Sensors: From Biosensors to the Electronic Nose

Senso s: F om Biosenso s o he Elec onic Nose By Diego L. Ga cía-González and Ramón Apa icio Ins i u o de la G asa A da. Pad e Ga cía Teje o, 4, 41012 Se illa Spain E-mail: [email protected] CONTENTS 1. In oduc ion. 2. Senso y quali y. 3. Senso s. 3.1. Me al oxide semiconduc o senso s (MOS). 3.2. Me al oxide semiconduc o ield-e ec ansis o (MOSFET) senso s. 3.3. Conduc ing polyme senso s. 3.4. Acous ic senso s. 3.5. Biosenso s. 4. The elec onic nose. 5. The elec onic ongue. 6. Da a analysis. 6.1. P e-p ocessing da a analysis. 6.2. Mul i a ia e da a analysis. 6.2.1. Clus e analysis. 6.2.2. Fac o ial analysis. 6.2.3. Disc iminan analysis. 6.2.4. Reg ession analysis. 6.3. A i icial neu al ne wo ks. Re e ences. RESUMEN Senso es: De los biosenso es a la na iz elec ónica. Los ecien es a ances en los sis emas de senso es han pe - mi ido el desa ollo de nue as aplicaciones en muchos campos ecnológicos. Es e a ículo de e isión desc ibe el es ado ac ual de es a nue a ecnología, con especial én asis en las aplicacio- nes alimen a ias. El diseño, la ecnología y el mecanismo senso- ial de cada ipo de senso son analizados en el a ículo. También se desc iben las p incipales ca ac e ís icas de la na iz y la lengua elec ónica (senso es de sabo ). Finalmen e, se desc ibe b e e- men e el uso de algunos p ocedimien os es adís icos en sis emas de senso es. PALABRAS-CLAVE: A he nálisis senso ial – Biosenso es - Na iz elec ónica – Senso es. SUMMARY Senso s: F om Biosenso s o he Elec onic Nose. The ecen ad ances in senso de ices ha e allowed he de eloping o new applica ions in many echnological ields. This e iew desc ibes he cu en s a e-o - he-a o his senso echnology, placing special emphasis on he ood applica ions. The design, echnology and sensing mechanism o each ype o senso a e analysed. A desc ip ion o he main cha ac e is ics o he elec onic nose and elec onic ongue ( as e senso s) is also gi en. Finally, he applica ions o some s a is ical p ocedu es in senso sys ems a e desc ibed b ie ly. KEY-WORDS: Biosenso s – Elec onic nose – Senso s –Sen- so y analysis. 1. INTRODUCTION Fo ages, he e has been much in e es in he design o de ices o he sensing o ood la ou . The de ices, o ins umen s, a e claimed o ope a e on p inciples simila , in many aspec s, o he human ol ac o y sys em. Today, he me hodologies de o ed o he senso y assessmen o oods u s a e based on ei he he classical panels o ained human beings o he analysis o some chemical compounds by headspace-gas ch oma og aphy (-mass spec ome y) o he global e alua ion o he odou in ensi y o ola ile compounds by he eme gen senso echnologies. The expe s’ panel, usually he only o icial me hod o he senso y e alua ion, allows he senso y quali ica ion o oods u s by hei own po en ial consume s, as well as he e alua ion o all he senso y pe cep ions in only one analysis. I has, howe e , he g ea disad an age o being a leng hy and expensi e me hodology whose inal esul depends on many ac o s as, o example, he panellis s’ aining and he speci ic ocabula y used in he senso y analysis. An al e na i e me hod is he iden i ica ion and quan i ica ion o he chemical compounds esponsible o he senso y pe cep ions. Non- ola ile compounds a e mainly ocused on phenolic compounds and aglycons, esponsible o as ing pe cep ions (as ingen , bi e and pungen ) (Mo ales and Tsimidou 2000), while ola ile compounds a e esponsible o he a oma o oods u s (Apa icio e al . 1996). The i s se o compounds is quan i ied by high pe o mance liquid ch oma og aphy (HPLC) (Mon edo o e al . 1992) while he quan i ica ion o ola ile compounds can be ca ied ou by a ious analy ical echniques (Mo ales e al . 1992). The la e can be summa ised as: 1. hose no in ol ing concen a ion: • di ec injec ion. • s a ic headspace 2. hose wi h concen a ion-dis illa ion. • simul aneous dis illa ion-ex ac ion. • dynamic headspace/pu ge-and- ap. • supe c i ical luid ex ac ion. • solid phase mic o ex ac ion G asas y Acei es 96 Vol. 53. Fasc. 1 (2002), 96-114 The use ulness o hese echniques in senso y analysis is based on he ac ha ola ile compounds a e esponsible o he oods u odou s. Thus, an in e es ing app oach has been ca ied ou in i gin oli e oil o es ablish he ela ionships be ween ola ile compounds and la ou (Apa icio e al. 1996), and be ween phenols and as e (Mo ales and Tsimidou 2000) using he ma hema ical algo i hm o he s a is ical senso y wheel (Apa icio e al . 1994). These echniques can assess he senso y quali y ou ine es s bu canno be used on-line because hey a e ime-consuming and hey p e iously need o a sample p e- ea men o a ola ile concen a ion phase. An al e na i e, ecen ly de eloped, is he use o senso s ha ha e been widely used in many ields o analy ical chemis y. Concep ually speaking, his kind o senso is a de ice ha is able o gi e a signal p opo ional o he physical o chemical p ope y o which he de ice esponds (K ess-Roge s 1997); o example, he senso s ansduce he signals p oduced by magne ic, he mal and chemical ields. Based on hese ac s, he mos common de ini ion is ha a senso is a single ansduce in which, due o a physical e ec , a comple e ansduc ion p ocess akes place (Middelhoek 2000). The elec onic in eg a ion o a ious senso s inside one se cons i u es an a ay o senso s, o he so-called elec onic nose, ha is he kind o low cos s uc u e habi ually used o analyse he a oma o he oods u s. The senso s ha e some ad an ages o e o he ci ed echniques. They do no use eagen s o need o any kind o p e- ea men , hey ha e good le els o sensi i i y and speci ici y, and besides hey a e apid supplying esul s o he analys . This is a non-des uc i e echnique ha can be used in an on line quali y con ol du ing he manu ac u ing p ocess. Bu senso s ha e ob iously disad an ages mainly ocused on he ep oducibili y, eco e y, and he e ec ha humidi y and empe a u e ha e on he senso esponses. The analysis o hese aspec s, ad an ages and disad an ages o he senso echnologies, oge he wi h he applica ions o he di e en kinds o senso s in ood science a e he objec i es o his e iew. Finally, he e is a schema ic analysis o he ma hema ical p ocedu es used o analyse he senso esul s as well as he applica ion ields and he e e ences. 2. SENSORY QUALITY Senso y quali y consis s on a se o quali y a ibu es, ha can be de ec ed h ough sense o gans, like appea ance, size and shape, colou , iscosi y, kines he ic sensa ions, ac ile sensa ions, and chemes he ic sensa ions (s imuli elici ed om some chemical i i an s), odou and as e (Ange osa 2000). Colou and la ou a e, howe e , he sensa ions ha mainly con ibu e o ood accep abili y by consume s (Mo ales and Tsimidou 2000) o when moni o ing ood indus y p ocesses (Table I). Fla ou is a complex sensa ion consis ing p ima ily o odou and as e, bu is complemen ed by ac ile and kines he ic sensa ions (Reineccious 1993). The impo ance o la ou in he global senso y quali y lies in he ac ha i plays a c ucial ole in ood ecogni ion, selec ion, accep ance and nu i ion, as i is pa ly esponsible o aiding he diges ion o ood humans (Enso 1989). The main human senso con ibu ing o la ou is he ol ac ion sys em ha de ec s he odo an compounds. The basic cha ac e is ics o odo an molecules a e o ha e low ela i e molecula masses (up o app oxima ely 300 Da), o be small and pola , and also o be hyd ophobic (C a en e al. 1996). In he nasal mucus, he e a e small odou an binding p o eins (OBPs) ha accommoda e he hyd ophobic molecules in solu ion (B ee 1997). The OBPs wo k as binde s o he odou an compounds and s a e s o he ansduc ion eac ion. A e c ossing he mucus laye , he odou an s each he chemosenso y cilia y memb ane o he ol ac o y neu on in which he G- ecep o p o eins a e loca ed (Ga dne and Ba le 1994). Mo e han 100 million ol ac o y cells wi h hese ecep o p o eins a e inside he ol ac o y epi helium. The ecep o s in e ac wi h odou an molecules and gene a e a second se o messenge s ha cause exci a ion in he neu ones. The signal is p ocessed in he ol ac o y bulb and hen he in o ma ion is ansmi ed o he limbic sys em whe e he associa ions be ween odou in o ma ion and ecalled memo ies ake place (Kelle e al . 1999). The in o ma ion o he ecep o s is also ansmi ed o he ce eb al co ex whe e is combined wi h he as e o p oduce he la ou sensa ion. On he o he hand, he as e ecep o s a e mainly si ua ed on he ongue su ace, o e he en i e o al Table I Applica ion o he senso y panel in di e en c i ical poin s o he ood indus y Moni o ing o he ood ipening (e.g., wine, cheese) Quali y assu ance o he selec ed aw oods Moni o ing o he cooking p ocesses Moni o ing o he e men a ion p ocesses Moni o ing o he indus ial p ocesses (e.g., la ou ing, blending, colou ing). Moni o ing he s o age o oods and ood p oduc s (e.g., eshness and ageing con ol) Quali y assu ance o he manu ac u ed ood p oduc s Moni o ing o he p oduc -packaging in e ac ions Moni o ing o he o e all quali y o he inal ood o ood p oduc . Vol. 53. Fasc. 1 (2002) 97 ca i y, down he oesophagus and e en on he achea and la ynx. When an app op ia e soluble subs ance eaches he as e cell, he in e ac ion o he molecule wi h he G- ecep o p o ein leads o chemical changes ha esul in a neu al impulse (Winquis e al. 1999). The esul is he as ing pe cep ion ha has ou basic as es (bi e , swee , sal and sou ) al hough he senso y assesso s a e able o dis inguish be ween dozens o di e en as es, o ins ance quali ying i gin oli e oil (Apa icio e al . 1994). 3. SENSORS Today, se e al comme cial senso s a e a ailable on he ma ke besides he g ea numbe o p o o ype senso s ha a e being designed by esea ch ins i u ions yea by yea . All hese ypes o senso s exhibi physical and chemical in e ac ions wi h he chemical compounds when hey low o e , o a e in con ac wi h, he senso s. The choice o senso s is hence qui e la ge, so ha we ha e classi ied hem in o b oad classes ollowing an a bo escen s uc u e. The odou senso s, as ing senso s and biosenso s, cons i u e he i s se . The la e wo g oups a e he esul o a e y ecen esea ch, and hence he e is no a axonomic s uc u e o hese senso s ye . The senso s o he o me g oup can howe e be clus e ed in o wo g ea g oups acco ding o hei wo king empe a u e: ho senso s and cold senso s. The g oup o he so-called “ho senso s” mainly clus e s he di e en kinds o me al oxide gas senso s (based on doped o undoped semiconduc o s), while he g oup o he “cold senso s” clus e s he piezoelec ic c ys al senso s (bulk acous ic wa e senso s, su ace acous ic wa e senso s) and he conduc ing o ganic polyme senso s. Figu e 1 shows schema ic diag ams o he main odou senso s while Table II shows he physical esponses o hese senso s o he concen a ion o ola ile compounds. When he analys s wo k wi h an a ay o senso s o he same class (e.g., me al oxide gas senso s made wi h doped plus undoped semiconduc o s) o combining di e en classes (e.g., conduc ing polyme senso s plus piezoelec ic c ys al senso s), he esul ing ins umen is habi ually called “elec onic nose”. Fu he mo e, Winquis e al . (1999) ha e ecen ly gone beyond desc ibing he combina ion o an elec onic nose and an elec onic ongue. 3.1. Me al oxide semiconduc o senso s (MOS) These senso s a e he mos comme cially ex ended in he sensing sys ems. They consis o a ce amic o me ( ubula o la ) hea ed by wi e and coa ed wi h a me al oxide semiconduc ing ilm. Fo his eason, hey a e usually called me al oxide, o ce amic, gas senso s (Schalle e al . 1998) al hough hey can also be called Taguchi and Figa o because o he in en o and he i s company espec i ely. The oxide coa ing may be ei he n- ype o p- ype semiconduc o s. The n- ype semiconduc o (SnO2, ZnO, In2O3, WO3, Fe2O3, Ga2O3, TiO2) esponds o oxidising compounds and i s he mal o pho oly ic exci a ion esul s in an excess o elec ons ha inc eases he eac i i y wi h oxidising molecules. The p- ype semiconduc o (oxides o nickel o cobal ) esponds o educing compounds and i s exci a ion p omo es eac ions wi h educing compounds due o an elec on de iciency in i s alence band. These me al oxides (mos ly SnO2, Ga2O3 and TiO2) can also be doped wi h me als such as Cu (Zhang and Liu 2000), P plus Nb (Comini e al . 2000) and Sb, Bi, Fe, Au, Cd (Mo imi su e al . 2000), o me al oxides o Co, C , La, Ce (Fukui and Ka suki 2000), Mg, Al (Xu e al . 2000), Ru, Rh and I (Lang e al. 2000) among o he s . These doping me als help o diminish he humidi y and empe a u e dependence (Fukui and Ka suki 2000; Mo imi su e al. 2000), and o imp o e he sensi i i y and he selec i i y (Comini e al. 2000; Mielle 1996). Fla MOS senso Tubula MOS senso MOSFET senso CP senso BAW senso SAW senso Figu e 1 Schema ic diag ams o he main odou senso s: Fla and ubula me al oxide semiconduc o (MOS), me al oxide semiconduc o ield e ec ansis o (MOSFET), conduc ing polyme (CP), bulk acous ic wa e (BAW) and su ace acous ic wa e (SAW) senso s. Table II Physical p ope ies o he senso s desc ibed in he e iew Senso Physical p ope y Me al-oxide senso s (MOS, MOSFET) Conduc ing polyme senso s Elec ochemical senso s Acous ic senso s (SAW, BAW, QMB, Can ile e ) Calo ime ic senso s Op ical senso s Resis ance and impedance Resis ance and impedance Conduc ance, in ensi y and ol age Mass and equency shi Tempe a u e Op ical p ope ies 98 G asas y Acei es The me al ilm hickness u he classi ies senso s in o hin ilm (6-1000 nm) and hick ilm (10-300 µm) senso s. The hickness is only ela ed o he echnique used (physical o chemical apou deposi ion o hin ilms e sus sc een p in ing o pain ing deposi ions o hick ilms) bu also wi h he esponse ( as e o hin ilms), sensi i i y (highe o hin ilms), and ep oducibili y (highe o hick ilms) and cos (much mo e low o hick ilms). The sc een- p in ing echniques, based on a plana in eg a ed mic oelec onic echnology, allow ob aining low size senso s wi h high pe o mance in a ious applica ion ields. Ne e heless, sc een-p in ed ce amic gas senso s need imp o emen s as lowing he powe consump ion, ge ing bes selec i i y and acili a ing moun ing echnology. These aspec s a e eached wi h mic omachined me al oxide gas senso s, whe e sensi i e laye is deposi ed in a hin dielec ic ilm o low he mal conduc i i y, ge ing isola ion be ween subs a e and gas sensi i i y hea ed a ea (Simon e al. 2001). The mechanism o sensing o hese senso s is ela ed o he semiconduc o beha iou when exchanging oxygen be ween he ola iles and he me al ilm. In non-ope a ing, when only he ca ie gas (e.g., syn he ic ai ) lows in he senso chambe , he oxygen is adso bed on he su ace and inside he me al coa ing, illing he senso la ice acancies, and la e oxygen a ac s ee elec ons om he semiconduc o ma e ial wi h which he senso conduc ance inc eases. In ope a ion, when he in e ac ions be ween adso bed ola ile molecules and me al ilm ake place, he conduc ance dec eases due o he eac ion wi h he adso bed oxygen. Two mechanisms, ha can occu simul aneously, ha e been desc ibed in his phase (Schalle e al . 1998). A ela i ely low empe a u es, when educing ola iles compounds a e adso bed on he su ace, he cha ges a e ans e ed be ween he ola iles and adso bed oxygen. A high empe a u e, howe e , he loaded oxygen eac s wi h he adso bed ola ile compounds, and conduc ance dec ease in a non-linea manne . Zhang and Liu (2000) ha e s udied he pa icula cases o NO2 and CO2 o e a ious ansi ion me als. Fu he mo e, he au ho s ha e sugges ed some mechanisms ha p oduce mo e oxygen acancies in he ma e ial, leading o a dec ease in he ee elec on concen a ion and an inc ease in sensi i i y o he ma e ial (e.g., he subs i u ion o in ions o coppe ions). Recen ly o he measu emen pa ame e s ha e been s udied, in addi ion o he elec ical esis ance change. The empe a u e d op du ing he sensing eac ion has been p oposed (Liu e al . 2000; Takada 2000). The empe a u e dec ease is caused by a change in he he mal conduc i i y o he senso s and i is ela ed o he eac ion be ween educing gases and su ace oxygen ad-ions. They ha e a wo king empe a u e be ween 200oC and 400oC (Simon e al. 2001), and he o ganic ola iles a e comple ely combus ed o wa e and ca bon dioxide when se les he senso su ace. These small size senso s a e obus , hey ha e a qui e good sensi i i y, hei signal p ocessing is simple, and hey ha e a e y low cos (Simon e al. 2001). Thei sensi i i y is, howe e , a ec ed by humidi y and poisoning ma e ials. Al hough he senso s a e made o ma e ials ela i ely esis an o humidi y and ageing, he humidi y is s ill he main p oblem as he senso sensi i i y dec eases when humidi y inc eases. Ano he d awback is he possibili y o he senso poisoning by sulphu compounds (e.g. H2S o SOx) ha p oduces he o al o pa ial i e e sible des uc ion o he senso . Sulphu is known as an inhibi o o me al and oxide ca alys s. On me als, sulphu chemiso bs an unoccupied o bi al o he me al ia i s occupied lone pai o bi al, while, on oxides, sulphu is subs i u ed o la ice oxygen a oms. This i e e sible p ocess can be caused by he p esence o sulphu compounds o some acids (e.g. ace ic) in he samples, e.g. cheeses, wines, inega s, c uci e ous ege ables o lampan i gin oli e oils. E hanol also a ec s o senso bu only p oducing a empo a y blinding e ec o any o he analy es p esen in he sample. O he undesi able cha ac e is ics o hese senso s a e: hei slow baseline eco e y when compounds wi h high molecula weigh a e analysed, hei high wo king empe a u e ha makes hem inapp op ia e in en i onmen s con aining lammable chemicals, hei poo speci ici y and selec i i y, and he senso d i mainly caused by sensi i i y loss. Mos o he me al-oxide senso s show a non-linea esponse o a gi en chemical compound o odou . Se e al a emp s o ind a sa is ac o y Figu e 2 Responses o a MOS senso (Alpha MOS 4000) o i gin oli e oils cha ac e ised by di e en le els o oxida ion. Oxidized oli e oil samples we e p epa ed placing i gin oli e oil samples in he beam o an ul a iole ligh o di e en days (0, 3, 8, 12 days). Vol. 53. Fasc. 1 (2002) 99 ela ion be ween he senso esponses and he analy e concen a ion o he odou in ensi y ha e been done. The o mula R=1/ACα (whe e, R is he esponse in ensi y, A and α a e cons an , and C is he concen a ion) has been p oposed o explain he senso esponse a mode a e concen a ion and below he sa u a ion le el o he senso (Wilson e al. 2000). Khol e al . (2000) ha e also exp essed he senso esponse as R-R0/R0=K(C)1/2 (whe e K is a senso -dependen cons an ). Mielle e al . (2000) ha e ound ano he ela ion be ween he compound concen a ion and he senso esponse: R-R0/R0=10(a +b)logC, whe e “a” and “b” a e cons an s o each compound. The applica ion o hese senso s has been mainly ocused on he de ec ion o gas leakage, combus ible and oxic gases (Shu me and Ga dne 1992; Xu e al. 2000; Ménil e al. 2000; Simon e al. 2001) bu hei applica ions on ood echnology ha e always been o ele ance. Thus, hese senso s ha e been applied o de ec ion o ola ile compounds o di e en oods u s as co ee (Ga dne e al. 1992), milk (Sbe eglie i e al. 1998), s awbe y (Hi sch elde e al. 1998), juices (Bazemo e e al. 1997), mea (B aggins e al. 1999) and edible oils (Bazzo e al. 1998). Figu e 2 shows he esponses o a MOS senso (AlphaMOS 4000) o i gin oli e oils cha ac e ised by di e en le els o oxida ion. 3.2. Me al oxide semiconduc o ield-e ec ansis o (MOSFET) senso s These senso s comp ise h ee laye s, a silicon semiconduc o , a silicon oxide insula o and a ca aly ic me al (Pd, P , I , Rh) called he ga e (Schalle e al. 1998). This design co esponds o he ansis o s designed by Lunds öm in 1975 (Middelhoek 2000). This kind o ansis o s ope a es by means o h ee con ac s, wo allow he cu en in (sou ce) and ou (d ain), and hi d ac s as he ga e and egula es he cu en h ough he ansis o . When a ol age is applied on he ga e, an elec ic ield is gene a ed which a ec s he ansis o conduc i i y. I apou (pola compounds) in e ac s wi h he ga e, hen he cu en lowing h ough he ansis o (senso ) changes and a shi o he conduc ance is p oduced. The in e ac ion be ween he apou and he ga e depends on he ga e s uc u e; i can be hin, made o a po ous me al ilm (6-20 nm), o hick, buil wi h a dense me al ilm (100-200 nm). The la e kind o ga e ( ansis o ) wo ks well wi h apou s con aining molecules ha can dissocia e hyd ogen, because hese a oms lead o a po en ial change in his kind o ansis o . Almos all kind o compounds, howe e , espond well when he ga e is hin, due o he p obable mechanisms o ol age shi s desc ibed by Lunds öm e al . (1975; 1990) and Spe z e al . (1992). MOSFET a e obus senso s wi h low sensi i i y o humidi y bu whose selec i i y and sensi i i y can be a ec ed by he wo king empe a u e (75-200 oC), kind o me al ga e (as desc ibed) and mic os uc u e o he ca aly ic me al. Thei obus cha ac e is ic makes hem pa icula ly use ul in en i onmen al applica ions (Spe z e al. 2000). They show good sensi i i y o oxic and lammable subs ances (Zhao e al. 2000) and because o hei sensi i i y o e hanol hey a e also used in ood cooking, e men ing, and wine-making p ocesses (Zhao e al. 2000). Finally, hese senso s ha e been used o he e hylene measu emen du ing ui ipening p ocess (Winquis e al. 1990). 3.3. Conduc ing polyme senso s Since Diaz e al . (Pe saud and T a e s 1997) designed he i s polypy ole senso , he in e es in conduc ing polyme s (CPs) has inc eased yea by yea . These senso s a e made o a b oad ange o monome s ha a e polyme ised by chemical o elec ochemical me hods. A hin ilm o polyme is deposi ed on o a sensing subs a e o med by wo pa allel pla inum o gold pla ed elec odes wi h a insula ing base such as oxidized silicon (Pa idge e al. 1996). The mos common polyme s used in he syn hesis o CPs a e polypy oles, polyanilines, poly iophenes, bu ane o decane sulphonic acids, pa a - oluenesulphonic acid and i s e a-e hylammonium sal , sodium monohyd ogenosulpha e and e ae hylammonium e a luo obo a e (Mielle 1996) al hough new polyme s a e being syn hesized o open hese senso s o new applica ions. Thus, polyme s wi h a conjuga ed bonding sys em a e being now used due o he excellen elec ical cha ac e is ics o hese ma e ials o mic oelec onic de ice manu ac u ing (Rella e al. 2000). Senso s based on aniline and de i a es can be doped a speci ic a ios wi h o he compounds, as sulphonic acids (Koul e al. 2001) o HCl (Campos e al. 2000), so enhancing hei elec ical-op ical p ope ies and solubili y in o ganic sol en s. The dopan s p o ide ex a cha ges in he chain o he polyme wi h he o ma ion o cha ged de ec s, such as spinless bipola ons (Rella e al. 2000) o small egions o posi i e cha ge in he polyme chain ha p o ide mobile holes o elec on anspo . The CP sensing cha ac e is ics a e based in a measu able change in he elec ical conduc i i y when CPs a e exposed o ola ile compounds. A ixed ol age among he elec odes p oduces a cons an cu en ha passes h ough he conduc ing polyme . When gas lows on he senso , he ola ile compounds adhe e o i s su ace, and an adso p ion- deso p ion p ocess occu s on he polyme laye which al e s he elec on low in he sys em and hence he senso conduc i i y. The e o e hese 100 G asas y Acei es changes in conduc i i y a e based in he mobili y o cha ges along he conjuga ion o he π-backbone o he polyme . Coun e ions (s abilised ions ha a e pu inside he polyme by he manu ac u e in o de o balance o he cha ges inside he mic os uc u e o polyme laye ) can p oduce some modi ica ions such as changes in mo phology o polyme -dopan complex, in he concen a ion o chemical de ec s (pola ons o bipola ons) o he oxida ion s a e o he polyme (Tan and Blackwood 2000) al hough he ope a ing mechanism is no ye well unde s ood. Spec al s udies, as UV-VIS spec oscopy, FTIR spec oscopy, and elec on spin esonance spec oscopy (ESR), a e being ca ied ou o know mo e abou he in e ac ions occu ing du ing he sensing p ocess (Tan and Blackwood 2000; Koul e al. 2001). The senso cha ac e is ics explain why hey a e ones o he mos used in senso a ays. They a e easily ab ica ed, hei powe consump ion is low, and hey a e obus and can wo k a oom empe a u e. As i is said abo e, he e a e many ypes o polyme s, including a b oad ange o de i a es, and mo e ypes o polyme s appea each yea . The na u e o he dopan s also modi ies hei ini ial p ope ies (Koul e al. 2001), and i means he possibili y o designing senso s ailo ed o speci ic se ies o ola iles (e.g. ke ons, aldehydes o alcohols) o owa ds a pa icula applica ion (e.g. cha ac e isa ion o au hen ica ion). The oxida ion s a e o he polyme can also be changed a e deposi ion o ailo he sensing cha ac e is ics (Pa idge e al. 1996). CPs a e esilien o poisoning by he compounds ha inac i a e he desc ibed ino ganic semiconduc o senso s bu su p isingly hei li e ime, only abou 9-18 mon hs, is sho e . The andom na u e o he deposi ion p ocess is, howe e , a p oblem o he p oduc ion o ep oducible senso s, al hough he use o ou elec odes, ins ead o wo, seems o imp o e he esul s (Pa idge e al. 1996). The senso s ha e also o he d awbacks as hei oo long esponse ime (20-40 s), in compa ison wi h MOS (Mielle 1996), and hei inhe en d i o e ime o wi h changes in empe a u e. The main p oblem is, howe e , ha hey a e ex emely sensi i e o mois u e. Because CPs a e sensi i e o many ypes o ola iles, many applica ions a e desc ibed in he li e a u e. They a e sensi i e o alcohols, ace one, benzene and o he pola ola ile chemicals (Ha ield e al. 1994). Re e ences show applica ions o CPs in oods u s like bee and o he alcoholic be e ages (Pea ce e al. 1993; Sla e e al. 1993), oli e oil (Apa icio e al. 2000), o ange juices (Bazemo e e al. 1998), ishes (Luzu iaga e al. 1999), milk (Magan e al. 2001). Figu e 3 shows he esponses o polyme senso s (A omaScan A20S) o i gin oli e oils cha ac e ised by di e en le els o oxida ion. Conduc ing polyme s a e o en associa ed o o he ansduce mechanisms as sensing ma e ials. This is he case o he op ical senso s based on CPs, since hese polyme s show spec al changes in he UV- isible egion when he doping/dedoping p ocess happens (Kond a owicz e al. 2001). Op ical senso s inco po a ing CPs sensi i e o ammonia, humidi y, pH and ace ic acid ha e been ecen ly de eloped (Kond a owicz e al. 2001; de Ma cos e al. 2000). Senso s based on ield e ec ansis o (FET) inco po a ing a CP as he ga e ma e ial (Ha ield e al. 2000) o qua z-c ys al mic obalance associa ed o CP (Sla e e al. 1992) a e o he examples. Finally, CPs p o ide a su ace o immobilise biological elemen s like enzymes o an ibodies, in o de o p epa e biosenso s. Recen ly, a dielec ic moni o ing has been p oposed o s udy he adso p ion p ocess o biomolecules on o CPs, such as DNA on o polypy oles (Saoudi e al. 2000). 3.4. Acous ic senso s Elec oacous ic de ices based on he piezoelec ical p ope ies o qua z ma e ial ha e been success ully used o implemen senso s since 1964 (King 1964). These senso s a e based on he p opaga ion o acous ic wa es p oduced by piezoelec ical ma e ials (e.g. qua z o LiNbO3 o Si-SiO2-ZnO) in a mul ilaye s uc u e (Lucklum and Haup mann 2000). Su ace acous ic wa e (SAW) and bulk acous ic wa e (BAW) a e he mos common o hose acous ic senso s. The ope a ion mode o hese senso s is based on he physical changes p oduced on hei memb anes made o chemically in e ac i e ma e ials (CIM). The physical changes ha can be de ec ed a e ela ed o he piezoelec ic ma e ials used o he acous ic ansduc ion, o he mass densi y and ( isco-)elas ic p ope ies o CIM. The ope a ion p inciples o SAW and BAW senso s a e e y simila (D’Amico e al . 1997). Howe e , SAW- ype senso exploi s he p opaga ion Figu e 3 Mean o he aw esponse o conduc ing polyme senso s o a pu e i gin oli e oil and samples spiked wi h di e en pe cen ages o a i gin oli e oil ancid s anda d. Vol. 53. Fasc. 1 (2002) 101 o adio equency (102-103MHz) oscilla ions (e.g. Rayleigh wa es) along a laye consis ing a leas o a subs a e co e ed by he CIM. The acous ic wa es a e ansmi ed by he elec ical ield gene a ed by elec odes deposi ed a he su ace o he solid. SAW de ices p oduce bi-dimensional wa es a elling h ough he su ace o he c ys al. Depending on he geome y o he acous ic s uc u e and on he equency scaling, SAW senso s can p oduce di e en ypes o wa es: Rayleigh wa es, su ace ans e se wa es, Bleus ein-Gulyae wa es, and Lamb and Lo e wa es. The choice o one o hese wa es o he piezoelec ic ma e ial (ZnO, AIN, Bi12GeO20, α-SiO2, LiTaO3, LiNbO3) is in luenced by he cha ac e is ics o he memb ane and by he ope a ion in gaseous o liquid en i onmen s. F om he sensing poin o iew, he p esence o ola ile compounds p oduces he adso p ion o molecules on he senso and hence changes on he p ope ies o he CIM ha a ec bo h he phase eloci y and he p opaga ion loss o he acous ic wa e. The esul ing is a equency shi whose magni ude is ela ed o he amoun o ma e ial adso bed on o he laye senso . A BAW senso is a piezoelec ic esona o , o en made o qua z, wi h one o bo h su aces co e ed wi h a sensi i e coa ing (memb ane) o a ew µm (1µm-10nm) hickness (D’Amico e al . 1997), whose chemical and physical ( he mal) cha ac e is ics a e simila o he gas-ch oma og aphy s a iona y phases. The s uc u e is connec ed o an ampli ie o o m an oscilla o whose h ee-dimensional wa es a el h ough he c ys al (Schalle e al. 1998) a a equency o 10-30MHz. In he p esence o ola ile compounds he e is a p ocess o adso p ion o abso p ion o he chemical species ha al e he physical p ope ies o he memb ane and hence a ec s he esonan equency o he s uc u e. The changes a e de ec ed by measu ing he shi in he esonance equency; he magni ude o he shi is ela ed o he mass o he compound al hough he esponse can be no-linea due o iscoelas ic beha iou o he memb ane. Because o he esponse is p opo ional o he mass adso bed, his de ice (a qua z disk sandwiched be ween wo elec odes) is also called qua z c ys al mic obalance (QMB o QCM) (Fe a i e al. 2000) and i was al eady p oposed as a p omising senso o gas de ec ion in 1964 (King 1964). O he kinds o acous ic senso s a e he lexu al pla e wa e (FPW) senso s and can ile e -based senso s. FPW senso s a e simila o SAW senso s al hough hey a e coa ed wi h a memb ane whose hickness is much smalle han he acous ic wa eleng h (Cai e al. 2000) and, in consequence, he mechanical lexu e occu s in he en i e memb ane. Can ile e -based senso s, which de i e om he a omic o ce mic oscopy, consis o ee-s anding can ile e beams coa ed wi h a senso laye ha in e ac s wi h he analy e (Ba is on e al. 2001). The can ile e -based senso oscilla es a a ixed esonance equency, and any change o he mechanical p ope ies is de ec ed by measu ing he shi o he esonance equency (dynamic mode) o he s a ic de lec ion (s a ic mode). The mechanism o sensing o mic ocan ile e is simila o SAW and BAW senso s. The molecules a e abso bed in he ac i e laye , inc easing he mass o he can ile e (Po e e al. 2001). This change o mass loading esul s in a change o he ib a ion equency o ampli ude. Thus, he esponse o a can ile e is cha ac e ized by i s geome ical dimensions and he ib a ion equency among o he s (Ba is on e al. 2001). Acous ic senso s ha e a ious ad an ages o e o he senso s, as high sensi i i y and sho esponse ime, low powe consump ion and size, and obus ness (Gu ie ez e al. 2000). The sensi i i y o he acous ic senso is ela ed o he ope a ing equency (SAW: 100MHz-1GHz; BAW: 10-30 MHz), and i inc eases as he squa e o he undamen al equency does (Ballan ine and Wohl jen 1989). Fu he mo e, he equency shi is di ec ly p opo ional o he sensing laye hickness o a speci ic compound (Li and Ma 2000). Disad an ages o acous ic senso s a e he empe a u e and humidi y dependence, he di icul y o eplacing senso s, he poo ep oducibili y in he deposi ion o he coa ing ma e ial (Mielle 1996) and ce ain le el o noise (o e all in SAW) because o hei high equency (Schalle e al. 1998). In he case o QCM senso s, hei main disad an age is ha hey need o a hin c ys al o inc ease he sensi i i y ha p e en s hei inclusion in mul isenso a ays (Fe a i e al. 2000). The acous ic senso applica ions can be di ided be ween physical and chemical applica ions (Ballan ine and Wohl jen 1989). Some physical p ope ies can be measu ed by acous ic senso s, as empe a u e, p essu e, elec ic ield, displacemen and low (Cai and G imes 2000). The common cha ac e is ic o hese senso s is ha hey ha e magne oelas ic hick- ilm as coa ing ma e ial. The chemical applica ions equi e wo kinds o senso s, liquid-phase senso s and apou -phase senso s, acco ding o he medium in which hey ope a e (Ballan ine and Wohl jen 1989). The apou -phase acous ic senso s ha e been used o de ec gases and ola iles as NO2 (Mülle e al. 2000), NH3 (Ballan ine and Wohl jen 1989), xylene, oluene, e achlo oe hylene (Dicke e al. 1999), H2, oxic gases (McGill e al. 2000), o ganophospho ous compounds (Zimme mann e al. 2001) and SO2 (Qin e al. 2000). Acous ic senso s ha e also been used o analyse he headspace o oods u s as oli e oil samples (Di Na ale 2001). SAW senso s wi h a sensi i e ma e ial con aining a high deg ee o 102 G asas y Acei es hyd oxyl g oups is used o de ec he humidi y because i p oduces changes in he pola i y o he coa ing ma e ial (Dicke e al. 1999). Humidi y senso s ha e been manu ac u ed using polyme sensi i e o we ai , such as polyimid (Be nou e al. 2000). Al hough SAW senso s a e no sui able o be used in liquid-phase because o hei p opaga ion loss inside he liquid in e ace (Tong and Zhang 1999), some au ho s ha e used hem o measu e he low o liquid samples (Nomu a e al . 2001), o o de e mine he liquid densi y and iscosi y wi h a Lo e-mode acous ic senso (He mann e al . 1999). Finally, SAW senso s a e also used in scien i ic a eas as di e en as biochemis y, elec ochemis y (Ballan ine and Wohl jen 1989) and immunosensing (F eudenbe g e al. 2001). 3.5. Biosenso s A biosenso consis s o an immobilised biologic molecule (enzymes, cellules o an ibodies) nex o a ansduce , which ans o ms chemical signal in o an elec ic signal o in o o he kind o ou pu as op ical, acous ic and hea signal when an analy e eaches o i . (A oyo 1998). A nold and Meye ho (K ess- Roge s 1997) de ined biosenso s as “a sel - con ained analy ical de ice ha esponds selec i ely and e e sibly o he concen a ion o ac i i y o chemical species in biological samples”. Biosenso s hen consis o a biological sensing elemen and a ansduc ion elemen . The biological elemen (G i i hs and Hall 1993) can be made o ca aly ic (enzymes, mic oo ganisms and issues) o non-ca aly ic componen s (an ibodies, ecep o s and nucleic acids). The cha ac e is ics o biological elemen s allow clus e ing he biosenso s in h ee g ea g oups: me abolism biosenso s (enzymes, combina ions o enzymes and co ac o s, and cells like algae and bac e ia), a ini y biosenso s (an ibodies, “immunosenso s”) and ecombinan biosenso s (DNA p obes). The ansduc ion elemen allows he ans o ma ion o he chemical signal, ob ained om a biological p ocess, in o ano he kind o signal. Acco ding o he kind o he ansduced signal, biosenso s can also be g ouped in elec ochemical (ampe ome ic, po en iome ic, conduc ime ic), op ical, calo ime ic and acous ic biosenso s (G i i hs and Hall 1993). The biosenso sensing mechanisms depend on he ansduc ion echnology. In he case o elec ochemical biosenso s o example, he ansduc ion sys em is based on an elec ochemical na u e p ocess, and so he ansduced signal is elec ic. Mos o he comme cial biosenso s a e elec ochemical; ampe ome ic biosenso s being he mos widesp ead because he elec ochemical ins umen a ion is ela i ely simple and inexpensi e. The cu en lows a cons an po en ial wi h espec o a e e ence elec ode in he ampe ome ic senso s, and he cu en gene a ed by he oxida ion o educ ion o elec oac i e species a he su ace o he wo king elec ode is measu ed (G i i hs and Hall 1993). In ampe ome ic biosenso s, edox enzymes (oxido educ ases) ha e a special impo ance, since he ans o ma ion o he subs a e akes place wi h a ans e o elec on be ween edox enzymes and elec odes. These edox enzymes a e oxidases, pe oxidases and dehyd ogenases (Re iejo and Pinga ón 2000). The e a e ou main ca ego ies o ampe ome ic senso s (Palleschi and Cubadda 2001): Oxygen elec odes, hyd ogen pe oxide elec odes, ca bon-base media ed elec odes and NADH elec ochemical senso s. The po en iome ic biosenso s measu e he changes o po en ial a cons an cu en (usually ze o). Th ee kinds o ansduce a e used in he po en iome ic biosenso s: ion selec i e elec odes (ISEs), gas sensing elec odes, and ield e ec ansis o s (FETs) (G i i hs and Hall 1993). In he case o conduc ime ic biosenso s changes o conduc i i y be ween wo elec odes a e measu ed. This kind o ansduc ion echnology is no widely used due o low selec i i y. Nowadays, lo s o s udies ha e been done o ab ica e elec on- ans e ing in e aces be ween edox enzymes and elec odes in o de o pe o m an enhanced elec ochemical biosenso (Shinoha a e al. 2000). These media o s a e able o eplace oxygen as an elec on accep o and o ope a e a a po en ial ha g ea ly educes he e ec s o o he elec ochemically ac i e species (Whi e and Tu ne 1997). Some o hese media o s a e e ocene and i s de i a i es, cobal ph alocyanine, e a hia ul alene, elec opolyme s (Alba eda-Si en 2000), and p o eins (Shinoha a e al. 2000). Op ical biosenso s a e based on op ical changes while he biochemical eac ion akes place. I a ligh beam is p opaga ed h ough a wa eguide, hen he in e nally e lec ed ligh gene a es an elec omagne ic “e anescen wa e” ha can be used o exci e luo escen molecules a he su ace o he wa eguide. The “e anescen wa e” can couple wi h he elec on plasma o a me al when he wa eguide is coa ed wi h a me al hin laye . I causes he elec ons o oscilla e and hus gene a ing a su ace plasmon wa e (G i i hs and Hall 1993) ha is cha ac e ised by high sensi i i y and speci ici y. The su ace plasmon is an elec omagne ic wa e associa ed wi h he longi udinal oscilla ion o he ee elec on gas on he in e ace o he me al and he dielec ic (Caide and Sui 2000). Recen ly, su ace plasmon esonance in e e ome y (SPRI) has been es ablished as a no el me hod (Niki in e al. 2000) ha gains in sensi i i y and esolu ion o he de ec ion o a mul iplici y o analy es. Finally, he e a e calo ime ic and acous ic biosenso s. The calo ime ic biosenso s a e based Vol. 53. Fasc. 1 (2002) 103 on he ac ha almos all biochemical eac ions a e exo he mic. Thei comme cial success is e y limi ed o he da e, he mopiles and enzyme he mis o s being he mos used. The acous ic biosenso s measu e changes in he acous ic p ope ies o he senso s. The e a e wo ypes: su ace acous ic wa e o bulk wa e (G i i hs and Hall 1993). The p incipal ad an age o enzyme biosenso s is ela ed wi h hei high speci ici y (s e eoselec i i y and egiospeci ici y) and ca aly ic ac i i y. Howe e , hey ha e he disad an ages o low s abili y and possibili y o being eusable (A oyo 1998). These incon eniences can be pa ially a oided by he immobilisa ion o he enzymes, which consis s o ixing enzymes on o a su ace by physical o chemical me hods in o de o ob ain an insoluble by-p oduc wi h ca aly ic ac i i y. The immobilisa ion p o ides ad an ageous e ec s o e enzymes such as an ac i i y inc emen , high eusabili y and easibili y o handling and con ol. Bu he enzyme biosenso s also ha e inhe en disad an ages such as he e ogenei y o he sys em, high cos , and changes in he con o ma ion o he enzymes (A oyo 1998). The con en ional me hods o he enzyme immobilisa ion include co alen binding, encapsula ion in a polyme o c oss-linking o a sui able suppo ing ma ix (Lillis e al. 2000). Table III shows di e en me hods o immobilise he enzymes. The applica ions o biosenso s sp ead o e a lo o ields like clinical, ag icul u e, mili a y, indus ial and ood a eas al hough hei comme cial se up is being ca ied ou slowly. In medicine, biosenso s a e used in he assessmen o d ug o biomolecule concen a ions in blood, se um o o he co po al luid in- i o o in- i o measu emen s, and da a a e used in diagnos ic o on-line moni o ing (Liu and Li 2000). The ag icul u al applica ions o biosenso s a e ocused on he de ec ion o i al, bac e ia and ungal in ec ions, he use o e ilize s and pes icides (G i i hs and Hall 1993), and ola iles eleased by damaged plan s in glasshouses (Schü z e al. 2000). In mili a y de ence, biosenso s could be used o de ec ne e gases and o he p oduc s used in chemical wa a e (G i i hs and Hall 1993). Biosenso s a e applied o he con ol p ocesses in pha maceu ical, chemical and pe oleum indus ies, mining and cons uc ion (G i i hs and Hall 1993). The en i onmen al con ol o oxic compounds and pollu an gases (Köhle e al. 2000) and was ewa e oxici y (Liao e al. 2001) can also be done by biosenso s. Finally, biosenso s ha e been also used in Food Technology o moni o he quali y con ol o aw ma e ials and on-line o o -line manu ac u ing p ocesses. Thus, glucose and suc ose biosenso s ha e been used o con ol he p ocess o alcohol e men a ion (Fe ei a e al. 2001), and o de ec en e o oxins (Tempelman e al. 1996) and bac e ia (Su e al. 2001) in many oods u s. Recen ly, a biosenso based on ca alase enzyme has been de eloped o hyd ope oxide de e mina ion in ex a i gin oli e oil samples (Campanella e al. 2001). 4. THE ELECTRONIC NOSE Recen ly, senso s wi h sensi i i y o a wide ange o ola iles compounds ha e been in eg a ed in a sys em called elec onic nose. The name o ‘elec onic nose’ is due o he simila i ies be ween his ins umen and he physiological sys em. Ga dne and Ba le (1994) de ined elec onic nose as an ins umen , which comp ises an a ay o elec onic chemical senso s wi h pa ial speci ici y and an app op ia e pa e n- ecogni ion sys em, capable o ecognising simple o complex odou s. Using his de ini ion i can be hough ha elec onic noses a e composed o pa s such as a ays o non-speci ic solid s a e gas senso , a a ie y o ansduce s, da a collec o s and da a analysis ools, all o ien ed o he classi ica ion and quan i ica ion o chemical clus e s o ola ile compounds, including, in pa icula , odou s (D’Amico e al. 2000). Ne e heless, he e m “elec onic nose” is being inco ec ly used o Table III Me hods o immobilising he enzymes o biosenso s Me hod o enzyme immobilisa ion B ie desc ip ion Re e ences En apmen Mic oencapsula ion Physical adso p ion Co alen binding C oss-linking Immobilisa ion in ma ices like gels, polyme s o pas es. The enzymes a e enclosed wi hin semipe meable polyme memb anes. an de Waals a ac ion o ces be ween an enzyme and a solid suppo su ace. Enzymes and ca ie s a e bound by co alen bond. Fo ma ion o h ee-dimensional linkings be ween he biological ma e ial and bi- o mul i unc ional eagen s. O ega e al. 1998 Ramos e al. 2001 Lü h e al. 2000 Senillou e al. 1999 Dzyade ych e al. 2001 104 G asas y Acei es Chaud y, A.N., Hawkins, T.M., T a e s, P.J. (2000). A me hod o selec ing an op imum senso a ay. Sens. Ac ua o s B , 69, 236-242. Comini, E, Faglia, G., Sbe eglie i, G., Li, Y.X., Wloda ski, W., Ghan asala, M.K. (2000). Sensi i i y enhancemen owa ds e hanol and me hanol o TiO2 ilms doped wi h P and Nb, Sens. Ac ua o s B , 64, 169-174. C a en, M.A., Ga dne , J.W., Ba le , P.N. (1996). Elec onic noses-de elopmen and u u e p ospec s. T ends in Anal. Chem. , 15, 9, 486-493. D’Amico, A., Di Na ale, C., Paolesse, R. (2000). Po ai s o gases and liquids by a ays o non-speci ic chemical senso s: ends and pe spec i es. Sens. Ac ua o s B , 68, 324-330. D’Amico, A., Di Na ale, C., Ve ona, E. (1997). Acous ic de ices. In: Handbook o Biosenso s and Elec onic Noses. Medicine, Food and En i onmen , p. 197-223. E. K ess-Roge s (Ed). CRC P ess: Boca Ra on, FL. De Ma cos, S., Asensio, C., U uñuela, I., Galla a, F., Galbán, J., Cas illo, J.R. (2000). New app oaches o polyaniline op ical senso s: pH, ace ic acid and amonia de e mina ion. Química Analí ica , 19, 1, 99-104. Di Na ale, C., Macagnano, A., Na dis, S., Paolesse, R., Falconi, C., P oie i, E., Siciliano, P., Rella, R., Tau ino, A., D’Amico, A. (2001). Compa ison and in eg a ion o a ays o qua z esona o s and me al-oxide semiconduc o chemo esis o s in he quali y e alua ion o oli e oil. Sens. Ac ua o s B , 78, 303-309. Di Na ale, C., Paolesse, R., Macagnano, A., Man ini, A., D’Amico, A., Ubigli, M., Legin, A., L o a, L., Rudni skaya, A., Vlaso , Y. (2000). Applica ion o a combined a i icial ol ac ion and as e sys em o he quan i ica ion o ele an compounds in ed wine. Sens. Ac ua o s B , 69, 342-347. Dicke , F.L., Hayden, O., Zenkel, M.E. (1999). De ec ion o ola ile compounds wi h mass-sensi i e senso a ays in he p esence o a iable ambien humidi y. Anal. Chem. , 71, 1338-1341. Dzyade ych, S., A khypo a, V.N., Ko pan, Y.I., El’skaya, A.V., Solda kin, A.P., Ja ezic-Renaul , N., Ma ele , C. (2001). Conduc ome y o maldehyde sensi i e biosenso wi h speci ically adap ed analy ical cha ac e is ics. Anal. Chim. Ac a , 445, 47-55. Enso , D.R. (1989). The con ibu ion o la ou chemis y o he ood indus y In: Fla ou Chemis y o Lipid Foods, p. 1-12. D.B. Min and T.H. Smouse (Eds.). The Ame ican Oil Chemis s’ Socie y, Champaign. Fe a i, V., Ma ioli, D., Ta oni, A., Ranucci, E. (2000). Mul isenso a ay o mass mic obalances o chemical de ec ion based on esonan piezo-laye s o sc een-p in ed PZT. Sens. Ac ua o s B , 68, 81-87. Fe ei a, L.S., De Souza, M.B., Folly, R.O.M. (2001). De elopmen o an alcohol e men a ion con ol sys em based on biosenso measu emen s in e p e ed by neu al ne wo ks. Sens. Ac ua o s B , 75, 166-171. F eudenbe g, J., on Schick us, M., Hunklinge , S. (2001). A SAW immunosenso o ope a ion in liquid using a SiO2 p o ec i e laye . Sens. Ac ua o s B , 76, 147-151. Fukui, K., Ka suki, A. (2000). Imp o emen o humidi y dependence in gas senso based on SnO2. Sens. Ac ua o s B , 65, 316-318. Ga cía Pin o, C., Fe nández Laespada, M.E., Pé ez Pa ón, J.L., Mo eno Co de o, B. (2001). Elec onic ol ac ome y. A new ool in analy ical chemis y. Química Analí ica , 20, 3-11. Ga dne , J.W., Ba le , P.N. (1994). A b ie his o y o elec onic noses. Sens. Ac ua o s B , 18-19, 211-220. Ga dne , J.W., C a en, M., Dow, C. Hines, E.L. (1998). The p edic ion o bac e ia ype and cul u e g ow h phase by an elec onic nose wi h a mul i-laye pe cep on ne wo k. Meas. Sci. Technol ., 9, 120-127. Ga dne , J.W., Hines, E.L., Wilkinson, M. (1990). Applica ion o a i icial neu al ne wo ks o an elec onic ol ac o y sys em. Meas. Sci. Technol ., 1, 446-451. Ga dne , J.W., Shu me , H.V., Tan, T.T. (1992). Applica ion o an elec onic nose o he disc imina ion o co ees. Sens. Ac ua o s B , 6, 71-75. González Ma ín, Y., Pé ez Pa ón, J.L., Mo eno Co de o, B., Ga cía Pin o, C. (1999). Classi ica ion o ege able oil by linea disc iminan analysis o Elec onic Nose da a. Anal. Chim. Ac a , 384, 83-94. Göpel, W. (2000). F om elec onic o bioelec onic ol ac ion, o : om a i icial “moses” o eal noses. Sens. Ac ua o s B , 65, 70-72. G i i hs, D., Hall, G. (1993). Biosenso s-Wha eal p og ess is being made? T ends Bio echnol. , 11, 122-130. Guada ama, A., Fe nández, J.A., Íñiguez, M., Sou o, J., de Saja, J.A. (2001). Disc imina ion o wine a oma using an a ay o conduc ing polyme senso s in conjunc ion wi h solid-phase mic o-ex ac ion (SPME) echnique. Sens. Ac ua o s B , 77, 401-408. Gu ié ez, F.J., Fon echa, J.L., Ho illo, M.C., Sayago, I., Fe nández, M.J., Gómez-Espinosa, R., Cané, C., G ácia, I. (2000). Mic osenso es de gases basados en ondas acús icas supe iciales. P oceeding Cong eso TEDEA, Almag o (Ciudad Real, Spain) 26-9-2000. Gu ié ez-Osuna, R., Nagle, H.T. (1999). A me hod o e alua ing da a-p ep ocessing echniques o odo classi ica ion wi h an a ay o gas senso s. IEEE T ans. Sys . Man Cybe ., 29, 626-632. Ha ield, J.V., Co ing on, J.A., Ga dne , J.W. (2000). GasFETs inco po a ing conduc ing polyme s as ga e ma e ials. Sens. Ac ua o s B , 65, 253-256. Ha ield, J.V., Nea es, P., Hicks, P.J., Pe saud, K., T a e s, P. (1994). Towa ds an in eg a ed elec onic nose using conduc ing polyme senso s. Sens. Ac ua o s B , 18-19, 221-228. He mle, T, Weima , U., Mi o ics, J., Rosens iel, W., Göpel, W. (1999). Fea u e ex ac ion o a hyb id gas sensing sys em In: Elec onic Noses & Senso A ay Based Sys em , Design and Applica ions, p. 97-102. W.J. Hu s (Ed.), Technomic Co, Inc., Lancas e , PA. He mann, F., Hahn, D., Bü genbach, S. (1999). Sepa a e de e mina ion o liquid densi y and iscosi y wi h sagi ally co uga ed Lo e-mode senso s. Sens. Ac ua o s A , 78, 99-107. Hi sch elde , M., Han iede , D., Ul ich, D. (1998). Disc imina ion o s awbe y a ie ies by a gas senso a ay in co ela ion wi h human senso y e alua ion. Semin. Food Anal. , 3, 27-36. Hodgins, D. (1997). The elec onic nose: Senso a ay-based ins umen s ha emula e he human nose In: Techniques o Analyzing Food A oma , p. 331-371. R. Ma sili (Ed.), Ma cel Dekke Inc., New Yo k. Hong, S.K. (1999). Compensa ion o nonlinea he mal bias d i o Resonan Ra e Senso using uzzy logic. Sens. Ac ua o s A , 78, 143-148. Iiyama, S., Yahi o, M., Toko, K. (2000). Measu emen s o soy sauce using as e senso . Sens. Ac ua o s B , 66, 205-206. I a sson, P, Holmin, S., Höje , N.E., K an z-Rülcke , C., Winquis , F. (2001). Disc imina ion o ea by means o a ol amme ic elec onic ongue and di e en applied wa e o ms. Sens. Ac ua o s B , 76, 449-454. Vol. 53. Fasc. 1 (2002) 111 Ka o, K., Ka o, Y., Takama su, K., Udaka, T., Nakaha a, T., Ma suu a, Y., Yoshikawa, K. (2000). Towa d he ealiza ion o an in elligen gas sensing sys em u ilizing a non-linea dynamic esponse. Sens. Ac ua o s B , 71, 192-196. Kelle , P.E. (1999). Physiologically inspi ed pa e n ecogni ion o elec onic noses. P oceedings o he SPIE, 3722, No. 13. King, W.H. JR. (1964). Piezoelec ic so p ion de ec o . Anal. Chem. , 36, 1735-1739. Kohl, D., Heine , L., Bock, J., Ho mann, Th., Schiebe le, P. (2000). Sys ema ic s udies on esponses o me al-oxide senso su aces o s aigh chain alkanes, alcohols, aldehydes, ke ones, acids and es e s using he SOMMSA app oach. Sens. Ac ua o s B , 70, 43-50. Köhle , S., Bachmann, T.T., Schmi , J., Belkin, S., Schmid, R.D. (2000). De ec ion o 4-chlo obenzoa e using immobilized ecombinan Esche ichia coli epo e s ains. Sens. Ac ua o s B , 70, 139-144. Kond a owicz, B., Na ayanaswamy, R, Pe saud, K.C. (2001). An in es iga ion in o he use o elec och omic polyme s in op ical ib e gas senso s. Sens. Ac ua o s B , 74, 138-144. Koul, S., Chand a, R., Dhawan, S.K. (2001). Conduc ing polyaniline composi e: A eusable senso ma e ial o aqueous ammonia. Sens. Ac ua o s B , 75, 151-159. K ess-Roge s, E. (1997). Biosenso s and elec onic noses o p ac ical applica ion In: Handbook o Biosenso s and Elec onic Noses. Medicine, Food and En i onmen , p. 3-39. E. K ess-Roge s (ed). CRC P ess: Boca Ra on, FL. Lang, A.C., Fleische , M., Meixne , H. (2000). Su ace modi ica ions o Ga2O3 hin ilm senso s wi h Rh, Ru and I clus e s. Sens. Ac ua o s B , 66, 80-84. Lazze ini, B., Maggio e, A. Linguis ic uzzy classi ica ion o ol ac i e signals. 1998 IEEE In e na ional Con e ence on uzzy sys ems p oceedings. IEEE Wo ld Cong ess on Compu a ional In elligence (CAT No. 98CH36228), p. 1518-1523. Vol. 2. Legin, A., Rudni skaya, A., Vlaso , Y., Di Na ale, C., Mazzone, E., D’Amico, A. (2000). Applica ion o elec onic ongue o quali a i e and quan i a i e analysis o complex liquid media. Sens. Ac ua o s B , 65, 232-234. Li, D., Ma, M. (2000). Su ace acous ic wa e mic osenso s based on cyclodex in coa ings. Sens. Ac ua o B , 69, 75-84. Liao, J.D., Wang, S-H., Hsu, D-J. (2001). S udies on he ea ly de ec ion o was ewa e ’s oxici y using a mic obial sensing sys em. Sens. Ac ua o s B , 72, 167-173. Lillis, B., G ogan, C., Be ney, H., Lane, W.A. (2000). In es iga ion in o immobilisa ion o lac a e oxidase o imp o e s abili y. Sens. Ac ua o s B , 68, 109-114. Liu, J., Li, G. (2000). Applica ions o biosenso s o diagnos ic analysis and biop ocess moni o ing. Sens. Ac ua o s B , 65, 26-31. Liu, W., Ma, W., Li, L., Li, J. (2000). The empe a u e change in SnO2- based gas senso s du ing sensing eac ion. Sens. Ac ua o s B , 66, 222-224. Lucas, Q., Poling, J., Benincasa, V. (1998). Analysis wi h a hyb id sys em QCM/CP/MOS. Semin. Food Anal. , 3, 53-58. Lucklum, R., Haup mann, P. (2000). The qua z c ys al mic obalance: mass sensi i i y, iscoelas ici y and acous ic ampli ica ion. Sens. Ac ua o s B , 70, 30-36. Lunds öm, I., Shi a aman, M.S., S ensson, C., Lundk is , L (1975). Applied Physics Le e s , 26, 55. Lunds öm, I., Spe z, A., Winquis , F., Ackelid, U., Sundg en, H. (1990). Ca aly ic me als and ield-e ec de ices – a use ul combina ion. Sens. Ac ua o s B , 1, 15-20. Lü h, H., Thus , M., S e en, A., Ko dos, P., Schöning, M.J. (2000). Biochemical senso s wi h s uc u ed and po ous silicon capaci o s. Sens. Ac ua o s B , 69, 104-108. Luzu iaga, D.A., Balaban, M.O. (1999). Elec onic nose odo e alua ion o salmon ille s s o ed a di e en empe a u es. In: Elec onic Noses & Senso A ay Based Sys em, Design and Applica ions. W. J. Hu s (Ed.), Technomic:Lancas e , PA, pp. 219-279. Magan, N., P lou, A., Ch ysan hakis, I. (2001). Milk-Sense: A ola ile sensing sys em ecognises spoilage bac e ia and yeas in milk. Sens. Ac ua o s B , 72, 28-34. Masila, M., B eime , M., Sadik, O. A. (1999). S a egies o imp o ing he analysis o ola ile o ganic compounds using GC-based elec onic nose In: Elec onic Noses & Senso A ay Based Sys em, Design and Applica ions , p. 26-42. W.J. Hu s (Ed.), Technomic Co, Inc., Lancas e , PA. McGill, R.A., Nguyen, V.K., Chung, R., Sha e , R.E., DiLella, D., S epnowski, J.L., Mlsna, T.E., Venezky, D.L., Dominguez, D. (2000). The ‘‘NRL-SAWRHINO’’: a nose o oxic gases. Sens. Ac ua o s B , 65, 10-13. Ménil, F., Coilla d, V., Luca , C. (2000). C i ical e iew o ni ogen monoxide senso s o exhaus gases o lean bu n engines. Sens. Ac ua o s B , 67, 1-23. Middelhoek, S. (2000). Celeb a ion o he en h ansduce s con e ence: The pas , p esen and u u e o ansduce esea ch and de elopmen . Sens. Ac ua o s A , 82, 2-23. Mielle, P. (1996). Elec onic noses: Towa ds he objec i e ins umen al cha ac e iza ion o ood a oma. T ends Food Sci. Technol. , 7, 432-438. Mielle, P., Ma quis, F., La asse, C. (2000). Elec onic noses: speci y o disappea . Sens. Ac ua o s B , 69, 287-294. Mon edo o, G. F., Se ili, M., Baldioli, M., Minia i, E. (1992). Simple and hyd olyzable phenolic compounds in oli e oil: 1. Thei ex ac ion, sepa a ion, and quan i a i e and semiquan i a i e e alua ion by HPLC. J. Ag ic. Food Chem . 40, 1571-1576. Mo ales, M.T., Apa icio, R., Gu ié ez, F. (1992). Técnicas de aislamien o y concen ación de olá iles de acei es ege ales. G asas Acei es, 43, 164-173. Mo ales, M.T., Tsimidou, M. (2000). The ole o ola ile compounds and polyphenols in oli e oil senso y quali y. In: Handbook o Oli e Oil: Analysis and P ope ies, p. 393-458. J. Ha wood and R. Apa icio (Eds.), Aspen:Gai he sbu g, MA. Mo imi su, M., Ozaki, Y., Suzuki, S., Ma sunaga, M. (2000). E ec s o su ace modi ica ion wi h pla inum and u henium on empe a u e and humidi y dependence o SnO2-based CO gas senso s. Sens. Ac ua o s B , 67, 184-188. Muenchmeye , W., Wal e, A., Ma z, G. (2000). Imp o ing elec onic noses using a ap and he mal deso p ion uni . Sens. Ac ua o s B , 69, 379-383. Mülle , C., Ni maie , T., Rügeme , A., Schick us, M. . (2000). Sensi i e NO2 de ec ion wi h su ace acous ic wa e de ices using a cyclic measu ing echnique. Sens. Ac ua o s B , 68, 69-73. Nakamo o, T., Nakahi a, Y., Hi ama su, H., Mo iizumi, T. (2001). Odo eco de using ac i e odo sensing sys em. Sens. Ac ua o s B , 76, 465-469. Niki in, P.I., G igo enko, A.N., Beloglazo , A.A., Valeiko, M.V., Sa chuk, A.I., Sa chuck, O.A., S eine , G., Kuhne, C., Huebne , A., Salze , R. (2000). Su ace 112 G asas y Acei es plasmon esonance in e e ome y o mic o-a ay biosensing. Sens. Ac ua o s A , 85, 189-193. Nomu a, T., Sai oh, A., Ho ikoshi, Y. (2001). Measu emen o acous ic p ope ies o liquid using liquid low SH-SAW senso sys em. Sens. Ac ua o s B , 76, 69-73. O ega, N., Bus o, M.D., Pe ez-Ma eos, M. (1998) Op imisa ion o -glucosidase en apmen in algina e and polyac ilamide gels. Bio esou . Technol. , 64, 105-111. Palleschi, G., Cubadda, R. (2001). Elec ochemical biosenso s o ood analysis and he ood indus y. I alian J. Food Science , 13, 137-157. Pa do, M., Niede jau ne , G., Benussi, G., Comini, E., Faglia, G., Sbe eglie i, G., Holmbe g, M., Lunds om, I. (2000b). Da a p ep ocessing enhances he classi ica ion o di e en b ands o Esp esso co ee wi h an elec onic nose. Sens. Ac ua o s B , 69, 397-403. Pa do, M., Sbe eglie i, G., Ga dini, S., Dalcanale, E. (2000a). A hie a chical classi ica ion scheme o an Elec onic Nose. Sens. Ac ua o s B , 69, 359-365. Pa idge, A.C., Ha is, P., And ews, M.K. (1996). High sensi i i y conduc ing polyme s senso s. Analys , 121, 1349-1353. Paulsson, N, La sson, E., Winquis , F. (2000). Ex ac ion and selec ion o pa ame e s o e alua ion o b ea h alcohol measu emen wi h an elec onic nose. Sens. Ac ua o s A , 84, 187-197. Pea ce, T.C., Ga dne , J.W., F iel, S., Ba le , P.N., Blai , N. (1993). Elec onic nose o moni o ing he la ou o bee s. Analys , 118, 371-377. Pe saud, K.C., T a e s, P.J. A ays o b oad speci ici y ilms o sensing ola ile chemicals In: Handbook o Biosenso s and Elec onic Noses. Medicine, Food And En i onmen , p. 563-592. E. K ess-Roge s (Ed.) CRC P ess, Inc. Boca Ra on, FL. Po e , T.L., Eas man, M.P., Pace, D.L., B adley, M. (2001). Senso based on piezo esis i e mic ocan ile e echnology. Sens. Ac ua o s A , 88, 47-51. Qin, S., Wu, Z., Tang, Z., Song, Y., Zeng, F., Zhao, D. (2000). The sensi i i y o SO2 o he SAW gas senso wi h ie hanolamine modi ied wi h bo ic acid. Sens. Ac ua o s B , 66, 240-242. Ramos, M.C., To ijas, M.C., Na as Díaz, A. (2001) Enhanced chemiluminiscence biosenso o he de e mina ion o phenolic compounds and hyd ogen pe oxide. Sens. Ac ua o s B , 73, 71-75. Reineccius, G. (1993). Biases in analy ical la o p o iles in oduced by isola ion me hod In: Fla o Measu emen , p. 61-67. C.T. Ho and C.H. Manley (Eds.), Ma cel Dekke , New Yo k. Rella, R., Siciliano, P., Qua an a, F., P imo, T., Valli, L., Schene i, L., Mucci, A., Ia ossi, D. (2000). Gas sensing measu emen s and analysis o he op ical p ope ies o poly[3-(Bu yl hio) Thiophene] Langmui -Blodge ilms. Sens. Ac ua o s B , 68, 203-209. Re iejo, A.J., Pinga ón, J.M. (2000). Biosenso es elec oquímicos. Una He amien a ú il pa a el análisis medioambien al, alimen a io y clínico. Anales de la Real Sociedad Española de Química , 96, 2, 5-15. Romain, A-C., Nicolas, J., Wie z, V., Ma e no a, J., And é, Ph. (2000). Use o a simple in oxide senso a ay o iden i y i e malodou s collec ed in he ield. Sens. Ac ua o s B , 62, 73-79. Rong, L., Ping, W., Wenlei, H. (2000). A no el me hod o wine analysis based on senso usion echnique. Sens. Ac ua o s B , 66, 246-250. Saoudi, B., Despas, C., Chehimi, M.M., Jammul, N., Delama , M., Bessié e, J., Walca ius, A. (2000). S udy o DNA adso p ion on polypy ole: In e es o dielec ic moni o ing. Sens. Ac ua o s B , 62, 35-42. Sbe eglie i, G., Comini, E., Faglia, G., Niede jau ne , G., Benussi, G., Con a ini, G., Po olo, M. (1998). A no el elec onic nose based on semiconduc o hin ilm gas senso o dis inguish di e en hea ea men s o milk. Semin. Food Anal. , 3, 67-76. Schaak, R.E., Dahlbe g, D.B., Mille , K.B. (1999). The elec onic nose: s udies on he undamen al esponse and disc imina i e powe o me al-oxide senso s In: Elec onic Noses & Senso A ay Based Sys em , Design and Applica ions , p. 43-50. W. J. Hu s (Ed.), Technomic Co, Inc., Lancas e , PA. Schalle , E., Bosse , J. O., Esche , F. (1998). Elec onic nose and hei applica ion o ood. Lebensm.-Wiss. u.-Technol. , 31, 305-316. Schü z, S, Schöning, M.J., Sch o h, P., Malkoc, Ü, Weibecke , B., Ko dos, P., Lü h, H., Hummel, H.E. (2000). An insec -based BioFET as a bioelec onic nose. Sens. Ac ua o s B , 65, 291-295. Senillou, A., Ja ezic-Renaul , N., Ma ele , C., Cosnie , S. (1999). A minia u ized u ea senso based on he in eg a ion o bo h ammonium based u ea enzyme ield e ec ansis o in a single chip. Talan a , 50, 219-226. Shinoha a, H., Kusaka, T., Yoko a, E., Monden, R., Sisido, M. (2000). Elec on ans e be ween edox enzymes and elec odes h ough he a i icial edox p o eins and i s applica ion o biosenso s. Sens. Ac ua o s B , 65, 144-146. Shu me , H.V., Ga dne , J.W. (1992). Odou disc imina ion wi h an elec onic nose. Sens. Ac ua o s B , 8, 1-11. Simon, I, Bâ san, N., Baue , M., Weima , U. (2001). Mic omachined me al oxide gas senso s: oppo uni ies o imp o e senso pe o mance. Sens. Ac ua o s B , 73, 1-26. Sinesio, F., Di Na ale, C., Quaglia, G.B., Buca elli, F.M., Mone a, E., Macagnano, A., Paolesse, R., D’Amico, A. (2000). Use o elec onic nose and ained senso y panel in he e alua ion o oma o quali y. J. Sci. Food Ag ic., 80, 63-71. Singh, S., Hines, E.L., Ga dne , J.W. (1996). Fuzzy neu al compu ing o co ee and ain ed wa e da a om an elec onic nose. Sens. Ac ua o s B , 30, 3, 190-195. Sla e , J.M., Payn e , J., Wa , E.J. (1993). Mul i-laye conduc ing polyme gas senso a ays o ol ac o y sensing. Analys , 118, 379-384. Sla e , J.M., Wa , E.J., F eeman, N.J., May, I.P., Wei , D.J. (1992). Gas and apou de ec ion wi h poly(Py ole) gas senso s. Analys , 117, 1265-1270. Spe z, A., Winquis , F., Sundg en, H., Lunds öm, I. (1992). Field e ec gas senso s In: Gas senso s, p. 219-279. G. Sbe eglie i (Ed.), Kluwe Academic:Do d ech , Ge many. Spe z, A.L., Tobias, P., Unéus, L., S ennings o p, H., Ekedahl, L-G, Lunds öm, I. (2000). High empe a u e ca aly ic me al ield e ec ansis o s o indus ial applica ions. Sens. Ac ua o s B , 70, 67-76. STATISTICA. Use s Guide, elease 5.5; S a so : Tulsa, OK, 2000. S e e , J.R., S a hmann, S., McEn ega , C., Decas o, M., Pen ose, W.R. (2000). New senso a ays and sampling sys ems o a modula elec onic nose. Sens. Ac ua o s B , 69, 410-419. Su, X., Low, S., Kwang, J., Chew, V.H.T., Li, S.F.Y. (2001). Piezoelec ic qua z c ys al based e e ina y diagnosis o Salmonella en e i idis in ec ion in chicken and egg. Sens. Ac ua o s B , 75, 1-2, 29-35. Takada, T. (2000). Tempe a u e d op o semiconduc o gas senso when exposed o educing gases-simul aneous Vol. 53. Fasc. 1 (2002) 113 measu emen o changes in senso empe a u e and in esis ance. Sens. Ac ua o s B , 66, 1-3. Tan, C.K., Blackwood, D.J. (2000). In e ac ions be ween polyaniline and me hanol apou . Sens. Ac ua o s B , 71, 184-191. Tempelman, L.A., King, K.D., Ande son, G.P., Ligle , F.S. (1996). Quan i a ing S aphylococcal en e o oxin B in di e se media using a po able ibe -op ic biosenso . Anal. Biochem., 233, 50-57. Toko, K. (2000). Tas e senso . Sens. Ac ua o s B , 64, 205-215. Tong, X., Zhang, D. (1999). No el p opaga ion di ec ion o quasi-longi udinal leaky su ace acous ic wa e on qua z and i s po en ial as liquid senso s. Sens. Ac ua o s A , 78, 160-162. We lein, H-D. (2001). Disc imina ion o chocola es and packaging ma e ials by an elec onic nose. Eu . Food Res. Technol ., 212, 529-533. Whi e, S.F., Tu ne , A.P.F. (1997). Enzymes, co ac o s, and media o s In: Handbook o Biosenso s and Elec onic Noses. Medicine, Food and En i onmen , p. 43-57. E. K ess-Roge s (Ed). CRC P ess: Boca Ra on, FL. Wilson, D.M., Dunman, K., Roppel, T., Kalim, R. (2000). Rank ex ac ion in in-oxide senso a ays. Sens. Ac ua o s B , 62, 199-210. Winquis , F., Lunds öm, I., Be gk is , H. (1990). E hylene p oduc ion om ui s measu ed by a simple ield-e ec s uc u e and compa ed wi h a gas ch oma og aphic me hod. Anal. Chim. Ac a , 231, 93-100. Winquis , F., Lunds öm, I., Wide, P. (1999). The combina ion o an elec onic ongue and an elec onic nose. Sens. Ac ua o s B , 58, 512-517. Xu, Y., Zhou, X., So ensen, T. (2000). Oxygen senso s based on semiconduc ing me al oxides: an o e iew. Sens. Ac ua o s B , 65, 2-4. Zhang, G., Liu, M. (2000) E ec o pa icle size and dopan on p ope ies o SnO2-based gas senso s. Sens. Ac ua o s B , 69, 144-152. Zhao, S., Sin, J.K. O., Xu, B., Zhao, M., Peng, Z., Cai, H. (2000). A high pe o mance e hanol senso based on ield-e ecc ansis o using a LaFeO3 nano-c ys alline hin-Film as a ga e elec ode. Sens. Ac ua o s B , 64, 83-87. Zimme mann, C., Rebiè e, D., Déjous, C., Pis é, J., Chas aing, E., Planade, R. (2001). A lo e-wa e gas senso coa ed wi h unc ionalized polysiloxane o sensing o ganophospho us compounds. Sens. Ac ua o s B , 76, 86-94. Zonde an, C., Mu esan, S., de Jonge, H.G., an Velzen, E.U.T., Wilkinson, C., Nijhuis, H.H., Leguij , T. (1999). Con olling Mailla d eac ions in he hea ing p ocess o blockmilk using an elec onic nose. J. Ag ic. Food Chem ., 47, 4746-4749. 114 G asas y Acei es