ADVANCES IN ELECTRONIC - NOSE TECHNOLOGIES

Authors Name: * G.Sujatha1, N. Dhi !a", #. A!!a$urai% a&$ D.Th!a'araja&(

Assistant Professor (Electrical Engineering), College of Food and Dairy Technology, Koduvalli, Chennai- !" P# $tudent (%"Tech, Food Technology), College of Food and Dairy echnology,Koduvalli,Chennai- !" & 'etd" Professor and (ead, De)art*ent of +eterinary ,ioche*istry, %adras +eterinary College, Chennai . Dean, Faculty /f Food $ciences, College of Food and Dairy Technology, Koduvalli, Chennai- !" *Emai): [email protected] A*STRACT: Food quality analysis is one among the most important, complex and challenging discipline in the food sector. Due to its strict interaction with the quality of life it is extremely important to improve the performances of the methods in the field of food quality analysis. Now a days electronic noses have provided a plethora of benefits to a variety of commercial industries, including the agricultural, biomedical, cosmetics, environmental, food, manufacturing, military, pharmaceutical, regulatory, and various scientific research fields using mainly metal oxide !"#$, metal oxide field effect transistor !"#F%&$, acoustic bul' ()*$ and surface #)*$ wave$ and polymer sensor. &hese sensors has unique advantage of providing fast results by identifying, characteri+ing and quantifying the target analytes of interests present in the flavor emitted by foods without destructing them. )ll these sensors are highly sensitive and inexpensive. ,owever, these sensors face a numerous problems during operations in different aspects. )mong the four sensors, polymer based on nano material may overcome the problems encountered by the environment including the temperature and humidity. Key Words: E-nose, metal oxide, acoustic and polymer sensors, as sensor, aroma detection. 1. INTROD+CTION Now, many of the food industries are loo'ing forward for the non destructive food quality analysis techniques. &he emerging non destructive food quality analysis techniques are capable of evaluating the finished products quality by analy+ing their sensory outputs which may be in the form of flavor, odor, color, texture and taste. %lectronic Nose % - nose$ seems to be a new electronic device which can report on the finished products quality by analy+ing the head space gas which is generally composed of the volatile organic chemicals that contributes to the typical flavor for the different food materials.

&he term -electronic nose. was coined in /011 by 2ardner and (artlett, /003$ who later defined it as -an instrument which comprises an array of electronic chemical sensors with partial specificity and appropriate pattern recognition system, capable of recogni+ing simple or complex odors.. ". HISTORICAL ,ILESTONES IN DEVELO-,ENT O. E/ &ose &he first real tool for measuring aromas was developed by ,artman in /043. &he sensing element was a microelectrode, a simple platinum wire of 5.1 mm in diameter, which measured the flow of current by a sensitive milli voltmeter. !etal oxide semiconductor gas sensors !"#$ were first used commercially. 6n /0758s !"# were used as household gas alarms in 9apan under the names of &aguchi or Figaro. !oncrieff /07/$ wor'ed on the concept that different coatings materials, such as polyvinyl chloride, gelatin, and vegetable fats could be capable of providing different and complementary data for the discrimination of simple and complex aromas. ,is studies were limited to the use of a single temperature-sensitive resistor, but postulated that an array with six thermistors, provided with six different coatings, could discriminate large numbers of different aromas. &he principle of ()* sensors was introduced by :ing in 101. with his ;ie+oelectric #orption Detector. (uc' et al. /074$ studied the modulation of conductivity as an answer to differentiating aromas bouquets, while Dravnie's and &rotter /074$ used the modulation of contact potential to monitor aromas. &hese studies have been considered as a first approach to aromas evaluation because of the lac' of analytical instruments. &he hydrogen sensitive ;d!"# palladium metal oxide semiconductor$ device was developed in /0<= by a group of #wedish researchers >undstr?m et al", /0<4, /005 and /00=$. &he first gas sensor based on a #)* oscillator was introduced by *ohlt@en and Dessy /0<0$. Aonducting organic polymer A;$ sensors have been under development for approximately /5 years ((odgins, 100-) since /0058s. A; sensors rely on changes of resistance by the adsorption of gas. &able: / #ensor &ypes, sensitivity and their detection principle S.No / B = 3 Se&sor T!0e !etal "xides #emi-conducting !"#, &aguchi$ !etal "xide #emiconductor field-effect sensors !"#F%&$ Aonducting polymer sensors )coustic sensors: #urface and (ul' acoustic wave #)* and, ()*$ Se&siti e ,ateria) Doped semi-conducting metal oxides #n"B, 2a"$ Aatalytic metals !odified Aonducting ;olymers "rganic or inorganic film layers Dete1tio& -ri&1i0)e Cesistance change %lectric field change Cesistance change !ass change frequency shift$

%. *IOLOGICAL AND ARTI.ICIAL OL.ACTION &he human olfactory system is more complex and contains thousands of receptors that bind odor molecules and can detect some odors at parts per trillion levels (,reer,100-) and include between /5 and /55 million receptors (Deisingh et al", !22-)" )pparently some of the receptors in the olfactory mucus can bind more than one odor molecule and in some cases one odor molecule can bind more than one receptor. &his results in a mind-boggling amount of combinations that send unique signal patterns to the human brain. &he brain then interprets these signals and ma'es a @udgment andDor classification to identify the substance consumed, based in part, on previous experiences or neural networ' pattern recognition. &he % - nose often consists of non-selective sensors that interact with volatile molecules that result in a physical or chemical change that sends a signal to a computer which ma'es a classification based on a calibration and training process leading to pattern recognition Fig. /$. &he non-selectivity of the sensors results in many possibilities for unique signal combinations, patterns or fingerprints (aldwin et al", B5//$.&he greatest advantage of using % - nose is that it can be calibrated to be reliably consistent and can give ob@ective data for important functions li'e quality and safety control. &hese instruments can also test samples that are unfit for human consumption. &he disadvantage of the % - nose is that they are affected by the environment including temperature and humidity, which can cause sensor drift.

Fig: / #equence of events that occurs in order for chemical recognition in air phase compounds to be detected in both biological and artificial olfactory system (Kauer et al", !22&) (. E / &ose SENSORS

)ll types of sensors exhibit interactions with the gas to be measured so that a series of physical andDor chemical interactions occurs when volatile compounds flow over the sensor. ) dynamic equilibrium develops as volatile compounds are constantly being adsorbed and desorbed at the sensor surface $hiers,P", 100 $.&he ideal sensors to be integrated in an electronic nose should fulfill the following criteria De*arne, et al. /00B, %ari and ,ar3i /00B, ,artlett, et al. /00=, (odgins,/004 and /00<$: high sensitivity towards chemical compoundsE low sensitivity towards humidity and temperatureE selectivity to respond to different compounds present in the headspace of the sampleE high stabilityE high reproducibility and reliabilityE short reaction and recovery timeE durableE easy calibrationE easy to process data output. Farious 'inds of gas sensors are available, but only four technologies are currently used in commerciali+ed electronic noses: metal oxide semiconductors !"#$E metal oxide semiconductor field effect transistors !"#F%&$E conducting organic polymers A;$E pie+oelectric crystals bul' acoustic wave G ()*$. "thers such as fibre optics (Dic4inson, et al"1001, 5a4aga6a, et al"100-, Eguchi, 100! and $utter and 7urs 100-), electrochemical (%ari, and ,ar3i 100! and ,altruschat, et al"100-) are still in the developmental stage and may be integrated in the next generation of % - Noses. #uch sensors can be divided into two main classes: hot !"#, !"#F%&$ and cold A;, #)*, ()*$. &he former operate at high temperatures and are considered to be less sensitive to moisture with less carryover from one measurement to another. &herefore, they should offer the best ratio of drift and lifetime to sensitivity $hiers, 100 $. !.1"etal oxide semiconductor sensors #"$%& &he metal oxide coating may be either of the n-type mainly +inc oxide, tin dioxide, titanium dioxide or iron 666$ oxide$ which responds to oxidi+ing compounds, or of the p-type mainly nic'el oxide or cobalt oxide$ which responds to reducing compounds %ielle, 1001, (uheey, 108& and #reen6ood and Earnsha6 1088$. N-type semiconductors increase the reactivity with oxidi+ing molecules whereas excited p-type promotes reactions with reducing compounds (uheey, 108& and #reen6ood and Earnsha6 1088$. &he film deposition technique further divides each sensor type into thin 7H/555 nm$ or thic' /5H=55 Im$ film !"# sensors. Film deposition includes physical or chemical vapour deposition, evaporation, or spraying for thin films, and screen printing or painting for thic' films De*arne and $an9ines /00B$. &he thin film devices offer a faster response and significantly higher sensitivities, but are much more difficult to manufacture in terms of reproducibility. &herefore, commercially available !"# sensors are often based on thic' film technologies. Due to the high operating temperature typically B55H

745JA$, the organic volatiles transferred to the sensors are totally combusted to carbon dioxide and water on the surface of the metal oxide, leading to a change in the resistance :;nsche et al", 100 $. &he mechanism is based on an oxygen exchange between the volatile molecules and the metal film. &o shift the selectivity of a metal oxide film towards different chemical compounds the film is doped with noble catalytic metals e.g. platinum or palladium$, or the wor'ing temperature is changed within the range of 45H355JA. )lthough the selectivity is also greatly influenced by the particle si+e of the polycrystalline semiconductor, the !"# sensors are usually less selective than other technologies such as A;, ()*, #)* or !"#F%&. ,owever !"# sensors are extremely sensitive to ethanol %ielle, /007 and :;nsche et al, 100 $. !.2"etal oxide semiconductor 'ield-e''ect transistor sensors #"$%(E)& ) !"#F%& sensor comprises three layers, a silicon semiconductor, a silicon oxide insulator and a catalytic metal usually palladium, platinum, iridium or rhodium$, also called the gate. ) normal transistor operates by means of three contacts, two allow the current in source$ and out drain$, and the third acts as the gate contact that regulates the current through the transistor. 6n the !"#F%& transistor, the gate and drain contacts are shortcut, giving a diode mode transistor with convenient electronics for operation, characteri+ed by an 6F-curve. &he applied voltage on the gate and drain contact creates an electric field, which influences the conductivity of the transistor. *hen polar compounds interact with this metal gate, the electric field, and thus the current flowing through the sensor, is modified. &he recorded response corresponds to the change of voltage necessary to 'eep a constant preset drain current <undstr=* et al", 100! and 100&$. &he metal oxide semi- conductor field-effect transistor !"#F%&$ sensors rely on a change of electrostatic potential. )s in the coating of !"# sensors, the gate structure of a !"#F%& sensor is either a thic', dense metal film /55HB55 nm$ or a thin, porous metal film 7HB5 nm$. &he thic', continuous metal gate responds almost exclusively to molecules that dissociate hydrogen on the catalytic metal surface. 6t is implicitly assumed that the insulator is not exposed to the ambient molecules. &he dissociated hydrogen atoms diffuse within microseconds through the metal causing a dipole layer at the metalHinsulator interface, leading to a potential change in the transistor. Detection of molecules such as ammonia or carbon monoxide is not possible with such a layer since no hydrogen atoms are released $)et> et al", 100!$. ,owever, it transpired that the latter compounds respond well when the metal gate is thinned. &he selectivity and sensitivity of !"#F%& sensors may be influenced by the operating temperature 45H B55JA$, the composition of the metal gate, and the microstructure of the catalytic metal <undstr=* et al, 10- and 100&$. !"#F%& sensors, li'e !"# sensors, have a relatively low sensitivity to moisture and 5

are thought to be very robust. ,owever, high levels of manufacturing expertise are necessary to achieve good quality and reproducibility. $challer et al",1008$. !.* +onductin or anic polymer sensors #+, %ensors& Aonducting organic polymer A;$ sensors are li'e !"# sensors, rely on changes of resistance by the adsorption of gas. &hese sensors comprise a substrate e.g. fibre-glass or silicon$, a pair of gold-plated electrodes, and a conducting organic polymer such as polypyrrole, polyaniline or polythiophene as a sensing element. &he polymer film is deposed by electrochemical deposition between both electrodes previously fixed to the substrate %ielle, 1001 and A*rani et al", 100 $. )s the conducting polymer is grown out of a solution, the deposited film contains cation sites balanced by anions from the electrolyte and the solvent residue :;nsche et al", /004 and (odgins et al", 100 and 100-)"&he cation sites probably consist of polar ions or bipolar ions which are small regions of positive charge in the polymer chain providing mobile holes for electron transport. *hen a voltage is passed across the electrodes, a current passes through the conducting polymer. &he addition of volatile compounds to the surface alters the electron flow in the system and therefore the resistance of the sensor $hiers, 100 )" &he volatiles may interact at least with i$ the polymer itself, ii$ the counter ion, or iii$ the solvent (odgins, 100-). &herefore, good selectivity in the A; sensors may be achieved by altering one of these parameters or the electrical growth of the polymer coating. 6n general, these sensors show good sensitivities, especially for polar compounds. ,owever, their low operating temperature K 45JA$ ma'es them extremely sensitive to moisture $hiers, 100 $. )lthough such sensors are resistant to poisoning ?annoni, 100 $, they have a lifetime of only about 0H/1 months. &his short life may be due to the oxidation of the polymer, or to exposure of the sensor to different chemicals that may develop contact resistances between the polymer and the electrodes. Lnli'e !"# sensors, the A; sensors are not yet widely mar'eted, and laboratory-scale manufacturing renders them expensive. &he difficulty of producing good batch- to-batch reproducibility and a pronounced drift of the response are their main disadvantages %ielle, 1001$ !.! -coustic sensors ;ie+oelectric sensors are based on a change of mass, which may be measured as a change in resonance frequency. &hese sensors are made of tiny discs, usually quart+, lithium niobate >iNb" =$ or lithium tantalite >i&a"=$, coated with materials such as chromatographic stationary phases, lipids or any

non-volatile compounds that are chemically and thermally stable #uil3ault and 7ordanel, /011, 5ieu6enhui>en and 5ederlof, /00B, (ol*3erg , /00<$. *hen an alternating electrical potential is applied at room temperature, the crystal vibrates at a very stable frequency, defined by its mechanical properties. Lpon exposure to a vapour, the coating adsorbs certain molecules, which increases the mass of the sensing layer and hence decreases the resonance frequency of the crystal. &his change may be monitored and related to the volatile present (odgins, /00<$. &he crystals may be made to vibrate in a bul' acoustic wave ()*$ or in a surface acoustic wave #)*$ mode by selecting the appropriate combination of crystal cut and type of electrode configuration :;nsche et al, /004$. ()* and #)* sensors differ in their structure. ()* are =-dimensional waves travelling through the crystal, while #)* are B-dimensional waves that propagate along the surface of the crystal at a depth of approximately one wavelength 5ieu6enhui>en et al, 100! and (ol*3erg , 100-$. &hese devices are also called Mquart+ crystal micro- balance8 NA! or N!($ because, similar to a balance, their responses change in proportion to the amount of mass adsorbed. ()* sensors vibrate with a frequency of /5H=5 ! ,+. &heir thin coating / ImH/5 nm$ is deposited by spin coating, airbrushing or in'@et printing :;nsche et al", /004, %ielle, /007, and (ol*3erg, /00<$. &he manufacturing technique includes photolithography and airbrushing, and is fully compatible with planar integrated circuits, fabrication, especially planar silicon technology. &his enables #)* structures and conditioning circuits to be incorporated on the same silicon substrate, resulting in robust and inexpensive #)* sensors Caliendo et al", /00B$. #ince pie+oelectric sensors may be coated with an unlimited number of materials, they present the best selectivity %ielle, /007 and (odgins, /00<$. ,owever, the coating technology is not yet well controlled, which induces poor batch- to-batch reproducibility %ielle, /007 and :;nsche et al, /004$. #)* sensors, though limited by the noise caused by their high operating frequency, are more sensitive than ()* sensors. ,owever, both sensors require a higher concentration of volatiles to elicit response levels comparable to other sensor types %ielle, 1001 and (odgins, 100-$. &he difficulty of integrating ()* and #)* sensors into an electronic nose resides in the more complex electronics and their high sensitivity to disturbances such as temperature and humidity fluctuations %ielle, /007 and :;nsche et al", /004$. Ta2)e: " Summar! o3 a$ a&ta'es o3 E)e1tro&i1 Nose Se&sor T!0e #.No #ensor &ype )dvantages Disadvantage 7

/ B

!etal oxides semiconducting !"#$

Fery high sensitivity, limited sensing range and rapid response.

!etal oxide field-effect #mall sensor si+e and inexpensive #ensors !"#F%&$ operating costs Aonducting polymer #ensors A;$ )mbient temperature operation, sensitive to many volatile organic compounds, short response time, inexpensive and resistive to sensor poisoning ,igh sensitivity, good response time, diverse sensor coatings, small, inexpensive, sensitive to virtually all gases

,igh temperature operation, high power consumption, sulfur and wea' acid poisoning. Cequires environmental control, baseline drift and low sensitivity to ammonia and carbon dioxide. #ensitive to humidity and temperature and sensor life is limited. Aomplex circuitry, temperature sensitive, specificity to analyte groups affected by polymeric- film sensor coating

)coustic sensors: surface and bul' acoustic wave #)* and ()*$

4. CONCL+SION )lthough these sensors are sensitive and inexpensive, they face different 'inds of problems, especially drift, noise, repeatability, environmental influence temperature and humidity$, poisoning and non linearity in sensor response. )mong the four categories discussed, the application of suitable polymer as a base material may help to overcome these problems. % - nose, with a nose H on H chip which is a single computer chip containing the polymer sensors and processing compounds for detection of flavors will suffice the requirements of various food industries in monitoring their products.

5. RE.ERENCES /. Amra&i, ,.E.H., -ersau$, #.C a&$ -a!&e. -.617748 ) high-frequency measurements of conducting polymers - Development of a new technique for sensing volatile chemicals. %easure*ent $cience and Technology, 7 /5$, /455H/45< B. *reer, H. 617798 #ense of smell - signal recognition and transductions in olfactory receptor neurons. 6n: (and3oo4 of ,iosensors and Electronic 5oses@ %edicine , Food,Eenviron*entE Cogers, %.:, %d.$E ACA press: (oca Caton, Fl, L#), 4B/-4=B. =. *art)ett, -.N., *)air. N a&$ Gar$&er,:. 6177%8 %lectronic Nose ;rinciples, )pplications and "utloo'. )#6A, /4%colloque, !ontpellier, 3<1H317 3. *a)$;i&, E.-., *ai, :., -)otto, A a&$ Dea, S 6"<118 %lectronic noses and tongues: )pplication for the food and pharmaceutical industries, $ensors, 11 /$, 3<33-3<77E

4. *a)trus1hat, H., #am0hause&, I., Oe)'e=)aus, R., Rose, :., a&$ >ah)=am0, ,., 617798 , Detection of volatile organic solvents using potentio dynamic gas sensors. Analytical Che*istry, 70 3$, <3=H<31 7. *u1=, T.,., A))e&, ..G., Da)to&, ,. 617548 , Detection of chemical species by surface effects on metals and semiconductors. 6n: $urface Effects An DetectionB ,reg*and, 7"A ., Dravnie's, )., %ds.$, #partan boo's inc.: *ashington, D.A., L#)E /-B<. <. Ca)ie&$o, C. a&$ Vero&a, E. 6177"8 #urface acoustic wave #)*$ 2as sensor. 6n: 2as #ensors, #berveglieri, 2. %d.$, Dordrecht: :luwer )cademic ;ublishers, B1/H=57 1. Deisi&'h, A.#., Sto&e, D.C a&$ Thom0so&, ,. 6"<<(8, )pplications of electronic noses and tongues in food analysis. Ant" 7" Food sci" Technol" =0 7$, 41<-753. 0. Demar&e, V., a&$ Sa&ji&?s, R. 6177"8, &hin film semiconducting metal oxide gas sensors. 6n: #berveglieri, 2. %d.$, 2as sensors. Dordrecht: :luwer )cademic ;ublishers, 10H//7 /5. Di1=i&so&, T. A., >hite, :., #auer, :. S. a&$ >a)t, D. R. A, 617758 , Ahemical-detecting system based on a cross reactive optical sensor array. 5ature, =1B, 70<H<55 //. Dra &ie=s, A a&$ Trotter, -.:. 617548, ;olar vapor detector based on thermal modulation of contact potential" 7" $ci" Anstru*., 3B, 7B3-7B<. /B. E'u1hi, #. 6177"8 "ptical gas sensors. 6n: 2as #ensors, #berveglieri, 2. %d.$, Dordrecht: :luwer )cademic ;ublishers, =5<H=B1 /=. Gar$&er, :.>.@ *art)ett, -.N. 6177(8, ) brief history of electronic noses. $ens" Actuat" 3@ che*., /1, B//-BB5. /3. Gui)2au)t, G. a&$ :or$a&, :. ,. 617AA8 )nalytical uses of pie+oelectric crystal: ) review. Aritical Ceviews, Analytical Che*istry, /0, /HB1 /4. Hartma&, :.D. 6174(8 ) possible method for the rapid estimation of flavors in vegetables. Proc" A*er" soc" hort" sci", 73, ==4-=3B. /7. Ho$'i&s, D. 617748, ) sixth sense. Dairy industries international, 75, =BH== /<. Ho$'i&s, D. 617798, &he %lectronic nose: #ensor array-based instruments that emulate the human nose. 6n: &echniques for analy+ing food aroma, !arsili, C. %d.$, New yor': !arcel De''er inc., ==/H=</ /1. Huhee!, :. E. 617A%8 &he #olid #tate. 6n: 6norganic Ahemistry, ,uheey, 9. %. %d.$, New Oor': ,arper P Cow ;ublishers, B5/HB5= /0. #auer,:., >hite,>., Tur&er, T. a&$ Ta)amo, * 6"<<%8, ;rinciple of odor recognition by olfactory system applied to detect low concentration explosives, &ufts Lniversity #chool of !edicine. B5. #i&', >. H. :R. 6175(8 ;ie+oelectric sorption detector. Analytical Che*istry, =7, /<=4H/<=0 B/. #oh), D. 6177"8 "xidic semiconductor gas sensors. 6n: 2as #ensors, #berveglieri, 2. %d.$, Dordrecht: :luwer )cademic ;ublishers, 3=H11 BB. Lu&$strBm, I., Shi arama&, ,. S., S e&sso&, C. a&$ Lu&$= ist, L. 617948, )hysics letters "5, 44 C 49. B=. Lu&$strBm, I., S0etD, A., >i&Euist, .., A1=e)i$, +. a&$ Su&$'re&, H. 6177<8 Aatalytic !etals )nd Field-%ffect Devices ) Lseful Aombination. $ensors and actuators, ,1, =/ 3$,/4HB5

B3. Lu&$strBm, I., S e&sso&, C., S0etD, A., Su&$'re&, H. a&$ >i&Euist, .. 6177%8 From ,ydrogen #ensors &o "lfactory 6mages H &wenty Oears *ith Aatalytic Field-%ffect Devices.$ensors And Actuators ,1, /=H/3, /7HB= B4. ,o&1rie33, R.>. 617518, )n instrument for measuring and classifying odors. 7" A))l" Physiol", /7, <3B-<30. B7. ,ari, C. ,. a&$ *ar2i, G. *. 6177"8, %lectrochemical 2as #ensors. 6n: 2as #ensors, #berveglieri, 2. %d.$, Dordrecht: :luwer )cademic ;ublishers, =B0H=7 B<. ,ie))e, -. 617758 %lectronic Noses: &owards the ob@ective instrumental characteri+ation of food aroma. Trends in Food $cience Technology, $)ecial Assue /n Flavor Perce)tion , <, 3=BH3=1 B1. Na=a'a;a, ,., #a;a2ata, S., Nishi!ama, #., +tsu&omi!a, #., Famamoto, I., >a$a, T., Famashita, F. a&$ Famashita, N. 617798, )nalysis of constituents in mixed odor vapors using an array of the &-)/B5= chemiluminescence-based sensors. 6n: ;roceedings of the =rd %uropean Aonference on optical chemical sensors and biosensors, :un+, C. % 2uest %d$, %uropt C$ode 666, Qurich. )msterdam, >ausanne: %lsevier, <7 B0. Nieu;e&huiDe&, ,. S. a&$ Ne$er)o3, A. :. 6177"8 #ilicon based surface acoustic wave gas sensors. 6n: #ensors and sensory systems for an %lectronic nose, 2ardner,9. *. and (artlett, ;. N. %ds$,. Dordrecht: :luwer )cademic ;ublishers, /=/H/34 =5. Shiers, V. -. 617748 %lectronic nose technology- %valuations and developments for the food industry. 6n: Aonference ;roceedings /004, !aarssen %d.$, Food 6ngredients %urope: Fran'furt: !iller Freeman &echnical, /01HB55 =/. S0etD, A., >i&Euist, .., Su&$'re&, H. a&$ Lu&$ StrBm, I. 6177"8 Field effect gas sensors. 6n: 2as #ensors, #berveglieri, 2. %d$,. Dordrecht: :luwer )cademic ;ublishers, B/0HB<0 =B. Sutter, :. ,. a&$ :urs, -. C. 617798 Neural networ' classification and quantification of organic vapors based on fluorescence data from a fiber-optic sensor array. Analytical che*istry, 70, 147H 17B ==. >oh)tje&, H. a&$ Dess!, R. #urface acoustic wave probe for chemical analysis. 66. 2as Ahromatography Detector. Analytical Che*istry, 4/, /374H/3<5 =3. >oh)tje&, H. a&$ Dess!, R. 617978 #urface acoustic wave probe for chemical analysis. 666. &hermo-mechanical ;olymer )naly+er. Analytical Che*istry, 4/, /3<5H/3<1 =4. >G&s1he, L. .., Vui))eumier, C., #e))er, +., *!3ie)$, ,. -., ,a!, I. -. a&$ #ear&e!, ,. :. 617748 #cent Aharacteri+ation: From human perception to %lectronic nose. ;roceedings "f &he /=th 6nternational Aongress "f Flavors, Fragrances and %ssential "ils, /4H/0 "ctober /004,6stanbul, &ur'ey, =, B04H=/= =7. Ha&&o&i, ,. 617748 ;reliminary results of employ of an artificial nose for the evaluation of cheese. $cien>a E Tecnica <attiero-Casearia, 37, B<<HB10

10