Artificial Organs 27(5):469–472, Blackwell Publishing, Inc. © 2003 International Society for Artificial Organs
Nano-Assembled Films for Taste Sensor Application *Antonio Riul Jr., *Roger R. Malmegrim, †Fernando J. Fonseca, and *Luiz H.C. Mattoso *EMBRAPA Instrumentação Agropecuária, São Carlos; and †Escola Politécnica da Universidade de São Paulo, USP, São Paulo, SP, Brazil
Abstract: An artificial taste sensor based on different types of ultra-thin films of conducting polymers (a special class of plastics that can conduct electricity) and their mixture with a lipid-like material has been able to mimic the human palate. In addition, this “electronic tongue” has been successfully employed in the analysis of tastants, sup-
pression effects, and commercial beverages throughout AC measurements (impedance spectroscopy) in a relatively low-cost, simple, and efficient way. Key Words: Nano-assembly—Taste—Sensor—Conducting polymers— Langmuir-Blodgett.
The field of nano-science and nano-technology is a broad and interdisciplinary world-wide research area, developing activities that have been growing explosively in the last few years. Through the use of nanostructured thin films of conducting polymers and a lipid-like material coating gold microelectrodes, we were able to mimic the human perception of taste by way of artificial sensors called “electronic tongues.” It is well known that the human tongue cannot discriminate each and every chemical substance it comes in contact with, but it groups all the information received in distinct patterns of response encoding the taste quality (global selectivity concept) (1–3). Therefore, unlike devices mimicking vision, hearing, and touch, which may in principle respond to a single physical stimulus, an artificial tongue must be able to recognize tastes (sweet, salty, sour, bitter, and umami) and respond to suppression effects like its human counterpart. Consequently, artificial taste sensors must have high sensitivity and stability, as selectivity is not a crucial requirement in this sort of application. To date, three types of “electronic tongues” have been reported in the literature. In the first, lipids were dispersed with polyvinyl chloride (PVC) and a
plasticizer (3–5), while in the second type chalcogenide glasses (6–8) were used as sensing units. The third one is based on pulsed voltammetry and it is composed of six different working electrodes and a silver counter electrode (9,10). It has been used in monitoring the treatment of potable water and in the analyses of citric juices and milk. These “electronic tongues” respond to the substances responsible for the basic tastes in much the same way as the biological system, and may detect differences in beverages with similar flavor. For instance, some of them can distinguish between brands of mineral water or brands of wine, in addition to detecting ionic metals in water (3,6–8). We will present here an electronic tongue composed of nanostructured films of polyaniline and polypyrrole, and their mixtures with a lipid-like material as transducers, as we have observed that a single material is unlikely to be sensitive to all basic tastes, and therefore a combination of sensors is generally used. The nano-assembled films were built using the Langmuir-Blodgett (LB) technique owing to the feasibility of depositing different conducting polymers with thickness in the order of about 2 nanometers each layer, also providing an excellent control of film thickness.
Received December 2002. Presented in part at the 2nd Latin American Congress for Artificial Organs and Biomaterials, held December 2001, in Brazil. Address correspondence and reprint requests to Dr. Luiz H.C. Mattoso, EMBRAPA Instrumentação Agropecuária, Rua XV de Novembro 1452, São Carlos (SP), CP 741, CEP 13560-970, Brazil. E-mail:
[email protected]
MATERIALS AND METHODS Six different sensing units were used: a bare interdigitated electrode, interdigitated electrodes coated with stearic acid (SA), a polyaniline oligomer 469
470
A. RIUL ET AL.
(16-mer), polypyrrole (PPy), and mixtures (1 : 1 w/w) of 16-mer/SA and PPy/SA. The synthesis of polypyrrole and the 16-mer polyaniline has been described elsewhere (11,12). The LB films were fabricated in a class 10,000 clean room using a KSV 5000 system. The spreading solutions were dispensed onto ultrapure water supplied by a Milli-Ro60 Reverse Osmosis cartridge feeding a Milli-Q system. The monolayers were compressed at a constant barrier speed of 10 mm/min until a target pressure of 25 mN/m, which was kept fixed during the whole transfer process. LB films composed of 5 layers transferred onto the interdigitated electrodes at a dipping speed of 2 mm/min were produced for each material. A detailed description of the LB film fabrication process can be found in references (13–16). AC measurements were carried out with an HP 4263A LCR Meter at the fixed frequencies of 100 Hz, 120 Hz, 1 kHz, 10 kHz, and 100 kHz. After each measurement, the electrodes were washed with copious amounts of ultrapure water. The standards for the various tastes employed were aqueous solutions prepared with ultrapure water to 5 mM NaCl (salty), 5 mM sucrose (sweet), 5 mM HCl (sour), and 0.1 mM quinine (bitter). Commercial beverages were also analyzed, and in order to obtain reproducible and comparable results the sensing units were left soaking in the solutions for at least 15 minutes prior to the measurements. A good reproducibility and stability of this sensor system has been observed over a year (17,18). RESULTS AND DISCUSSIONS Each electrode coating has a specific response for each solution analyzed, thus giving a distinct electrical signal for different concentrations of the same tastant, consistent with earlier reported work (18). Consequently, the sensor array provides a fingerprint ensuing an electrical pattern closely related with the characteristics of the sample analyzed. As a result, we are able to detect and differentiate substances quite easily with no need of counter electrode (3–10) (which might be a problem in the miniaturization of the system) and sample preconditioning processes to differentiate solutions with taste qualities. For that we use principal component analysis (PCA), which is a rigorous mathematical method statistically correlating the samples throughout a reduction in the dimensionality of the original data, avoiding redundancy of information (19). The 1st principal component (PC1) retains the majority of the important information associating the samples, the 2nd principal component (PC2) the second largest amount of Artif Organs, Vol. 27, No. 5, 2003
FIG. 1. PCA plot of the sensor distinguishing tastants (5 mM solutions of NaCl, HCl, and NaCl, and a 0.1 mM quinine solution) is shown.
information extracted, and so on. A PCA plot of the experimental data is shown in Fig. 1. It is clearly seen from Fig. 1 that tastants can be easily differentiated at 5 mM, which is below the human threshold for saltiness and sweetness (10 mM), for example (1). To illustrate, one must have 98.36% of precision to differentiate the tastants only looking on PC1 (x-axis) in Fig. 1 and 99.61% of acuteness using the first two principal components. The sensitivity of this sensor is also below other methods found in the literature (3–10) due to the physical principle of detection used, impedance spectroscopy, which allied to the properties of the ultrathin films used also makes possible the detection of non-polar and non-electrolyte substances such as coffee and sucrose. Besides, impedance measurements avoid the displacement of charged ionic species inside the sample, which might be troublesome in some applications. The sensor is also capable of differentiating several commercial beverages and substances with similar taste, as shown in Fig. 2. It is worth mentioning that the waters illustrated in Fig. 2 have their mineral and salt content differences in parts per billion. Suppression effects could also be detected, as illustrated in Figs. 3 and 4. When two compounds eliciting different taste qualities are mixed in solution, the taste of the mixture changes even when there are no chemical reactions between the materials involved (20,21). This mixture often yields a perceived taste sensation in humans less intense than the expected intensity from the simple sum of the single components
NANO-ASSEMBLED FILMS
FIG. 2. PCA plot of the sensor differentiating several commercial beverages is shown.
(20–23), similar to the behavior reported in Figs. 3 and 4. In both cases of suppression studied here, sucrose appears on the lefthand side of PC1 while quinine (Fig. 3) and NaCl (Fig. 4) appear on the righthand side of PC1. In other words, PC1 might be correlated to the tastes of the singular compounds, e.g., the more shifted to the right is the PCA data, the higher the saltiness (or sourness) will be, while the more shifted to the left is the PCA data, the higher the sweetness will be. The incremental addition of NaCl or quinine into the sucrose solution leads to a decrease in the taste
FIG. 3. PCA plot illustrating the suppression of sweetness due to an incremental sour addition is shown: 1 = sucrose 50 mM; 2 quinine 10 mM; 3 = quinine 50 mM; 4 = quinine 100 mM; 5 = quinine 500 mM; 6 = sucrose 50 mM + quinine 10 mM; 7 = sucrose 50 mM + quinine 50 mM; 8 = sucrose 50 mM + quinine 100 mM; 9 = sucrose 50 mM + quinine 500 mM.
471
intensity of the mixture. With no exception, the mixtures were always located below the pure compounds on the PC2 axis. This behavior is similar to the results previously reported for humans (20–23). Even at very low quinine addition (10 mM) to sucrose, the sensor is able to detect a shift in the “sweetness” of the mixture. Apparently, at lower NaCl concentrations the mixture is shifted slightly on the left of the pure NaCl solution, which might be indicative of the sweetness of dilute NaCl solutions (23), while at very high NaCl concentrations the mixture is shifted to the right side of the pure compound concentration, i.e., it is more “salty” than the diluted cases. Besides, the higher the NaCl concentration in the mixture, the higher the distance on PC2 between the mixture and the pure compounds. The response time for a taste measurement is almost immediate even when one goes from one solution to the other, due to a rapid change exchange between the electrolytic solution and the thin films composing the sensing units. The thickness control of the films during the sensor fabrication is a key factor to increase its sensitivity, corroborating previous results that the sensor sensitivity decreases with increasing film thickness (24). Indeed, the sensor was efficient in the detection and distinction of solutions presenting the same pH and very low molar concentrations, NaCl and sucrose for example, besides a good sensitivity to suppression effects. When compared with other devices this electronic tongue presents a higher sensitivity and simplicity of the data acquisition, distinguishing tastes and responding to suppression effects in a simple and efficient way that
FIG. 4. PCA plot illustrating the suppression of sweetness due to an incremental salt addition is shown: 1 = sucrose 32 mM; 2 = NaCl 3.2 mM; 3 = NaCl 32 mM; 4 = NaCl 320 mM; 5 = sucrose 32 mM + NaCl 3.2 mM; 6 = sucrose 32 mM + NaCl 32 mM; 7 = sucrose 32 mM + NaCl 320 mM. Artif Organs, Vol. 27, No. 5, 2003
472
A. RIUL ET AL.
might pave a new role in the quality control of foodstuffs in the food and beverage industries.
7.
CONCLUSIONS An electronic tongue comprising ultra-thin films of conducting polymers and a lipid-like material can detect the taste of liquid systems in a way similar to that observed in humans (global selectivity concept). The sensor was efficient in the detection and identification of tastants at concentrations below the human threshold and previous work in the literature. In addition, the distinction of beverages can be made even when the salt and mineral contents differ in parts per billion without the use of complex laboratory analysis throughout the simple impedance spectroscopy of an array of nanostructured films immersed in the liquid medium. A good similitude with the biological system was also observed in suppression effects, as the mixture often yielded a perceived taste sensation less intense than that expected from the single components. The obtained sensitivity might be valuable as well in the pharmaceutical industry where an unpleasant medicine has to be changed to a more acceptable taste without the exposure of humans to offensive substances during the test phase. This electronic tongue may be used continuously envisaging quality control applications in food and beverage industries. Acknowledgments: To FAPESP (São Paulo State Foundation of Support to the Research) and CNPq (National Research Center) for the financial support given. REFERENCES
8.
9.
10.
11. 12. 13. 14. 15. 16.
17.
18.
19. 20.
1. Pfaffman C. Handbook of Physiology, Vol 1, ed. J. Field. American Physiological Society, Washington DC, 1959. 2. Schiffman SS. Taste quality and neural coding: Implications from psychophysics and neurophysiology. Physiol Behav 2000;69:147–59. 3. Toko K. A taste sensor. Meas Sci Technol 1998;9:1919–36. 4. Toko K. Taste sensor. Sensors and Actuators B 2000;64:205–15. 5. Sakai H, Iiyama S, Toko K. Evaluation of water quality and pollution using a multichannel sensor. Sensors and Actuators B 2000;66:251–5. 6. Legin A, Rudnitskaya A, Vlasov Y, Di Natale C, Mazzone E, D’Amico A. Application of electronic tongue for quantitative
Artif Organs, Vol. 27, No. 5, 2003
21. 22. 23. 24.
analysis of mineral water and wine. Electroanalysis 1999; 11(10–11):814–20. Di Natale C, Paolesse R, Macagnano A, Mantini A, D’Amico A, Ubigli M, Legin A, Rudnitskaya A, Vlasov Y. Application of a combined artificial olfaction and taste system to the quantification of relevant compounds in red wine. Sensors and Actuators B 2000;69:342–7. Vlasov Y, Legin A, Rudnitskaya A, D’Amico A, Di Natale C. Electronic Tongue—new analytical tool for liquid analysis on the basis of non-specific sensors and methods of pattern recognition. Sensors and Actuators B 2000;65:235–6. Ivarsson P, Holmin S, Höjer NE, Krantz-Rülcker C, Winquist F. Discrimination of tea by means of a voltammetric electronic tongue and different applied waveforms. Sensors and Actuators B 2001;76:449–54. Krantz-Rülcker C, Stenberg M, Winquist F, Lundström I. Electronic tongues for environmental monitoring based on arrays and pattern recognition: a review. Anal Chim Acta 2001;426: 217–26. Lee JY, Kim CY. Synthesis of soluble polypyrrole of the doped state in organic solvents. Synthetic Metals 1995;74(2):103–6. Zhang WJ, Feng J, MacDiarmid AG, Epstein AJ. Synthesis of oligomeric anilines. Synthetic Metals 1997;84:119–20. Dhanabalan A, Riul Jr A, Mattoso LHC, Oliveira Jr ON. Composite Langmuir monolayers and Langmuir-Blodgett films from 16-mer polyaniline. Langmuir 1997;13:4882–6. Dhanabalan A, Mello SV, Oliveira Jr ON. Preparation of Langmuir-Blodgett films of soluble Polypyrrole. Macromolecules 1998:31(6):1827–32. Mello SV, Dhanabalan A, Oliveira Jr ON. LB films of parent Polypyrrole. Synthetic Metals 1999;102(1–3):1433–4. Riul Jr A, Mattoso LHC, Mello SV, Telles GD, Oliveira Jr ON. Langmuir and Langmuir-Blodgett-Films of parent polyaniline doped with functionalized acids. Synthetic Metals 1995; 71(1–3):2067–8. Riul Jr A, Mattoso LHC, Fonseca FJ, Taylor DM, Mello SV, Venâncio EC. Patent in Brazil (ref. number = PI 0103502–9); Mattoso LHC, Martin Neto L,Venâncio EC, Fonseca FJ, Mello SV, Riul Jr A, Taylor DM. International patent pending. Riul Jr A, dos Santos Jr DS, Wohnrath K, Di Tómmazo R, Carvalho ACPLF, Fonseca FJ, Oliveira Jr ON, Taylor DM, Mattoso LHC. Artificial taste sensor: Efficient combination of sensors made from Langmuir-Blodgett films of conducting polymers and a ruthenium complex and self-assembled films of an azobenzene-containing polymer. Langmuir 2002;18: 239–45. Beebe KR, Peel RJ, Seasholtz MB, eds. Chemometrics: A Practical Guide. New York: John Wiley and Sons, 1998. Bartoshuk LM. Taste mixtures: Is mixture suppression related to compression? Physiol Behav 1975;14:643–9. Breslin PAS. Interaction among salty, sour and bitter compounds. Trends in Food Science and Technology 1996;7:390–8. Walters DE. How are bitter and sweet tastes related? Trends in Food Science and Technology 1996;7:399–403. Bartoshuk LM, Murphy C, Cleveland CT. Sweet taste of dilute NaCl: psychophysical evidence for a sweet stimulus. Physiol Behav 1978;21:609–13. Stussi E, Stella R, De Rossi D. Chemoresistive conducting polymer-based odour sensors: influence of thickness changes on their sensing properties. Sensors and Actuators B 1997;43(1):180–5.
