymer membrane for saccharin sodium and acesulfame potassium as negatively charged high-potency sweeteners. We optimized the quantities and types of lipids and plasticizers for the fabrication of the sensor with high selectivity and sensitivity. We succeeded in the fabrication of the new sensor, the output of which could be suppressed under −5 mV for astringency substances. Moreover, the sensor has a good sensitivity and selectivity for saccharin sodium and acesulfame potassium.

Objective taste evaluation has been much in demand in the food, beverage and pharmaceutical industries. A taste-sensing system, which is an electronic tongue with “global selectivity,” is one of the methods used for objective taste evaluation. A taste sensor electrode responds to only one of the basic tastes (saltiness, sourness, sweetness, bitterness and umami) as a change in membrane potential caused by interactions with tastants. Sweet substances are compounds with diverse chemical structures and sizes. Since the taste-sensing system is a potentiometric measurement system using a change in membrane potential, three types of sweetness sensors are required, one for sweeteners with each type of electric charge (uncharged, positively charged and negatively charged). A sweetness sensor for uncharged sweeteners has been developed. Therefore, negatively charged sweeteners, such as saccharine sodium and acesulfame potassium, were chosen as the target substances in this study. We investigated the responses of various sensor membranes using a lipid and nine kinds of plasticizers to each basic taste sample. Furthermore, not only the selectivity of the membranes but also the concentration dependence of their response to sweeteners was investigated. As a result, one of the developed sensors showed responses of over 20 mV to 5 mM saccharine sodium and 10 mM acesulfame potassium in CPA value measurement (CPA: change in membrane potential caused by adsorption). On the other hand, the sensor also showed nearly zero responses to other basic taste samples. In addition, saltiness was the only interfering taste, and the responses to target substances in relative value measurement were over 140 mV. The developed sweetness sensor had high selectivity and concentration-dependent responses. Hence, we concluded that the sensor is suitable for use as a sweetness sensor for high-potency sweeteners with a negative electric charge.

Taste evaluation technology has been developed by several methods, such as sensory tests, electronic tongues and a taste sensor based on lipid/polymer membranes. In particular, the taste sensor can individually quantify five basic tastes without multivariate analysis. However, it has proven difficult to develop a sweetness sensor, because sweeteners are classified into three types according to the electric charges in an aqueous solution; that is, no charge, negative charge and positive charge. Using membrane potential measurements, the taste-sensing system needs three types of sensor membrane for each electric charge type of sweetener. Since the commercially available sweetness sensor was only intended for uncharged sweeteners, a sweetness sensor for positively charged high-potency sweeteners such as aspartame was developed in this study. Using a lipid and plasticizers, we fabricated various lipid/polymer membranes for the sweetness sensor to identify the suitable components of the sensor membranes. As a result, one of the developed sensors showed responses of more than 20 mV to 10 mM aspartame and less than 5 mV to any other taste. The responses of the sensor depended on the concentration of aspartame. These results suggested that the developed sweetness sensor had high sensitivity to and high selectivity for aspartame.

A taste sensor using a lipid/polymer membrane, i.e., an electronic tongue with global selectivity, has been developed for objective evaluation of the taste of foods and beverages. Moreover, the taste sensor has been also contributing to safety of foods, e.g., the sensor membrane with strong hydrophobicity was used to detect sodium dodecyl sulfate (SDS), a negatively charged surfactant, which was generally added into the pesticide because of its strongly melting effect. An immersion process in monosodium glutamate (MSG) solution, called “MSG preconditioning” was needed to obtain the change in membrane electric potential caused by adsorption (CPA) for sensor membrane before measurement. However, what happened to sensor membrane during MSG preconditioning is unclear. In this paper, we examined the relationship between the CPA value and the period of MSG preconditioning. The amount of adsorbed SDS and MSG was measured to figure out whether the CPA value is related to the amount of adsorption. As a result, with the precondition process progressed, the CPA values showed concentration dependence on SDS concentration, and increased to a peak by preconditioning for one day then decreased to a stable state after that. The amount of adsorbed SDS depended on the SDS concentration but did not change with the increasing of preconditioning time. In conclusion, we revealed that the most suitable time of MSG preconditioning for the membrane for SDS was one day. The CPA value was affected by both the surface charge density and the amount of absorption.

The current study demonstrated that theasinensin A (TSA) had a potential to form the complex with hydrophobic Trp-containing dipeptides, and to reduce their membrane potential by artificial-lipid membrane taste sensor. At a 1:3 molar ratio of the 6 Trp-containing dipeptides together with TSA, we observed a significant chemical shift of the protons of the dipeptides (Δδ) to a high magnetic field, when analyzed using 1H-nuclear-magnetic resonance (NMR) spectroscopy. The Δδ values were correlated with the hydrophobicity (log P) of the dipeptides and significant correlations were obtained (P = 0.022, R2 = 0.77); e.g., Trp-Leu with the highest log P value of 1.623 among the tested dipeptides showed the highest Δδ value of 0.105 ppm for the H7 proton of Trp-Leu, while less chemical shifts were observed in theasinensin B and epigallocatechin-3-O-gallate. Diffusion-ordered NMR spectroscopy revealed that the diffusion coefficient of 3 mM of Trp-Leu (7.6 × 10−11 m2/s) at a pulse field gradient in the range 0.05–0.3 T/m decreased in the presence of 3 mM TSA (6.6 × 10−11 m2/s), suggesting that Trp-Leu forms a complex with TSA. Quantum mechanical calculations and rotating frame nuclear Overhauser effect-NMR spectroscopy provided configuration information on the geometry of the complex that Trp-Leu formed with TSA (1:1 complex) with a ΔG energy of –8.7 kJ/mol. A sensor analysis using artificial-lipid membranes demonstrated that the changes in membrane potential of 1 mM Trp-Leu (21.8 ± 1.3 mV) and Leu-Trp (5.3 ± 0.9 mV) were significantly (P < 0.001) reduced by 1 mM TSA (Trp-Leu, 13.1 ± 2.4 mV; Leu-Trp, 3.5 ± 0.5 mV; TSA alone, 0.2 ± 0.01 mV), indicating the effective suppression of hydrophobicity of dipeptides by TSA-formed complex.

This paper reports the development of membrane sensors based on an artificial lipid and plasticizers with high selectivity and sensitivity to drug bitterness by using bis(1-butylpentyl) adipate (BBPA), bis(2-ethylhexyl) sebacate (BEHS), phosphoric acid tris(2-ethylhexyl) ester (PTEH), and tributyl o-acetylcitrate (TBAC) as a plasticizer and phosphoric acid di-n-decyl ester (PADE) as an artificial lipid to optimize surface hydrophobicity of the sensors. In addition, a sensor with highly correlated bitterness sensory score was developed by blending BBPA and TBAC to detect the bitterness suppression effect of sucrose, and other bitter-masking materials. Therefore, this sensor can be used to evaluate the bitterness of various drug formulations with high accuracy. Copyright © 2009 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

Taste sensor responds to different taste qualities by unique patterns of output signals. The sensor uses lipid membranes as recognition element of tastes. Food constituents affect the membrane potential of the sensor mainly with shielding effect and/or adsorption. However suitable concentration of lipid in membrane has not been studied up to now. In this paper, we changed the concentration of lipid dioctyl phosphate (C) and trioctylmethyl ammonium chloride (T) in membranes, and investigated the response of the sensor to taste substances. Dilution of lipid improved the detection limit of the sensor to adsorptive taste substances, whereas the detection limit to non-adsorptive taste substances did not remarkably changed.

A taste sensor is composed of several kinds of lipid/polymer membranes as transducers which convert taste information to electric signal. Thus, the role of membranes is very important to detect various taste components. In this paper, we developed novel membranes which specifically respond to quinine that is typical bitter substances. These membranes were composed of hydrophobic ionic liquid such as N, N, N-trimethyl-N-propylammonium bis(trifluoromethansulfonyl)imide, 1-butyl-3-methylimidazolium hexafluorophosphate and 1-butylpyridinium hexafluorophosphate, a plasticizer, 2-nitrophenyl octyl ether and a polymer, polyvinyl chloride. In addition to quinine, they also showed response to both several kinds of alkaloids such as caffeine and strychnine, and non-alkaloid such as phenylthiocarbamide. The order of these responses was equal to that of the tongue glossopharyngeal nerve of flog. Furthermore, there were the other alkaloids which response to these membranes. Especially in these alkaloids, they showed high response to denatonium benzoate and berberin chloride which have a strong bitter taste.

Recent concern on international terrorism and weapons of mass destruction demands the development of novel analytical methods for identification and quantification of explosive molecules. In this article, we describe the development of high-performance immunosensors for detection of 2,4,6-trinitrotoluene (TNT), a prime component of the landmines and bombs used by terrorist and military forces. The immunosensors were constructed by physical adsorption and self-assembly methods, and their binding interactions with a monoclonal anti-TNT antibody were evaluated for TNT detection using the surface plasmon resonance technique. A home-made 2,4,6-trinitrophenyl-keyhole limpet hemocyanine conjugate was used for physical adsorption. A poly(ethylene glycol) hydrazine hydrochloride thiolate was used in the construction of self-assembled monolayer surface and was immobilized with trinitrophenyl-β-alanine by the amide coupling method. The immunosensors were highly selective, regenerable, rapid, and exhibited remarkable...

It is possible to evaluate the taste of foods or medicines using a taste sensor. The taste sensor converts information on taste into an electrical signal using several lipid/polymer membranes. A lipid/polymer membrane for bitterness sensing can evaluate aftertaste after immersion in monosodium glutamate (MSG), which is called "preconditioning". However, we have not yet analyzed the change in the surface structure of the membrane as a result of preconditioning. Thus, we analyzed the change in the surface by performing contact angle and surface zeta potential measurements, Fourier transform infrared spectroscopy (FTIR), X-ray photon spectroscopy (XPS) and gas cluster ion beam time-of-flight secondary ion mass spectrometry (GCIB-TOF-SIMS). After preconditioning, the concentrations of MSG and tetradodecylammonium bromide (TDAB), contained in the lipid membrane were found to be higher in the surface region than in the bulk region. The effect of preconditioning was revealed by the above analysis methods.

A taste sensor using lipid-polymer membranes has been developed to evaluate the taste of foods, beverages and medicines. The response of the taste sensor, measured as a change in the membrane potential caused by adsorption (CPA), corresponds to the aftertaste felt by humans. The relationships between the CPA value and the amount of adsorbed taste substances, quinine and iso-α acid (bitterness), and tannic acid (astringency), have been studied so far. However, that of epigallocatechin gallate (EGCg) has not been clarified, although EGCg is abundantly present in green tea as one of its astringent substances. This study aimed at clarifying the response of the taste sensor to EGCg and its relationship with the amount of EGCg adsorbed onto lipid-polymer membranes. The lipid concentration dependence of the CPA value was similar to that of the amount of adsorbed EGCg, indicating a high correlation between the CPA value and the amount of adsorbed EGCg. The CPA value increased with increasing amount of adsorbed EGCg, however, the CPA value showed a tendency of leveling off when the amount of adsorbed EGCg further increased.

This paper reports a new measurement method to detect ppb levels of noncharged organic substances using lipid/polymer membrane sensors. Noncharged organic substances have large influences on the adsorption of positively-charged lipids to negatively-charged membranes. Organic solvents (trichloroethylene) and endocrine disrupting chemicals (di-2-ethylhexylphthalate) were detected by utilizing the sensor output, which is the change of membrane potential caused by interactions among the lipid membrane, noncharged organic substances, and lipids in solution. This new potentiometric method has a possibility of detection of a trace amount of noncharged toxic substances.

A novel odor sensing system which imitates the process of reception of odorants in biological olfactory system had been developed, utilizing surface polarity controlling method as a transducer of the sensor. In biological system, odorants are received by a large number of receptors, which recognize not whole structures but partial structures of molecules. The present sensor responded partial structures of odorants and adsorption interaction of the partial structures were controlled by the surface polarity of the electrode. Sensor output signals were analyzed on the stand point of partial structures of odorants. As a result, the responses to alcohols, aromatic compounds and aromatic alcohols as odorants could be discriminated by similar molecular structures, with principal component analysis (PCA). In addition, regression models of the response to partial structures of odorants were developed by multiple regression analysis, and odor qualities of aromatic alcohols could be evaluated quantitatively by partial structures of odor molecules.

Monosodium glutamate (MSG) induces a favorable taste, termed umami. The action of MSG on lipid membranes of a taste sensor was investigated. In general, food constituents monotonously increase the electric potential of the negatively charged membranes. Contrary to this, MSG decreases the potential at lower concentration and increases it at higher concentration, thus the effect of MSG on lipid membrane is biphasic. On the decrease of the membrane potential, a model was proposed in which the dissociated carboxyl group of MSG accelerates the dissociation of phosphate group of the membrane lipid.

The bitterness of bitter substances can be measured by the change in the membrane electric potential caused by adsorption (CPA) using a taste sensor (electronic tongue). In this study, we examined the relationship between the CPA value due to an acidic bitter substance and the amount of the bitter substance adsorbed onto lipid/polymer membranes, which contain different lipid contents, used in the taste sensor. We used iso-α-acid which is an acidic bitter substance found in several foods and beverages. The amount of adsorbed iso-α-acid, which was determined by spectroscopy, showed a maximum at the lipid concentration 0.1 wt % of the membrane, and the same phenomenon was observed for the CPA value. At the higher lipid concentration, however, the amount adsorbed decreased and then remained constant, while the CPA value decreased monotonically to zero. This constant adsorption amount was observed when the membrane potential in the reference solution did not change with increasing lipid concentration. The decrease in CPA value in spite of the constant adsorption amount is caused by a decrease in the sensitivity of the membrane as the surface charge density increases. The reason why the peaks appeared in both the CPA value and adsorption amount is based on the contradictory adsorption properties of iso-α-acid. The increasing charged lipid concentration of the membrane causes an increasing electrostatic attractive interaction between iso-α-acid and the membrane, but simultaneously causes a decreasing hydrophobic interaction that results in decreasing adsorption of iso-α-acid, which also has hydrophobic properties, onto the membrane. Estimates of the amount of adsorption suggest that iso-α-acid molecules are adsorbed onto both the surface and interior of the membrane.

A multichannel taste sensor using lipid/polymer membranes has higher sensitivities to electrolytes such as sour and salty substances than nonelectrolytes such as sweet substances. The purpose of this study is to improve sensitivity to sweet taste substances. The response to sweetness is assumed to have a mechanism different from electrolytic taste substances due to its nonelectrolyte. We could make large improvement of the response to sweetness by mixing positively and negatively charged lipid/polymer membranes and utilizing their hydrophobic property.

This paper describes an application of taste sensor for detection of a trace amount of organic substances. This sensor consists of multichannel sensors with lipid membranes which are modeled on living organisms. One of characteristics of this sensor is a global selectivity. We discovered that organic substances have large influences on the adsorption of a plus charged lipid (additive in a solution) to a lipid membrane made by minus charged lipid. Using this method, some of organic substances, such as oil, Trichioroethylene and Di-2-ethylhexylphthalate, could be detected about 100ppb within five minutes. In conclusion, the taste sensor has a possibility for real-time monitoring of water quality using this new methods.

A multichannel taste sensor is composed of several kinds of lipid/polymer membranes for transforming information of substances producing taste into electric signals, which are input to a computer. The sensor output shows different patterns for chemical substances which have different taste qualities such as saltiness, sourness and bitterness, while it shows similar patterns for chemical substances with similar tastes. The sensor responds to the taste itself, as can be understood from the fact that taste interactions such as the suppression effect, which appears between sweet and bitter substances, can be reproduced well. The suppression of the bitterness of quinine and a drug substance by sucrose can be quantified. Amino acids can be classified into several groups according to their own tastes based on sensor outputs. The taste of foodstuffs such as beer and mineral water can be discussed quantitatively using the taste sensor, which provides the objective scale for the human sensory expression. The taste sensor will open doors to a new era of membrane, food and environmental sciences.

We studied and discussed the characteristics and the application of a multichannel electrode type sensor. It was based on eight membranes using various lipids as an effective and a simple potentiometric sensor for detecting water quality. The response of the sensor was calibrated using a standard solution, adjusted as an artificial river water. Especially, we researched the response patterns of the sensor to the river water at Kitakyushu in Japan. In general, cations increase the membrane potential of negatively charged membranes, while anions decrease that of positively charged membranes. The water quality of the upper stream was clearly distinguished from the water quality of the other sites. The original data obtained with the sensor were expressed on an eight-dimensional space, so they were visualized on the two-dimensional plane using principal component analysis. That is, the water quality of the upper stream was the same, and the quality of middle and down stream was different. The multichannel sensor system will contribute to control water quality and water pollution.

The present study investigates adsorption of taste substances to lipid membranes for the multichannel taste sensor. To clarify the mechanism of adsorption of taste substances to a lipid membrane, we measured adsorption quantities of taste substances, such as amino acids and quinine-HCl, with fluorometry, and the surface of the lipid membrane was analyzed with ESCA (Electron spectroscopy for chemical analysis). Furthermore, we measured potential responses of a lipid membrane and pH of monosodium glutamate and lipid solutions. The results showed that amino acids adsorb to the lipid membrane, but no specific adsorption site exists on the membrane. The adsorption occurred by weak interaction between a phosphoric acid group of lipid molecules and an amino group of amino acids. Quinine-HCl adsorbed to membrane with strong electrostatic coupling and weak hydrophobic effects to lipid molecules in the membrane.

This paper describes an application of taste sensor for detection of cyanide. The taste sensor consists of multichannel sensors with lipid membranes which are modeled on living organisms. One of characteristics of this sensor is a global selectivity. In this study, we have developed a new effective method which is able to control the charge of membranes for high sensitivity and selectivity of lipid membrane sensors. Using this method, sensors for free cyanide and complex cyanide were developed. We could detect total cyanide under the guideline limit. It was possible to detect free cyanide below the concentration 0.5ppm in river water. Measurement time is within five minutes. In conclusion, the taste sensor has a very powerful performance and advantage for real time monitoring of water quality.

Response mechanism of a membrane made of trioctylmethylammonium chloride (TOMA) to ethanol was studied. The response of the TOMA membrane to ethanol increased linearly with increasing ethanol concentration (0-20%) at constant KCl concentration. The slope of response to ethanol was-1.73 (mV/ethanol%) using the multiple regression analysis (0-20% ethanol, 1-100 mMKCI). The value of this slope calculated using the transfer activity coefficient of chloride ion from water to ethanol showed a good agreement with the experimental value. On the other hand, the results obtained in AgCl and anion exchange membrane disagreed with the theoretical values.

Interactions between a lipid-protein membrane and taste substances were investigated using SPR (surface plasmon resonance). The used membrane was the LB (Langmuir-Blodgett) multi-layered film composed of DHP (dihexadecyl phosphate) and the sugar-binding protein ConA. The LB membrane including ConA showed larger responses to almost all the taste substances, such as quinine, caffeine, sucrose and NaCl than the LB membrane without ConA. Suggestion was also made on the interactions of astringent and pungent substances with the charged lipid LB membrane.

The technique using the phenomenon called surface plasmon resonance can make it possible to detect the change of refractivity (or dielectric constant) of a solution in a thin local region (the typical thickness of the region is about 1 μm), which is very close to the surface of a thin basal metal plate. Interactions between lipid molecules and flavor substances were investigated by using the surface plasmon resonance. In this study, the effects of various kinds of flavor substances—which can be regarded as representative of the five kinds of basic tastes (sweet, salt, sour, savory, and bitter)—on the lipid membranes were examined. The lipid membrane utilized as a receptor for taste substances was the LB (Langmuir-Blodgett) multilayered film composed of dihexadecyl phosphate. It was found that almost no flavor substances had any observable interactions to the lipid membrane system, while quinine hydrochloride, the representative of bitterness, was found to be strongly adsorbed to the lipid membrane. It was shown by doing measurements for various numbers of layers of the membrane system that the quinine hydrochloride is absorbed not only onto the surface but into the internal region of the lipid layers. It was observed that the sweet substance or salty substance suppressed the adsorption of the bitter substance. Similar suppression effects can also be found in biological systems. © 1997 Scripta Technica, Inc. Electron Comm Jpn Pt 2, 80(4): 1–8, 1997

We have developed a new portable taste sensor with a lipid/polymer membrane and conducted experiments to evaluate the sensor’s performance. The fabricated sensor consists of a taste sensor chip (40 mm × 26 mm × 2.2 mm) with working and reference electrodes and a portable sensor device (80 mm × 25 mm × 20 mm). The working electrode consists of a taste-sensing site comprising a poly(hydroxyethyl)methacrylate (pHEMA) hydrogel layer with KCl as the electrolyte layer and a lipid/polymer membrane as the taste sensing element. The reference electrode comprises a polyvinyl chloride (PVC) membrane layer with a small hole and a pHEMA layer with KCl. The whole device is the size of a USB memory stick, making it suitable for portable use. The sensor’s response to tannic acid as the standard astringency substance showed good accuracy and reproducibility, and was comparable with the performance of a commercially available taste sensing system. Thus, it is possible for this sensor to be used for in-field evaluations and it can make a significant contribution to the food industry, as well as in various fields of research.

The titrable acidity of and the ethanol concentration in sake mash were measured using a multichannel taste sensor system. A positively charged membrane responded to ethanol and the electric response of the membrane showed good agreement with those determined by gas chromatography in the measurement of sake mash. It was considered that the responses were induced by a change in the activity coefficient of anions. The titrable acidity of sake mash was measured using an electrode with a negatively charged membrane, because the electric response of the membrane showed a strong correlation with the titrable acidity of sake. The result showed good agreement with the value determind by titration, and the agreement was slightly increased by multiple regression together with the electric response of the positively charged membrane. Therefore, it is concluded that this method is useful for determination of the ethanol concentration in and the titrable acidity of sake mash. The results show that the taste sensor can be used for monitoring of those values in sake fermentation.

Electric characteristics of hybrid polymer membranes composed of two lipid species were studied, where one lipid species is positively charged in aqueous solution and the other is negatively charged. As a result, it was found that the hybrid membranes responded to taste substances in different ways according to the molar mixing ratio of these two kinds of lipids, and also showed different response characteristics from those of the single-lipid membranes. The membranes with the mixing ratio around 50% exhibited the largest responses to HCl (sourness) and monosodium glutamate (umami). Moreover, good quantitative agreements with the observed data on the response electric potential were obtained using a theory describing both the changes in surface electric potential and surface charge density with taste substances.

Pungent and astringent substances have been studied using a multichannel taste sensor. The electric-potential pattern made up of eight outputs from the membranes of the sensor has information about taste quality and intensity. Pungent substances including capsaicin, piperine and allyl isothiocyanate have no effect on the electric potentials of the liquid membranes. On the other hand, astringent substances such as catechin, tannic acid, chlorogenic acid and gallic acid change the potentials remarkably. A principal-component analysis of the patterns in electric-potential change caused by the taste substances reveals that the taste quality of astringency is located between bitterness and sourness. A model for astringency is presented.

Lipid membranes are useful materials to transform information about taste substances into electric signals. A lipid monolayer membrane for a taste sensor has been prepared by adsorbing lipid materials to the hydrophobic surface of a polymer membrane. The lipid-modified membranes respond to such electrolytic taste substances as HCl (sourness) or NaCl (saltiness) with large response magnitudes and such non-electrolytic substances as caffeine (bitterness) or sugar (sweetness) with high sensitivities. Thus the present membrane has a high ability to detect taste.

A taste sensor that uses lipid/polymer membranes can evaluate aftertastes felt by humans using Change in membrane Potential caused by Adsorption (CPA) measurements. The sensor membrane for evaluating bitterness, which is caused by acidic bitter substances such as iso-alpha acid contained in beer, needs an immersion process in monosodium glutamate (MSG) solution, called "MSG preconditioning". However, what happens to the lipid/polymer membrane during MSG preconditioning is not clear. Therefore, we carried out three experiments to investigate the changes in the lipid/polymer membrane caused by the MSG preconditioning, i.e., measurements of the taste sensor, measurements of the amount of the bitterness substance adsorbed onto the membrane and measurements of the contact angle of the membrane surface. The CPA values increased as the preconditioning process progressed, and became stable after 3 d of preconditioning. The response potentials to the reference solution showed the same tendency of the CPA value change during the preconditioning period. The contact angle of the lipid/polymer membrane surface decreased after 7 d of MSG preconditioning; in short, the surface of the lipid/polymer membrane became hydrophilic during MSG preconditioning. The amount of adsorbed iso-alpha acid was increased until 5 d preconditioning, and then it decreased. In this study, we revealed that the CPA values increased with the progress of MSG preconditioning in spite of the decrease of the amount of iso-alpha acid adsorbed onto the lipid/polymer membrane, and it was indicated that the CPA values increase because the sensor sensitivity was improved by the MSG preconditioning.

Astringent substances and pungent substances were studied using a multichannel taste sensor with lipid membranes. The electric-potential pattern constructed of eight outputs from the membranes has information of taste quality and intensity. Pungent substances, such as capsaicin, piperine and allyl isothiocyanate, had no effect on the membrane potentials of the lipid membranes. On the other hand, astringent substances such as tannic acid, catechin, gallic acid and chlorogenic acid changed the potentials remarkably. A principal component analysis of the patterns in electric potential changes caused by the taste substances revealed the astringency is located between bitterness and sourness.

A taste sensor with a multichannel electrode was developed by using lipid membranes as a transducer of taste substances. The sensor can detect the taste in a similar manner to the human gustatory sensation by response patterns of electric potential to taste substances. The sensitivity, reproducibility and durability were superior to those of humans. The same taste as that elicited by some commercial aqueous drink was reproduced by making aqueous solution mixed from four kinds of basic taste substances, the concentrations of which were determined so that the electric-potential pattern of this mixed solution could agree well with the pattern by the drink. Different brands of beer were easily distinguished by the electric-potential patterns.

The mechanism of the oscillatory electric potential of a dioleyl phosphate (DOPH) layer in a single hole sandwiched by two KCl solutions, 100 mM and 5 mM, was investigated using two types of cells for axial and transverse observations. It was discovered that the electrical oscillation was related to a macroscopic behavior of the interfaces between DOPH and KCl solutions caused by a chemically driven surface tension effect.

A series of n-alkanols from ethanol to tridecanol interacted with a negatively charged lipid membrane composed of dioleyl phosphate, which exhibited a

The taste sensor was widely used in distinguishing various taste substances. However, the taste sensor has poorer sensitivity to uncharged molecules such as caffeine, a bitter substance, than to charged taste substances. In the present study, we discussed the sensitivity of caffeine detection using a taste sensor with lipid/polymer membranes that were formed with different length of methyl group of lipid, namely, tetra-n-ctylammonium bromide (R8), tetrakis-(decyl)-ammonium bromide (R10), tetradodecylammonium bromide (TDAB; R12), and tetrahexadecylammonium bromide (R16). As a result, we observed that the electric responses of the lipid membranes to caffeine were associated with the length of alkyl chain of a lipid and an optimum concentration of the lipids in membranes was also observed to enhance the sensitivity of caffeine with taste sensor.

We have developed a taste sensor consisting of a multichannel electrode with transducers composed of lipid membranes immobilized with a polymer, i.e., eight different kinds of lipid analogs mixed with poly(vinyl chloride) and plasticizer. This multichannel sensor responds to five basic taste substances in five different ways; the taste patterns composed of eight electric-potential responses could be clearly distinguished from each other for the five basic taste substances. Moreover, similar but distinguishable patterns were obtained for substances eliciting a similar taste in humans. The reproducibility is noticeably improved from those of lipid membranes which were previously investigated for application in a taste sensor. The present results imply that the multichannel lipid membrane device has the ability to sense tastes like a human being.

The local anesthetics, tetracaine, procaine and lidocaine, interacted with a negatively charged lipid membrane composed of dioleyl phosphate (DOPH), which exhibited a self-sustained oscillation of the membrane potential. The anesthetics depolarized the membrane potential when present in increasing concentrations, whereas they increased the membrane resistance at low concentrations and decreased it at high concentrations. The above results were analyzed on the basis of electrochemical theory taking into account ion flux across the membrane. The electrical characteristics are affected by both the hydrophobicity and the diffusion constant of local anesthetics within the membrane.

This paper describes a measurement method to detect a ppb level of agricultural chemicals within five minutes using a lipid/polymer membrane sensor. Noncharged organic substances have large influences on the adsorption of positively charged lipids to negatively charged membranes. In this study, noncharged agricultural chemicals in the washed vegetables (spinach, broccoli and Japanese basil) solution can be detected. These substances can be detected above 2000 ppb in these solutions. In the case of spinach and broccoli, it can be considered that agricultural chemicals inhibit the adsorption of the adsorbent in solution to the positively charged lipid membrane, because the change rate is below zero. In the case of Japanese basil the agricultural chemicals have a synergetic effect on the adsorption of the adsorbent to the lipid membrane. These results show that noncharged agricultural chemicals can be detected indirectly. We obtained a possibility that agricultural chemicals can be classified before an accurate chemical analysis. It is useful for the evaluating safety of vegetables and fruits.

In order to imitate the mechanism of biological odor reception, an integrated multi-channel odor sensor was developed here. In this study, we attempted to control the response characteristics of this odor sensor by the composition of the water membrane and surface modification technology. As a result, the addition of P cellulose, ion-exchange resin, to the water membrane enabled large response to ethanol, which has hydroxyl group in its structure, and the formation of benzene-patterned self-assembled monolayer (SAM) enhanced the sensitivity to benzene, which has aromatic ring in its structure. In addition, both technologies were also appropriate for the detection of phenethyl alcohol, which has both hydroxyl group and aromatic ring; consequently, these channels could recognize the molecular substructures.

Describes a measurement method to detect ppm level of noncharged agricultural chemicals using a lipid/polymer membrane sensor within five minutes. In the case of this new measurement method, non-charged toxic substances are detected indirectly. Sensor output is change of membrane potential caused by interaction among lipid membrane, agricultural chemicals and lipid in a solution. This, method has a possibility of detection of a trace amount of agricultural chemicals. It is useful for the evaluation of safety of vegetables and fruits.

The higher sensitivity for sweeteners can be achieved by newly developed lipid/polymer membranes. The membrane is composed of lipids such as phosphoric acid di-n-hexadecyl ester and tetradodecylammoniumbromid, and a plasticizer, dioctyl phenylphosphonate. As a result of changing electric charge of the membrane surface, the newly developed membrane shows 5-10 times higher sensitivity for sucrose than the conventional ones. We also applied the sensor to other sugars such as sugar alcohol which is used as alternative sweetness or food additives. The experimental results of other sweeteners relatively correspond to human sensory evaluation, though the sensitivity for some sugars need to be improved.