Microscale titration of acetic acid using digital colorimetry and paper-based analytical devices

Talan a 276 (2024) 126254 A ailable online 12 May 2024 0039-9140/© 2024 The Au ho s. Published by Else ie B.V. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by- nc-nd/4.0/). Mic oscale i a ion o ace ic acid using digi al colo ime y and pape -based analy ical de ices I a i Be asa e a , b , * , Ane Bo daga ay a , Rosa Ga cia-A ona a , Mi en Os a a , William Reis de A aujo b , Maide Vidal a a Depa men o Applied Chemis y, Uni e si y o he Basque Coun y (UPV/EHU), 20018, Donos ia/San Sebas ian, Spain b Po able Chemical Senso s Lab, Depa men o Analy ical Chemis y, Ins i u e o Chemis y, S a e Uni e si y o Campinas (UNICAMP), 13083-970, Campinas, S˜ ao Paulo, B azil ARTICLE INFO Keywo ds: Pape -based de ice Mic o- i a ion Pape mic o luidics Colo ime y Digi al image analysis P ocess au oma ion ABSTRACT A quan i a i e me hod o acid-base i a ions in pape -based de ices (PADs) is desc ibed o analyze ace ic acid in inega samples. In his wo k, wo di e en ypes o PADs we e de eloped: a de ice o indi idual spo es ing and a mic o luidic de ice. Digi al colo ime y was used as he de ec ion me hod, and he images we e acqui ed using a sma phone and a homemade box wi h LED ligh s o con olled image acquisi ion. Ti a ion cu es we e buil wi h jus eigh poin s, using he R channel based on he g adual colo ansi ion om ed o blue o li mus, a na u al indica o . The endpoin was accu a ely de e mined by second de i a i e calcula ions. Bo h sys ems we e applied o i een inega samples o di e en ypes, and good concen a ion esul s we e ob ained in compa ison o he e e ence me hod. The p oposed me hodology is simple, as , en i onmen ally iendly, and su passes he need o calib a ion cu e cons uc ion. Mo eo e , he subjec i e endpoin iden i ica ion is elimina ed, and he me hod was au oma ed o p o ide a high h oughpu wo k low, sui able o quali y con ol p ocesses and eal- ime measu emen s. 1. In oduc ion One o he mos popula adi ional me hodologies ha enables quan i a i e analysis is olume ic analysis. As a esul , he concen a- ion o a pa icula analy e can be measu ed using a solu ion wi h a known concen a ion (usually used as i an ). The e a e many di e en i a ion ypes, and hey can be ca ego ized based on he measu ing echnique, he o de o he i an s, o he ype o esponse. The esponse can be based on an acid-base, edox, complexome ic, o p ecipi a ion eac ion [1]. Focusing on acid-base i a ions, he mos common me hods employed a e manual i a ion wi h isual indica ion o he endpoin (he ea e EP, se o co espond o equi alence poin ). Ins umen al and au oma ed lab-based me hods using spec opho o- me ic de ec o s and he use o pH elec odes (po en iome y) p o ide accu a e measu es, exemp ing he analys om in e p e ing he colo change. Howe e , o e he las decades, signi ican echnological ad- ances ha e been made, enabling he implemen a ion o digi al colo i- me ic analysis o ollow hese i a ions and o de ec he endpoin a he poin -o -need [2,3]. Digi al colo ime y cons i u es a powe ul ool due o i s simplici y, low cos , and he di e si y o imaging op ions, anging om desk op scanne s and webcams o digi al came as and sma phones [4]. An image can be decomposed in o di e en colo models, such as RGB, HSV o CIELAB, and hen, colo in ensi y can be ela ed o analy e concen- a ion [5]. Many examples can be ound whe e digi al colo ime y has been applied. Fo ins ance, o he sc eening o disease ma ke s in se um [6], he de e mina ion o sul onamides in wa e [7], nickel in elec oless coa ing ba hs [8], and o al phenolic con en in ea and in usion samples [9]. Rega ding he i a ion-based de e mina ion o o al acidi y, i has been calcula ed using a mic opla e, ei he in wines wi h a la bed scanne [10], o in ci ic ui s employing a sma phone [11], among o he s. In bo h cases, a ec o is c ea ed wi h RGB alues ex ac ed om he image, which is plo ed agains he added i an olume o ob ain a i a ion cu e. Then, EP is calcula ed wi h i s o second de i a i es. In ac , de i a i es a e he employed ma hema ical calcula ions o de e - mine he in lec ion poin o sinusoidal cu es as he ones ob ained in i a ions. Sodium chlo ide has also been de e mined by i a ion and employing a mic opla e – sma phone sys em [12]. In his case, G * Co esponding au ho . Depa men o Applied Chemis y, Uni e si y o he Basque Coun y (UPV/EHU), 20018, Donos ia/San Sebas ian, Spain. E-mail add ess: [email p o ec ed] (I. Be asa e). Con en s lis s a ailable a ScienceDi ec Talan a jou nal homepage: www.else ie .com/loca e/ alan a h ps://doi.o g/10.1016/j. alan a.2024.126254 Recei ed 21 Feb ua y 2024; Recei ed in e ised o m 6 May 2024; Accep ed 11 May 2024 Talan a 276 (2024) 126254 2 channel alues we e plo ed agains added i an olume and EP was calcula ed as he in e sec ion be ween wo s aigh lines. Pape -based analy ical de ices (PADs) ha e also d awn a en ion as a p omising ool in analy ical chemis y in o de o de ec and quan i y he analy e o in e es . Pape is made o cellulose, which gi es he abili y o anspo liquids wi hou ins umen a ion, by capilla i y [13]. The e a e many ways o manu ac u e hese de ices, including pho oli hog aphy, plasma ea men , wax-p in ing, ink-p in ing and lase ea men [14]. The de ec ion can also be done by se e al me hods, such as luo escence, Raman sca e ing, elec ochemis y, o colo ime y. Because o i s lex- ibili y and malleabili y, many di e en designs can be c ea ed, including spo es s, mic o luidic analysis (known as μ PADs), and dis ance-based measu emen s [15]. In gene al, PADs a e used in many ields, including en i onmen al moni o ing, poin -o -ca e diagnos ics, o en- sics, clinical analysis, and ood sa e y [16–19]. In addi ion, li e a u e shows ha hey can be combined wi h i ime ic analyses: chela e i a ion o Ca 2+ and Mg 2+ de e mina ion [20], edox i a ion o alcohol con en analysis [21], acid-base i a ion o o maldehyde quan i ica ion [22] and iodome ic i a ion o asco bic acid and dopamine de e mina ion [23], o example. Tep ek e al. p esen ed ou di e en sys ems including alkalini y de e mina ion by acid-base i a- ion, ha dness quan i ica ion by complexome ic i a ion, edox i a- ion o hiosul a e quan i ica ion and chlo ide analysis by p ecipi a ion i a ion [24]. In hese e e ences, EP was iden i ied isually on he PADs based on he colo change o he indica o . Scaling om semi-quan i a i e o quan i a i e analysis can be achie ed wi h digi al colo ime y, as i can be employed o build a calib a ion cu e. Fo example, by he use o he R channel in ensi y [25] o Euclidean no m [26], which can be ela ed o concen a ion. Ace ic acid, also known as e hanoic acid o me hylca boxylic acid, is a weak o ganic acid ha can be used in he p oduc ion o la ex emulsion esins, syn hesis o e eph halic acid, ui inega p oduc ion, dyes o pha maceu ical o mula ions, among o he s [27]. I is also he p ima y o ganic componen o inega , esponsible o i s scen and bi e la o . Vinega is de ined as a liquid p oduc ha is p oduced by alcoholic and ace ic e men a ion o ag icul u al-o igin subs ances. I is mos ly known o i s applica ions in he ood indus y as salad seasoning, pickling, la o ing, o as a p ese a i e agen in bo led sauces [28]. Table 1 shows some wo ks ela ed o ace ic acid de e mina ion ound in he li e a u e. Rega ding i a ion-based echniques, a pH ISFET cell o wine analysis has been epo ed [29], as well as isual de ec ion o endpoin o inega samples [30,31], and digi al colo ime y o ace ic acid de e mina ion in wa e [32]. In ela ion o spec oscopic me hods, UV–Vis [33] and NIR [34] ha e been used o inega analysis, and luo escence-based de e mina ions o bac e ia [35] and Chinese wine [36]. Finally, HPLC o inega , p obio ic and wa e ke i be e ages [37–40], and 1 H-RMN o mul i a ia e analysis o s awbe ies [41] ha e been epo ed. In he majo i y o he cases, analy e quan i ica ion was ca ied ou using an analy ical cu e, ei he wi h uni a ia e cali- b a ion using po en ial di e ence, abso bance o peak a ea, o example; o by mul i a ia e analysis me hods. Li e a u e shows ha ace ic acid can also be de e mined using PADs. Acid-base i a ions ha e been ca ied ou by di e en me hods such as spo es s [43] o μ PADs, ei he by s a -like designs [44] o dis ance-based measu emen s [45]. Howe e , he majo i y o he sys- ems a e based on quali a i e o semi-quan i a i e EP de e mina ion, some wo ks equi e building a calib a ion line p e iously, which do no employ he i a ion cu e i sel o acid de e mina ion. In o he cases, images need o be aken a e each i an addi ion, which make he me hods labo ious and signi ican ly inc eases expe imen al wo k and da a analysis complexi y. In his wo k, mic oscale acid-base i a ions o ace ic acid de e - mina ion ha e been ca ied ou in wo ypes o pape -based de ices, a spo es ing de ice and a s a -like design ollowing a mic o luidic app oach. In he p oposed sys ems, jus an eigh -poin i a ion is necessa y o ob ain a quan i a i e and accu a e de e mina ion o EP based on second de i a i e calcula ions, which is achie ed using a single image. To ou knowledge, i is he i s μ PAD sys em whe e EP iden i- ica ion is comple ely objec i e and based on he i a ion cu e, wi hou he need o isual de ec ion o p e iously building a calib a ion cu e. A simple, low-cos , and en i onmen ally iendly me hod is p oposed, au oma ed o as sample analyses. 2. Expe imen al 2.1. Chemicals and ma e ials All eagen s we e o analy ical g ade and used wi hou addi ional pu i ica ion. Li mus, ace ic acid, po assium hyd ogen ph hala e and absolu e e hanol we e pu chased om Pan eac (Ba celona, Spain). So- dium hyd oxide was supplied by Sigma-Ald ich (S . Louis, USA) and Labkem (Ba celona, Spain). Phenolph halein was p o ided by Me ck (Da ms ad , Ge many). Deionized wa e ( esis i i y ≥18 MΩ cm a 25 ◦C) was ob ained om a Milli-Q Ad an age-0.10 pu i ica ion sys em (Millipo e, Ge many). Solu ions o li mus 0.1%, li mus 0.166% and phenolph halein 0.1% we e p epa ed by dissol ing he app op ia e weigh ed mass o he in- dica o s in a small amoun o absolu e e hanol and dilu ing o ma k wi h deionized wa e . S anda d solu ion o ace ic acid 4% was p epa ed om 96% glacial ace ic acid and NaOH 0.1 mol L −1 was s anda dized wi h 0.056 mol L −1 po assium hyd ogen ph hala e. 2.2. Fab ica ion o pape -based de ices Pape -based de ices we e designed using Mic oso O ice Powe - Poin 2016 so wa e and we e p in ed on quali a i e il e pape using a wax p in e (Xe ox Colo Qube 8570, CT, USA). The p in ed shee s we e placed in a he mal p ess (Maquina ec, MG, B azil) o 15 s a 100 ◦C. Upon hea ing, wax pene a es he pape and de ines hyd ophobic ba - ie s a ound he pa e n. The back o he p in ed de ice was co e ed wi h anspa en ape o p e en solu ions om leaking h ough he de ice. Two app oaches we e used in his wo k, spo es analysis and mic o luidic analysis, so wo PADs we e designed and ab ica ed (Fig. 1). Spo es s we e pe o med in 5 mm diame e ci cles, pa e ned in a 3 by 8 design in o de o analyze eigh i a ion poin s in iplica e (Fig. 1a). Mic o luidic de ices we e ab ica ed modi ying he s a -like design epo ed by Ka i a and Kane a [44], wi h a sample ese oi loca ed a he cen e , eigh b anches wi h a eac ion ese oi and a de ec ion zone a he end o each b anch (Fig. 1b). Table 1 Summa y o analy ical me hods used o ace ic acid de e mina ion ound in he li e a u e. Technique Sample De e mina ion Re . Ti a ion Wine pH ISFET cell and analy ical cu e [29] Vinega Visual endpoin de ec ion [30,31] pH-me e and analy ical cu e [42] Wa e Digi al colo ime y and i a ion cu e de i a i es [32] Spec oscopy Vinega UV–Vis and analy ical cu e [33] NIR and mul i a ia e eg ession [34] Bac e ia Fluo escence biosenso and analy ical cu e [35] Chinese wine Fluo escen u anyl-o ganic amewo k and analy ical cu e [36] HPLC Vinega , p obio ic and wa e ke i be e ages UV–Vis and analy ical cu e [37–40] 1 H-RMN S awbe y Mul i a ia e analysis (PCA, HCA and PLS-DA) [41] I. Be asa e e al. Talan a 276 (2024) 126254 3 2.3. Colo ime ic acid-base i a ion In bo h app oaches, eigh -poin acid-base i a ions we e ca ied ou . In o de o ind he op imal measu emen condi ions, some pa ame e s we e op imized. These pa ame e s include de ice size, indica o , i an and sample olumes, d ying me hod and image acquisi ion ime. Rega ding spo es s, i s , mix u es o li mus, NaOH and wa e we e added o he spo s, and hen, sample was added. Immedia ely, PADs we e le o d y and images we e aken a di e en imes o ind he op imal condi ions. Fo mic o luidic analyses, inal concen a ions in spo es s we e aken as e e ence. Fi s , a ying NaOH concen a ion solu ions we e added o all eac ion ese oi s. Then, li mus was added o all de ec ion zones. Finally, sample was placed in he cen e o he μ PAD. De ices we e le o d y and images we e aken a di e en imes o achie e he op imal condi ion. In bo h cases, digi al images o PADs we e acqui ed wi h a sma - phone OnePlus No d CE ( h ee came as, 64 +2 +8 MP) in au oma ic mode. In o de o main ain image acquisi ion condi ions cons an , a home-made ligh box was used. The po able wooden box (dimensions: 10 cm heigh , 12 cm wid h, 15 cm leng h) was ab ica ed o elimina e ex e nal ligh in e e ences, wi h 24 V whi e LED ligh s along he in e- io . Bo h PADs and he sma phone we e always placed in ixed posi- ions o a oid u he a ia ion be ween pho og aphs. RGB alues we e ex ac ed using he ee image-edi ing so wa e GIMP ( e sion 2.10.34). Image analysis was also ca ied ou using MATLAB R2020b e sion (The Ma hwo ks Inc., Massachuse s, USA). 2.4. Vinega samples Bo h o he p oposed me hods we e es ed in i een inega samples. Fi e samples we e pu chased in local ma ke s om Campinas (S˜ ao Paulo, B azil), including lemon, ice and alcohol inega (Cas elo Ali- men os S.A.) and wo di e en apple inega samples (Cas elo Alimen os S.A. and Alma omi, bo h om S˜ ao Paulo, B azil). The o he 10 samples we e pa o an indus ial p ocess, p o ided by JR Saba e S.A. (Mu cia, Spain), and included i e samples de i ed om alcoholic e men a ion and ano he i e samples o whi e wine e men a ion. Comme cial samples we e dilu ed 10 imes, as indica ed concen a ion on he label is 4%, whe eas indus ial samples, wi h a heo e ical concen a ion o a ound 10%, we e dilu ed 25 imes. Acid-base po en iome ic i a ion was used as he e e ence me hod. Ace ic acid was de e mined using he 794 Basic Ti ino (Me ohm, Swi ze land) au oma ic i a o coupled o a pH elec ode. Fi s ly, NaOH was s anda dized wi h po assium hyd ogen ph hala e and phenolph halein in iplica e. Then, 0.25 mL o 0.5 mL o inega sample (depending on i s heo e ical concen a ion) and 2.0 mL o li mus 0.1% indica o we e added o a beake . Finally, solu ions we e i a ed wi h s anda dized NaOH solu ion un il he colo changed om ed o blue. EP olume was used o he calcula ion o ace ic acid concen a ion. Mea- su emen s we e pe o med in iplica e. 3. Resul s and discussion 3.1. Op imiza ion o PAD-based sensing 3.1.1. Colo channel selec ion In his wo k, li mus was used as acid-base indica o (pK a 6.5), which colo change occu s be ween pH 4.5–8.3 om ed o blue, wi h pu ple colo a neu al alues. Ace ic acid is a weak acid o which i a ion EP lies be ween pH 6.0–9.0. Fo ha eason, li mus was chosen o e some common indica o s such as phenolph halein, as i allows us o obse e a b oade colo change close o he EP, p o iding be e p ecision and sensi i i y o his weak acid-s ong base i a ion. A syn he ic solu ion o ace ic acid 0.4%, a e a 1:10 dilu ion o he 4% s ock solu ion, was used o p ocedu e op imiza ion, and i s ly, some p elimina y es s we e pe o med in 96-well mic opla es o selec he op imal colo channel. Digi al images we e analyzed o ex ac ed (R), g een (G) and blue (B) alues o each spo , which we e plo ed agains he i an olume added o achie e a classic acid-base i a ion cu e (Fig. S1). The h ee channels had mo e o less cons an alues be o e EP. Then, R and G channels d as ically dec ease, while B channel sligh ly inc eases. A e EP, R alues become cons an , while G and B alues inc ease a bi . Fo his eason, R channel was selec ed as he colo channel o moni o ing he i a ions, as i gi es he mos sigmoidal cu e. 3.1.2. Op imiza ion o spo es s p ocedu e Fab ica ion o PADs was done by wax p in ing and he mal hea ing, and no ela ed p oblem was iden i ied du ing he expe imen s. Fi s and o emos , he op imal indica o olume o he designed and wax- p in ed de ice was in es iga ed. Two indica o olumes we e es ed, 5 μ L and 7 μ L. I was easily obse ed by naked eye ha 7 μ L was oo much, as d ying ime inc eased signi ican ly and some imes he d op exceeded he wax, so 5 μ L was chosen as op imal indica o olume. Then, 96-well mic opla es we e used o es ima e i an olumes, which we e scaled o PAD p opo ions. Once NaOH olumes we e ixed as 1.0–6.0 μ L, i a ions we e ca ied ou using h ee di e en concen- a ions o li mus. Colo alues wi h li mus 0.05% we e oo ligh , and he Fig. 1. Design o pape -based de ices. (a) Wax-p in ed pape de ice wi h indi idual spo s o an eigh -poin i a ion, wi h h ee ows o analysis in iplica e. (b) Wax-p in ed μ PAD o mic o luidic analysis, wi h eigh iden ical b anches loca ed equidis an ly om he cen e (sampling zone). Leng hs a e indica ed in millime e s. I. Be asa e e al. Talan a 276 (2024) 126254 4 calcula ion o he concen a ion was mo e subjec ed o e o . In con as , a high concen a ion o li mus as 0.2% comp omised he i a ion, as EP was shi ed o highe alues, and hus, he calcula ion o acidi y was a ec ed. Wi h a concen a ion o 0.1%, bo h i a ion cu es and ace ic acid concen a ion showed good esul s, so i was selec ed as he op imal concen a ion. Impo an di e ences we e obse ed be ween adding he eagen s indi idually and adding hem oge he . When eagen s we e added indi idually, i a ion cu es we e no homogeneous and he me hod became un eliable. When bo h eagen s we e added oge he , epea - abili y was imp o ed, and o ha eason, li mus and NaOH we e added oge he . The mix u e is s able o a long ime, i can be p e iously p epa ed, and he co ec mix u e o he eagen s is ensu ed. Mix u es o 5 μ L li mus, 1–6 μ L NaOH, and 1–5 μ L wa e we e p epa ed in ials, and 11 μ L was added o each o he eigh i a ion spo s. Then, 4 μ L o a s anda d o ace ic acid 0.4% was added o s a he eac ion. Thus, inal concen a ions o each compound we e: NaOH 0.007–0.04 mol L −1 , li mus 0.03%, and sample 0.1%. Two o he mos impo an ac o s o conside a e he d ying me hod and he d ying/ eac ion ime. On he one hand, de ices we e d ied a oom empe a u e and in he o en (Fig. 2a). PADs le a oom em- pe a u e ook a lo o ime o d y, up o 90 min o mo e. This was caused by mode a e o high ai humidi y, which a ies g ea ly be ween days and seasons, impeding he s anda diza ion o he p o ocol. In con as , i he de ices we e d ied in an o en, PADs we e comple ely d y in 30 min maximum, so i was selec ed as d ying me hod. A empe a u e o 50 ◦C was used o a oid he de o ma ion o he wax-based ba ie . On he o he hand, ega ding d ying ime (Fig. 2b), i could be isually de ec ed ha spo s changed colo in he i s minu es a e addi ion, so images acqui ed immedia ely and a e 5 min did no gi e good esul s. A e 10 min, good EP and concen a ion esul s we e ob ained. Howe e , because he d ops we e no comple ely d y, pixel selec ion o da a ex ac ion was limi ed, and hus, he shape o he cu es was no comple ely sigmoidal. A 20 min, some spo s we e d y and o he s we e no , so i a ion cu es showed a iable R alues wi h high e o ba s, impac ing EP de ini ion and p o iding inaccu a e e- sul s. A e 30 min in he o en, all spo s we e comple ely d y and image quali y was good, which esul ed in good i a ion cu es and concen- a ion esul s in all cases, so his ime was se as op imal. Finally, i needs o be no ed ha he educ ion o o al eagen s and sample olume o each spo om 15 μ L o 10 μ L o 8 μ L, o example, could be in es iga ed. I will depend on he sys em unde s udy and he necessa y concen a ion o each eagen , bu ha way, he o al d ying ime would be educed, accele a ing he p ocess. 3.1.3. Op imiza ion o mic o luidic analysis p ocedu e Mic o luidic pape -based de ices ( μ PADs) we e ab ica ed as men ioned in he Expe imen al pa . Fi s ly, NaOH concen a ion and olume we e op imized. Mix u es wi h di e en concen a ions o i an we e p epa ed in small ials, conside ing he inal i an mmol added in spo es s. Reac ion ese oi s could be illed wi h 1 μ L in he o iginal design [44], bu as ou design consis ed o 8 b anches ins ead o 10, op imiza ion was ca ied ou wi h 0.50, 0.75, and 1.00 μ L o i an . Wi h he lowes olume, inaccu a e esul s we e ob ained due o he incomple e illing o he whole eac ion ese oi . In con as , he highes olume sp ead h ough he channels o he de ice, and in e - e ed wi h concen a ion calcula ions. I was de e mined ha 0.75 μ L (NaOH s anda ds be ween 0.02 and 0.12 mol L −1 ) was he op imal olume o i an o be added o he eac ion ese oi s, as i a oided bo h p oblems men ioned and o e ed good concen a ion esul s. Rega ding li mus olume and concen a ion, i was isually de ec - able ha 0.5 μ L was enough o ill he de ec ion zone co ec ly. The concen a ion was ecalcula ed acco ding o he li mus-NaOH ela ion used in spo es s con igu a ion, so a new s ock solu ion o li mus 0.166% was p epa ed and used. In his μ PAD sys em, sample was placed in he cen e o sp ead h ough he channels. P elimina y es s we e ca ied ou using 28, 24, 20 and 16 μ L o sample. I was obse ed ha he wo smalles olumes we e no enough o comple e he i a ion. By naked eye, colo s ob ained wi h 28 μ L and 24 μ L we e simila . Howe e , R channel alues and i a ion cu es showed ha he mos accu a e concen a ion alues we e ob- ained using 24 μ L o he sample. Once he op imal sample olume was se , eac ion ime was in es- iga ed. Images we e acqui ed using he ligh ing box a 0, 2.5, 5 and 10 min. I was obse ed ha 10 min was he necessa y ime o comple ely d y he de ice. Room empe a u e was selec ed as he d ying me hod due o he size o he μ PAD and he small olumes used in he analysis. 3.1.4. Summa y o colo ime ic p ocedu es Rega ding spo es s, each o he i a ion poin s was analyzed in iplica e wi h he same de ice. Fi s ly, 11 μ L o a eagen mix u e was added o all spo s. The mix u e con ained 5.00 μ L o li mus 0.1%, a ying olumes o NaOH 0.10 mol L −1 (1.00–6.00 μ L), and wa e up o 11.0 μ L. A e , 4.00 μ L o sample 0.4% was added. Immedia ely, PADs we e placed in he o en a 50 ◦C, and images we e aken a e 30 min. Fo μ PAD analysis, 0.75 μ L o a ying NaOH concen a ion solu ions (0.02–0.12 mol L −1 ) we e added o all eac ion ese oi s, and hen, 0.50 μ L o li mus 0.166% was added o all de ec ion zones. Nex , 24 μ L o sample 0.4% we e placed in he cen e o he μ PAD (sampling zone). The de ice was d ied a oom empe a u e and images we e acqui ed a e Fig. 2. (a) Ti a ion cu es ob ained wi h wo di e en d ying me hods. D ying a oom empe a u e ook up o 90 min, in con as wi h he o en, which ook only 30 min. (b) Ti a ion cu es ob ained om PADs d ied in he o en o di e en imes. I. Be asa e e al. Talan a 276 (2024) 126254 5 10 min. 3.1.5. Me hod alida ion Once op imal analysis pa ame e s we e iden i ied, a s anda d solu- ion o ace ic acid (0.4%) was used o me hod alida ion. In he case o spo es s, in a-day p ecision was calcula ed using h ee PADs measu ed on he same day (n =9), and in e -day p ecision wi h ou de ices om h ee di e en days (n =12). In he case o μ PADs, i een de ices di ided in o 5 images (3 de ices pe image) we e used o in a-day p ecision calcula ion (n =15), and in e -day p ecision was e alua ed using nine μ PADs om 3 images aken in h ee di e en days (n =9). Tes s pe o med wi h he s anda d solu ion showed good p ecision and accu acy esul s by bo h he p oposed me hods (Table 2), which come in ag eemen wi h he guideline o AOAC [46]. 3.2. Colo ime ic i a ion and ace ic acid calcula ion As p e iously s a ed, images o spo s PADs we e ob ained a e 30 min in he o en, while images o μ PADs we e cap u ed a e 10 min a oom empe a u e. In bo h cases, R channel alues we e used o build i a ion cu es, ob aining sigmoidal shapes. A e ex ac ing R channel alues, he i s and second de i a i es we e calcula ed and plo ed, as i is he co ec way o ma hema ically calcula ing he in lec ion poin o his ype o cu e. In his i a ion, he i s de i a i e esul s in a min- imum alue a he endpoin , whe eas i equals ze o on he second de i a i e. Assuming ha he eac ion be ween he weak acid and he s ong base is comple e, and due o he 1:1 s oichiome y o he eac ion, i can be assu ed ha he NaOH mmol numbe equals ace ic acid mmol exac ly a EP. Using he R colo alues and he i an olumes, expe imen al de i a i es we e calcula ed. Then, a eg ession line was buil be ween he maximum and minimum poin s o he second de i a i e cu e. Subsequen ly, EP olume o mmol numbe was calcula ed o y = 0 using he slope and in e cep o he eg ession line. Finally, ace ic acid concen a ion (%) was calcula ed. In he case o spo es s, sample olume is a known alue (4 μ L), so R alues we e plo ed agains he olume o NaOH added, and ace ic acid concen a ion was calcula ed di ec ly using ha alue. In ela ion o mic o luidic analyses, 0.75 μ L o di e en i an solu ions we e added o he eac ion ese oi s, and hen, he sample was added o he cen e o he de ice, so i could sp ead o he channels o he μ PAD. To build i a ion cu es, R channel alues we e plo ed agains he added NaOH mmol numbe . In his case, added NaOH olume was he limi ing ac o , so he amoun o sample ha ook pa in he i a ion is he amoun o i an ha was ixed (0.75 μ L). A schema ic illus a ion o he p ocedu e o EP calcula ion is shown in Fig. 3. 3.3. Vinega sample analysis Volume ic analysis pe o med wi h an au oma ic i a o and po en iome ic de ec o was he e e ence me hod used o inega sample analysis. In addi ion, he p oposed pape -based me hods, bo h spo es s and μ PADs, we e employed. All indus ial and comme cial samples we e analyzed in iplica e in all cases. Some inega ypes may con ain small amoun s o a a ic and ci ic acid. Howe e , due o he dilu ion applied o he samples (1:10 o 1:25), no in e e ence should be expec ed om hose compounds. Fig. 4 shows he i a ion cu es ob ained wi h bo h me hods o a sample (C101) selec ed as example. Rega ding spo es s, ace ic acid concen a ions we e 4.4%–4.9% o comme cial inega s and 10.1%–17% o indus ial samples. Mic o- luidic measu emen s ga e concen a ions be ween 4.2% and 4.8% o comme cial samples and 10.2%–17% o indus ial samples. P ecision alues (RSD%) anged be ween 0.6% and 13% in spo es ing, and be- ween 0.9% and 9.9% o μ PAD analysis. I needs o be no ed ha , e en i some RSD alues we e a bi high, hey we e below 10% o he g ea majo i y o samples. Conside ing he di e en measu ing loca ions and ha some samples we e no measu ed as soon as hey we e opened o p oduced, especially in he case o he indus ial samples, accep able esul s we e ob ained wi h he p oposed PADs. Besides, ela i e e o s ob ained by spo es s we e all below 10.5%, wi h nine samples below 5%. E o s ob ained by mic o luidic mea- su emen s we e simila , wi h 14% being he highes ela i e e o and wi h en samples below 5%. In conclusion, simila esul s we e ob ained wi h bo h me hods in compa ison o he e e ence me hod. De ailed alues a e summa ized in Table 3. Resul s ob ained by po en iome ic i a ion and he p oposed digi al colo ime y i a ion on PADs we e s a is ically e alua ed. Two di e en es s we e conduc ed. Fi s , since sample concen a ions a y in a wide ange, a join con idence ellipse es (EJCR) o slope and in e cep was pe o med [47,48]. In he case o spo es s, he con idence in e al o he slope was 1.00 ±0.09, and 0.2 ±0.8 o he in e cep . Fo he μ PAD app oach, he slope was 1.00 ±0.08, and in e cep was −0.1 ±0.8 (Fig. S3). Conside ing hese alues, he ideal poin (slope 1 and in e cep 0) lies wi hin hese anges in bo h PADs, so he p oposed me hods do no di e signi ican ly om he e e ence me hod. Then, a pai ed - es was pe o med wi h a 95% con idence le el. No signi ican di e ences we e ound be ween e e ence and spo es s ( exp 1.80 < c i 2.14), no be ween e e ence and mic o luidic measu emen s ( exp 1.20 < c i 2.14), since he calcula ed alue was below he abu- la ed one. In conclusion, i can be assu ed ha none o he p oposed me hods p esen s sys ema ic e o s and ha hey a e eliable me hods o ace ic acid de e mina ion in inega samples. Collec i ely, he p oposed mic oscale i a ion on PADs combined wi h digi al colo ime y p esen s signi ican imp o emen s om he G een Analy ical Chemis y poin o iew. I educes he amoun o e- agen s consumed and was e gene a ed due o i s minia u iza ion, and p o ides au oma ed analyses compa ed o classical i a ion, p o iding an AGREE sco e [49] o 0.71 o spo es ing, 0.78 o he μ PAD app oach and 0.49 o he classical i a ion me hod (Fig. S4). These alues demons a e ha mic oscale i a ions pe o med on PADs a e mo e sus ainable han he adi ional po en iome y me hod. In addi- ion, he use o digi al colo ime y o i a ion imp o es he accu acy o he me hod because i exemp s he analys om isually iden i ying he colo endpoin , which may be subjec i e. 3.4. Me hod au oma ion Me hod au oma ion in he ood indus y, speci ically o analy e moni o ing, is c ucial o ensu ing accu acy and e iciency in he es ing p ocess. Au oma ion educes human e o s, enhances p ecision, and allows o a highe h oughpu o samples. In his ega d, wo image analysis so wa e we e used and compa ed in his wo k. On he one hand, he ee so wa e GIMP was employed. A Mic oso Excel (Mic oso , 2016 e sion) empla e was c ea ed o simpli y and speed up da a analysis, and jus by inse ing he R channel alues ex ac ed manually by GIMP, he empla e makes he necessa y i s and second de i a i e calcula ions, co esponding i a ion g aphs, and ace ic acid concen a ion calcula ions. E en i analysis ime was educed, he colo ex ac ion wo k becomes edious when dealing wi h a la ge numbe o measu emen s. As an al e na i e, a use - iendly au oma ed unc ion was designed in MATLAB using he Image P ocessing Toolbox™. This way, manual ex ac ion is a oided, a single so wa e is used, and a s ack o images Table 2 Me hod alida ion esul s o he wo ypes o PADs de eloped in his wo k. PAD ype Acidi y (%) P ecision (RSD, %) Accu acy (RE, %) In a-day In e -day Spo s 0.427 ±0.007 0.4–1.9 4.8 2.7–9.1 μ PAD 0.39 ±0.02 1.0–6.4 5.4 1.4–3.8 I. Be asa e e al. Talan a 276 (2024) 126254 6 aken in he same posi ion can be analyzed in a single un, p o iding a high h oughpu wo k low. A schema ic illus a ion o he p ocedu e is shown in Fig. S2. A ew inpu s need o be speci ied o un he unc ion: image name o o ma , dilu ion ac o o he sample, and in he case o spo es s, NaOH concen a ion. Then, all necessa y s eps om dimen- sion selec ion o ace ic acid calcula ion a e pe o med au oma ically. Finally, a double g aph wi h all he esul s becomes isible ( i a ion cu e, 2 nd de i a i e g aph, EP, and ace ic acid concen a ion in he sample), as well as an ou pu ma ix in he wo kspace wi h he con- cen a ion esul s. The esul s ob ained wi h bo h me hods we e almos iden ical, indica ing he use ulness and e ec i eness o he moni o ing o ace ic acid concen a ion ei he in academia o indus ial p ocesses. Mo eo e , he p oposed PADs can be employed o pe o m o he ypes o i a ions, as he unc ion is a ailable unde eques and i is easily adjus able o o he eac ions jus by modi ying i an olumes o concen a ions. 3.5. Compa ison wi h o he wo ks As was men ioned in he in oduc ion, de e mina ion o ace ic acid using PADs has been p e iously epo ed in he li e a u e. Table 4 shows he wo ks whe e spo es ing and μ PADs ha e been used o ha pu pose. In ela ion o spo measu emen s, he wo ks ound we e majo ly ela ed o eaching and adap ing he pape -based me hod o expe i- men al labo a o y cou ses. The mos used i an was NaOH, and he Fig. 3. Schema ic illus a ion o EP calcula ion p ocedu e. Mean R alues we e ex ac ed and plo ed agains added NaOH ( μ L o mmol, he la e in his example igu e). Second de i a i es we e calcula ed o he cu es. A eg ession line was cons uc ed be ween he p e ious and he ollowing poin o he cu e c ossing a y =0, and om i , EP ( μ L o mmol) was calcula ed. Fig. 4. Ti a ion cu es ob ained o C101 sample selec ed as example. (a) Spo es ing, iplica e analysis in he same de ice, and (b) mic o luidic measu emen s pe o med h ee imes. I. Be asa e e al. Talan a 276 (2024) 126254 7 indica o s a ied om he classic phenolph halein o o he s ha o e g adual colo changes such as jabo icaba peel ex ac o li mus, as p oposed in his wo k. In gene al, he a iable pa ame e was he ol- ume o i an , bu he change in ace ic acid concen a ion has also been epo ed. PAD analysis has been ca ied ou ei he by naked eye, digi al came a o sma phone, and EP has been de e mined based on isual de e mina ion, bigges jumps, o he in e sec ion be ween wo lines. One wo k p oposed a me hodology based on he 1 s de i a i e, simila o he one p oposed. Howe e , hese me hods a e ei he subjec i e, semi- quan i a i e o labo ious, especially when hey a e based on images aken a e each i an addi ion, as p oposed by Noguei a and co- wo ke s [50], which signi ican ly inc eases expe imen al wo k and da a analysis complexi y. The main ad an age o he p oposed spo - es ing me hod is ha he need o cap u ing a ious images is elimina ed. Ins ead, mul iple spo s a e used simul aneously, expedi ing he en i e i a ion p ocess wi h a unique image. Rega ding μ PADs, only wo wo ks we e ound whe e ace ic acid de e mina ion was ca ied ou . In he wo k o Ka i a and Kane a [44], a s a -like design wi h 10 b anches was used, i an was placed in he eac ion ese oi s, and indica o in o he de ec ion zones, as pe o med in his wo k. In ha s udy, a semi-quan i a i e analysis was pe o med, and i a ion endpoin was de e mined isually based on he colo less o pink colo change o phenolph halein indica o . In his wo k, he design was modi ied and educed o lowe lobes in o de o in es iga e i i- a ions can be ca ied ou quan i a i ely wi h jus wo poin s a ound EP. As explained in p e ious sec ions, good esul s we e ob ained wi h an 8-poin i a ion, demons a ing he p ecise applicabili y o he me hod. In he case o Dias e al. [45], ace ic acid s anda ds we e used, placed on he μ PAD, and a eled dis ance was measu ed wi h a digi al calibe . Dis ance o he colo ed channel o he μ PAD was ela ed o he analy e concen a ion o build a calib a ion cu e. Then, samples we e measu ed ollowing he same p inciple. Al hough hese app oaches a e based on acid-base i a ion, he i a ion cu e i sel is no used o acid de e mina ion, he colo change o phenolph halein may lead o inac- cu a e EP de e mina ions, and he need o build a calib a ion line in- c eases da a analysis ime. As an al e na i e, he me hod p oposed in his wo k enables he isualiza ion o a g adual colo change, which can be de ec ed by sma phone imaging, quan i a i e de e mina ion o EP can be ca ied ou , and hus, ace ic acid concen a ion can be de e - mined accu a ely based on an eigh -poin i a ion and jus one image. As men ioned o spo es ing, as and easy- o-use me hodologies ha upli p e iously epo ed wo ks we e de eloped. 4. Conclusions I was demons a ed ha he p oposed me hods o mic oscale acid- base i a ions using pape -based de ices a e eliable o he quan i i- ca ion o ace ic acid in inega samples. Two ypes o de ices – based on spo es s and μ PADs – we e de eloped and op imized o analysis, elimina ing he subjec i e de ec ion o eac ion endpoin and he need o build a calib a ion cu e. Ad ances om o he wo ks we e ob ained: i a ions we e ca ied ou using jus eigh analysis poin s and one image ob ained wi h a sma phone, a comple ely quan i a i e μ PAD was Table 3 Ace ic acid concen a ion alues gi en as acidi y pe cen age and ob ained by he e e ence me hod, spo -based PADs, and mic o luidic de ices. Rela i e e o s in compa ison o po en iome y we e calcula ed in each case. Sample Re e ence me hod Spo es s Mic o luidic analysis C ace ic (%) C ace ic (%) RE (%) Cace ic (%) RE (%) A102 10.5 ±0.2 11.3 ± 0.5 7.6 10.2 ± 0.7 −2.8 E102 10.3 ±0.3 11.2 ± 0.3 8.5 10.3 ± 1.1 0.5 A106 12.1 ±0.1 13.1 ± 0.4 8.4 10.4 ± 0.8 −13.9 D106 11.0 ±0.1 10.1 ± 1.5 −8.1 10.4 ± 0.5 −5.1 B105 9.9 ±0.1 10.1 ± 1.3 2.6 10.3 ± 0.9 5.6 E105 11.0 ±0.2 10.8 ± 0.8 −1.8 11.0 ± 0.1 0.4 F105 11.15 ±0.03 12.2 ± 0.4 9.4 10.5 ± 0.6 −6.2 B101 12.95 ±0.04 12.8 ± 0.5 −1.4 12.9 ± 0.3 −0.4 C101 14.4 ±0.1 15.0 ± 0.1 2.8 14.8 ± 1.3 2.7 F101 17.0 ±0.3 16.8 ± 1.2 −1.2 17.3 ± 0.4 1.6 Lemon inega 4.30 ±0.01 4.4 ±0.3 3.3 4.2 ±0.3 −2.8 Rice inega 4.6 ±0.2 4.53 ± 0.05 −2.0 4.3 ±0.1 −6.7 Apple inega (1) 4.8 ±0.1 4.9 ±0.2 3.2 4.8 ±0.1 1.1 Apple inega (2) 4.35 ±0.04 4.8 ±0.2 10.4 4.39 ± 0.06 0.9 Alcohol inega 4.52 ±0.05 4.8 ±0.2 4.6 4.5 ±0.2 −1.8 Table 4 Summa y o wo ks epo ed in he li e a u e o ace ic acid de e mina ion using PADs. PAD Reagen s Colo change P ocedu e De ec ion EP V o al Samples P ecision and accu acy Re . Spo s NaOH and jabo icaba peel ex ac Magen a o g een Va ying V i an Image a e each addi ion Sma phone 1 s de i a i e 20 μ L Vinega (3) RSD 2.8–4.8% RE 4.6–7.5% [50] Na 2 CO 3 and phenolph halein Colo less o pink Va ying C ace ic One image Digi al came a and ligh ing box In e sec ion o wo lines a – – [51] NaOH and phenolph halein Colo less o pink Va ying V ace ic Naked-eye Naked-eye 90 μ L Foods (15) RSD 0–12.6% RE 0.5–12.8% [43] NaOH and uni e sal indica o a Va ying V i an Image a e each addi ion Sma phone Bigges pH jump 100 mL Foods and o he s RSD 7.1–28% RE 5.8–40% [52] NaOH and li mus Red o blue Va ying V i an One image Sma phone and ligh ing box 2 nd de i a i e 110 μ L Vinega (15) RSD 0.6–13% RE 1.2–10.4% This wo k μ PAD Na 2 CO 3 and phenolph halein Colo less o pink Va ying C i an Naked-eye Naked-eye 45 μ L – – [44] NaOH and phenolph halein Colo less o pink Calib a ion line Dis ance (digi al calibe ) – 160 μ L Vinega (3) RSD 4.3–4.7% RE 1.9–2.4% [45] NaOH and li mus Red o blue Va ying C i an One image Sma phone and ligh ing box 2 nd de i a i e 34 μ L Vinega (15) RSD 0.9–9.9% RE 0.4–14% This wo k a No men ioned. I. Be asa e e al. Talan a 276 (2024) 126254 8 p oposed o he i s ime, he use o li mus and second de i a i es enabled a highly accu a e endpoin calcula ion, and he me hod was au oma ed ei he o ee image-p ocessing so wa e o o MATLAB. This way, analysis ime and complexi y, and eagen and sample ol- umes we e conside ably educed, minimizing was e gene a ion and enabling a mo e en i onmen ally- iendly me hod. Bo h me hods a e p ecise, accu a e, easy o use and as , making hem sui able o quali y con ol p ocesses and eal- ime moni o ing. CRediT au ho ship con ibu ion s a emen I a i Be asa e: W i ing – e iew & edi ing, W i ing – o iginal d a , So wa e, Me hodology, In es iga ion, Concep ualiza ion. Ane Bo da- ga ay: W i ing – e iew & edi ing. Rosa Ga cia-A ona: W i ing – e- iew & edi ing. Mi en Os a: W i ing – e iew & edi ing, Supe ision, Resou ces, P ojec adminis a ion, Concep ualiza ion. William Reis de A aujo: W i ing – e iew & edi ing, Resou ces, Concep ualiza ion. Maide Vidal: W i ing – e iew & edi ing, Supe ision, Resou ces, Concep ualiza ion. Decla a ion o compe ing in e es The au ho s decla e ha hey ha e no known compe ing inancial in e es s o pe sonal ela ionships ha could ha e appea ed o in luence he wo k epo ed in his pape . Da a a ailabili y Da a will be made a ailable on eques . Acknowledgemen s This wo k was suppo ed by he Basque Go e nmen (Resea ch G oup IT1662/22), he Uni e si y o he Basque Coun y (p e-doc o al schola ship PIF 19/131), S˜ ao Paulo Resea ch Founda ion (FAPESP) [G an numbe s: 2018/08782–1, and 2022/03250-7], and Na ional Council o Scien i ic and Technological De elopmen – CNPq (G an numbe : 310282/2022-5). Appendix A. Supplemen a y da a Supplemen a y da a o his a icle can be ound online a h ps://doi. o g/10.1016/j. alan a.2024.126254. Re e ences [1] D.C. Ha is, Quan i a i e Chemical Analysis, se en h ed., W. H. F eeman and Company, 2007. [2] E. da N. Gaiao, V.L. Ma ins, W. da S. Ly a, L.F. de Almeida, E.C. da Sil a, M.C. U. A aújo, Digi al image-based i a ions, Anal. Chim. Ac a 570 (2006) 283–290, h ps://doi.o g/10.1016/j.aca.2006.04.048. [3] D.S.A. Beeckman, G. Meesen, P. Van Oos eld , D. Van ompay, Digi al i a ion: au oma ed image acquisi ion and analysis o load and g ow h o chlamydophila psi aci, Mic osc. Res. Tech. 72 (2009) 398–402, h ps://doi.o g/10.1002/ jem .20694. [4] S. Soa es, G.M. Fe nandes, F.R.P. Rocha, Sma phone-based digi al images in analy ical chemis y: Why, when, and how o use, T ends Anal. Chem. 168 (2023) 117284, h ps://doi.o g/10.1016/j. ac.2023.117284. [5] S.E. Umbaugh, Compu e Imaging: Digi al Image Analysis and P ocessing, CRC P ess, 2005. [6] R. Deng, X. Chao, H. Li, X. Li, Z. Yang, H.Z. Yu, Sma phone-based mic opla e eade o high- h oughpu quan i a ion o disease ma ke s in se um, Analys 148 (2023) 735–741, h ps://doi.o g/10.1039/d2an01571d. [7] D. Ba zallo, L. Fe e , E. Palacio, Eco- iendly sc eening me hod o sul onamides using a 3D handheld sma phone-based luo escence de ec ion de ice and g aphene nanopla ele -packed pipe e ip mic oex ac ion, J. En i on. Chem. Eng. 12 (2024) 111888, h ps://doi.o g/10.1016/j.jece.2024.111888. [8] G. Albizu, M. Vidal, A. Bo daga ay, R. Ga cía-A ona, M. Os a, Semi-au oma ic digi al image con ol o nickel concen a ion in a Ni-P elec oless acid ba h using he hexaamminenickel blue complex, Mic ochem. J. 199 (2024) 110073, h ps:// doi.o g/10.1016/j.mic oc.2024.110073. [9] M. Zugazua-Ganado, A. Bo daga ay, J. Ezena o, R. Ga cia-A ona, M. Os a, Adap a ion o he Folin-Ciocal eu and Fas Blue BB spec opho ome ic me hods o digi al image analysis o he de e mina ion o o al phenolic con en : educ ion o eac ion ime , in e e ences and sample analysis, LWT–Food Sci. Technol. 193 (2024) 115756, h ps://doi.o g/10.1016/j.lw .2024.115756. [10] D.A. Volme , L. Cu bani, T.A. Pa ke , J. Ga cia, L.D. Schul z, E.M. Bo ges, De e mina ion o i a able acidi y in wine using po en iome ic, Conduc ome ic, and Pho ome ic me hods, J. Chem. Educ. 94 (2017) 1296–1302, h ps://doi.o g/ 10.1021/acs.jchemed.6b00891. [11] L.M.A. de Oli ei a, V.B. dos San os, E.K.N. da Sil a, A.S. Lopes, H.A. Dan as-Filho, An en i onmen - iendly spo es me hod wi h digi al imaging o he mic o- i a ion o ci ic ui s, Talan a 206 (2020) 120219, h ps://doi.o g/10.1016/j. alan a.2019.120219. [12] K. Ru anako n, N. Phadungcha oen, W. Laiwa anapaisal, A. Chins iwongkul, T. Rojana a a, Sma phone-based echnique o he de e mina ion o a i a ion equi alence poin om an RGB linea -segmen cu e wi h an example applica ion o minia u ized i a ion o sodium chlo ide injec ions, Talan a 233 (2021) 122602, h ps://doi.o g/10.1016/j. alan a.2021.122602. [13] L.A. P adela-Filho, W.B. Veloso, I.V.S. A an es, J.L.M. Gongoni, D.M. de Fa ias, D. A.G. A aujo, T.R.L.C. Paix˜ ao, Pape -based analy ical de ices o poin -o -need applica ions, Mic ochim. Ac a 190 (2023) 179, h ps://doi.o g/10.1007/s00604- 023-05764-5. [14] E. No iana, T. Oze , C.S. Ca ell, J.S. Link, C. McMahon, I. Jang, C.S. Hen y, Mic o luidic pape -based analy ical de ices: om design o applica ions, Chem. Re . 121 (2021) 11835–11885, h ps://doi.o g/10.1021/acs.chem e .0c01335. [15] T. Oze , C. McMahon, C.S. Hen y, Ad ances in pape -based analy ical de ices, Annu. Re . Anal. Chem. 13 (2020) 85–109, h ps://doi.o g/10.1146/annu e - anchem-061318-114845. [16] L.F. de Lima, A.L. Fe ei a, I. Ranjan, R.G. Collman, W.R. de A aujo, C. de la Fuen e-Nunez, A bac e ial cellulose-based and low-cos elec ochemical biosenso o ul asensi i e de ec ion o SARS-CoV-2, Cell Repo s Phys. Sci. 4 (2023) 101476, h ps://doi.o g/10.1016/j.xc p.2023.101476. [17] M. Dong, J. Wu, Z. Ma, H. Pe e z-So oka, M. Zhang, P. Komenda, N. Tang i, Y. Liu, C. Riga o, F. Lin, Rapid and low-cos CRP measu emen by in eg a ing a pape - based mic o luidic immunoassay wi h sma phone (CRP-Chip), Senso s 17 (2017) 684, h ps://doi.o g/10.3390/s17040684. [18] M.S. Ka ai, L.F. de Lima, W.R. de A aujo, A disposable and low-cos lase -sc ibed g aphene elec ochemical senso o simul aneous de ec ion o hyd oquinone, pa ace amol and me hylpa aben, Ma e . Le . 330 (2023) 133211, h ps://doi.o g/ 10.1016/j.ma le .2022.133211. [19] M. S isa-A , K.E. Boehle, B.J. Geiss, C.S. Hen y, Highly sensi i e de ec ion o Salmonella yphimu ium using a colo ime ic pape -based analy ical de ice coupled wi h Immunomagne ic sepa a ion, Anal. Chem. 90 (2018) 1035–1043, h ps://doi.o g/10.1021/acs.analchem.7b04628. [20] S. Ka i a, T. Kane a, Chela e i a ions o Ca2+and Mg2+using mic o luidic pape - based analy ical de ices, Anal. Chim. Ac a 924 (2016) 60–67, h ps://doi.o g/ 10.1016/j.aca.2016.04.019. [21] S.A. Noguei a, A.D. Lemes, A.C. Chagas, M.L. Viei a, M. Talha ini, P.A.O. Mo ais, W.K.T. Col o, Redox i a ion on oldable pape -based analy ical de ices o he isual de e mina ion o alcohol con en in whiskey samples, Talan a 194 (2019) 363–369, h ps://doi.o g/10.1016/j. alan a.2018.10.036. [22] N. Tap ab, Y. Sameenoi, Rapid sc eening o o maldehyde in ood using pape - based i a ion, Anal. Chim. Ac a 1069 (2019) 66–72, h ps://doi.o g/10.1016/j. aca.2019.03.063. [23] O. Heida y, M. Akhond, B. Hemma eenejad, A mic o luidic pape -based analy ical de ice o iodome ic i a ion o asco bic acid and dopamine, Mic ochem. J. 182 (2022) 107886, h ps://doi.o g/10.1016/j.mic oc.2022.107886. [24] A. Tep ek, V. Poe i A ono, W. Waiyawa , A. Limsakul, J. Shiowa ana, A. Si ipinyanond, Semi-quan i a i e analysis by spo coun ing on o igami pape - based de ice o endpoin de ec ion in i ime ic analysis, Mic ochem. J. 158 (2020) 105284, h ps://doi.o g/10.1016/j.mic oc.2020.105284. [25] O.D. Oyewunmi, S.H. Sa iabadi-Tali, S. Jahanshahi-Anbuhi, Dual-modal assay ki o he quali a i e and quan i a i e de e mina ion o he o al wa e ha dness using a pe manen ma ke ab ica ed mic o luidic pape -based analy ical de ice, Chemosenso s 8 (2020) 1–19, h ps://doi.o g/10.3390/chemosenso s8040097. [26] E. Ghohes ani, J. Tashkhou ian, B. Hemma eenejad, Colo ime ic de e mina ion o pe oxide alue in ege able oils using a pape based analy ical de ice, Food Chem. 403 (2023) 134345, h ps://doi.o g/10.1016/j. oodchem.2022.134345. [27] F.S. Wagne , U.B. S a , Ace ic acid, Ki k-O hme Encycl, Chem. Technol. (2014) 1–21. [28] C.W. Ho, A.M. Lazim, S. Faz y, U.K.H.H. Zaki, S.J. Lim, Va ie ies, p oduc ion, composi ion and heal h bene i s o inega s: a e iew, Food Chem. 221 (2017) 1621–1630, h ps://doi.o g/10.1016/j. oodchem.2016.10.128. [29] P. Gim´ enez-G´ omez, M. Gu i´ e ez-Capi ´ an, J.M. Ríos, F. Capde ila, A. Puig-Pujol, C. Jim´ enez-Jo que a, Mic oanaly ical low sys em o he simul aneous de e mina ion o ace ic acid and ee sul u dioxide in wines, Food Chem. 346 (2021) 128891, h ps://doi.o g/10.1016/j. oodchem.2020.128891. [30] F. Benmeziane De adji, N. Ghanemi, L. Dje moune A koub, Compa ing physicochemical, nu i ional and biochemical p ope ies o h ee adi ional ig inega s p oduced in Alge ia, Ca pa hian J. Food Sci. Technol. 15 (2023) 147–158, h ps://doi.o g/10.34302/c pj s /2023.15.2.13. [31] K.M.J.P. Boua enin, K.A. Kouam´ e, M.E. Gueu-Kehi, N.T. Dj´ eni, K.M. Dj` e, O ganic p oduc ion o inega om mango and papaya, Food Sci. Nu . 9 (2021) 190–196, h ps://doi.o g/10.1002/ sn3.1981. [32] R.M. Ca aballo, L.M. Saleh Medina, S.G.J. Gomez, P. Vensaus, M. Hame , Tu me ic and RGB analysis: a low-cos expe imen o eaching acid-base equilib ia a home, I. Be asa e e al. Talan a 276 (2024) 126254 9 J. Chem. Educ. 98 (2021) 958–965, h ps://doi.o g/10.1021/acs. jchemed.0c01165. [33] S. ang Supha oek, K. Ponhong, W. Si iangkhawu , K. G udpan, Employing na u al eagen s om u me ic and lime o ace ic acid de e mina ion in inega sample, J. Food D ug Anal. 26 (2018) 583–590, h ps://doi.o g/10.1016/j. j da.2017.06.007. [34] S. Kasemsum an, A. Boondaeng, S. Jung hee apanich, K. Ngowsuwan, W. Apiwa anapiwa , P. Janchai, P. Vai hanomsa , Assessing e men a ion b o h quali y o pineapple inega p oduc ion wi h a nea -in a ed ibe -op ic p obe coupled wi h s abili y compe i i e adap i e eweigh ed sampling, Molecules 28 (2023) 6239, h ps://doi.o g/10.3390/molecules28176239. [35] M. Mo mino, V. Siewe s, Y. Nygå d, De elopmen o an Haa1-based biosenso o ace ic acid sensing in Saccha omyces ce e isiae, FEMS Yeas Res. 21 (2021) oab049, h ps://doi.o g/10.1093/ emsy / oab049. [36] G. Che, D. Gu, W. Yang, M. Li, Q. Pan, Tu n-on luo escence de ec ion o ace ic acid in wine using a u anyl-o ganic amewo k, C ys . G ow h Des. 22 (2022) 1984–1990, h ps://doi.o g/10.1021/acs.cgd.2c00016. [37] S.T. Alemneh, S.A. Emi e, M. Jekle, B. Hi zmann, E ec o e ige a ed s o age on some physicochemical cha ac e is ics o a e -based e men ed be e age and he iabili y o he e men ing Lac iplan ibacillus plan a um and Lac icaseibacillus hamnosus used, J. Food P ocess. P ese . 46 (2022) e17034, h ps://doi.o g/ 10.1111/j pp.17034. [38] Z.B. Güzel-Seydim, G. S ¸a ı , Ç. G¨ okı maklı, Use o Manda in and pe simmon ui s in wa e ke i e men a ion, Food Sci. Nu . 11 (2023) 5890–5897, h ps://doi.o g/ 10.1002/ sn3.3561. [39] A. Liu, W. Pan, S. Li, J. Li, Q. Li, L. He, S. Chen, K. Hu, X. Hu, G. Han, S. Li, J. Zhou, F. Chen, S. Liu, Seasonal dynamics o mic obio a and physicochemical indices in he indus ial-scale e men a ion o Sichuan Baoning inega , Food Chem. X. 16 (2022) 100452, h ps://doi.o g/10.1016/j. ochx.2022.100452. [40] S.H. Mohamed, A.I. Salim, Y.M. Issa, A.E. Ali, De ec ion and iden i ica ion o adul e a ion in inega samples based on e e sed-phase high-pe o mance liquid ch oma og aphic (RP-HPLC) s a egies, ACS Food Sci. Technol. 2 (2022) 21–30, h ps://doi.o g/10.1021/acs oodsci ech.1c00169. [41] S. Huang, S. Ying Lim, H. Lau, W. Ni, S. Fong Yau Li, E ec o glycinebe aine on me aboli e p o iles o cold-s o ed s awbe y e ealed by 1H NMR-based me abolomics, Food Chem. 393 (2022) 133452, h ps://doi.o g/10.1016/j. oodchem.2022.133452. [42] N.S. Oli ei a, A.F. Oli ei a, A.A. Ne es, M.E.L.R. Quei oz, De elopmen o a non- i a ion po en iome ic me hod o he de e mina ion o acidi y in inega , Talan a 256 (2023) 124261, h ps://doi.o g/10.1016/j. alan a.2023.124261. [43] J. Huang, L. Cai, A comp ehensi e expe imen o aining unde g adua es majo ing in nu i ion educa ion using a pape -based mic o acid-base i a ion, J. Chem. Educ. 99 (2022) 3607–3612, h ps://doi.o g/10.1021/acs. jchemed.2c00211. [44] S. Ka i a, T. Kane a, Acid −base i a ions using mic o l uidic pape -based analy ical de ices, Anal. Chem. 86 (2014) 12108–12114, 10.1021/ac5039384. [45] L.G.A. Dias, L.C. Dua e, K.M.P. Pinhei o, N.S. Mo ei a, W.K.T. Col o, Dis ance- based pape mic o luidics as en i onmen ally iendly pla o ms o moni o ing acid-base i a ions, Talan a Open 7 (2023) 100216, h ps://doi.o g/10.1016/j. alo.2023.100216. [46] Guidelines o S anda d Me hod Pe o mance Requi emen s, 2016, h ps://doi. o g/10.1093/9780197610145.005.006. [47] H.C. Goicoechea, A.C. Oli ie i, Sus ained p edic ion abili y o ne analy e p ep ocessing me hods using educed calib a ion se s. Theo e ical and expe imen al s udy in ol ing he spec opho ome ic analysis o mul icomponen mix u es, Analys 126 (2001) 1105–1112, h ps://doi.o g/10.1039/b100422k. [48] A.G. Gonz´ alez, M.A. He ado , A.G. Asue o, In a-labo a o y es ing o me hod accu acy om eco e y assays, Talan a 48 (1999) 729–736, h ps://doi.o g/ 10.1016/S0039-9140(98)00271-9. [49] F. Pena-Pe ei a, W. Wojnowski, M. Tobiszewski, Ag ee - analy ical GREEnness me ic app oach and so wa e, Anal. Chem. 92 (2020) 10076–10082, h ps://doi. o g/10.1021/acs.analchem.0c01887. [50] S.A. Noguei a, L.R. Sousa, N.K.L. Sil a, P.H.F. Rod igues, W.K.T. Col o, Moni o ing acid-base i a ions on wax p in ed pape mic ozones using a sma phone, Mic omachines 8 (2017) 1–10, h ps://doi.o g/10.3390/mi8050139. [51] P. Namwong, P. Ja ujam us, M. Ama a ongchai, S. Chai am, Fab ica ing simple wax sc een-p in ing pape -based analy ical de ices o demons a e he concep o limi ing eagen in acid-base eac ions, J. Chem. Educ. 95 (2018) 305–309, h ps:// doi.o g/10.1021/acs.jchemed.7b00410. [52] J. Li, M.L. O’Neill, C. Pa ison, J.H.W. Zhou, J.M. I o, C.S.T. Wong, H.Z. Yu, N. Me bouh, Mobile app o quan i y pH s ips and moni o i a ions: sma phone- aided chemical educa ion and class oom demons a ions, J. Chem. Educ. 100 (2023) 3634–3640, h ps://doi.o g/10.1021/acs.jchemed.3c00227. I. Be asa e e al.