RICERCA ITALIANA     

Developement, characterisation, and analytical applications of innovative electrochemical sensors

Università degli Studi di Modena e Reggio Emilia

Abstract

The Project collects eight Units of Research, among which active collaborations in the field of Electroanalytical Chemistry have been active for a long time. The aim is to convey efforts towards the development of new electrochemical sensor systems for the analysis of compounds of interest in the field of food chemistry and for the in situ analysis of food matrices. The developed systems, consisting of modified electrodes, will be for the most part amperometric sensors, though attention will be also devoted to potentiometric sensors. The modifiers will be ion exchanger polymers, redox polymers and conducting polymers, permselective coatings, self assembled monolayers, multilayer polyelectrolytes, carbon nanotubes, metal oxides and metal nanoparticles, as such, as well as included inside conducting polymers. Some of these coatings will also be suitable for anchoring enzymes to the electrode. This constitutes a modern approach to the development of efficient biosensors.
The new sensors will be characterised as to all aspects necessary to define the principles on which they work, by using the wide number of techniques available to the Units of Research involved. Together with the more or less conventional electrochemical techniques, sophisticated spectroscopic techniques, as well as techniques for defining surface morphology, will be employed. Their measures will be often coupled to the suitable electrochemical ones. After testing on conventional-size electrodes, the systems will be promptly transferred to microelectrodes, since they are only suitable to lead to faithful responses when working directly on the real matrix.
The sensors will be basically of two different types, designed and realised to satisfy two different requirements. A first class will consists of systems in which specificity of response will be sought, in order to perform quali- and quantitative definition of the chemical system. A second class will involve systems giving relatively complex reproducible responses that have to significantly vary for small changes in the nature of the matrix. These are non-specific sensors, in which characteristics are sought, such as, for an amperometric sensor, a corrent/potential signal that, though 'hidden', contains information on the nature of the matrix as a whole, rather than dealing with details on composition. The chemometric treatment, also performed by ad-hoc elaborated algorithms, should lead to discrimination among different matrices and allow identification of a matrix. Sensors like these will be used under the form of ensemble, and a suitable chemometric treatment will judge about the usefulness of adding or not one more element. Such an ensemble of sensors, once properly miniaturised, will constitute an 'electronic tongue'.
The matrices will be chosen among those of interest in the field of foodstuffs, both for characterising, and hence typicising them, and for identifying adulterations, sophistications, or dangerous additives. One aim is also that of managing to reach such a point of definition of the capabilities, to apply procedures for validation of systems and methodologies.
The Project, through very significant involvement of young researchers and of complementary expertise of the different Units of Research, will also pursue the objective of proper transfer of knowledge inside the Project itself, devoting particular attention to young people. On the other hand, Project Seminars, which will be open to anyone and suitably advertised also through a website, scientific publications and Communications to Congresses, and a final Project Workshop, to which Italian and foreign experts will be also invited, will be the most evident way for diffusing the results obtained. <<<

Principal Investigator

Renato SEEBER Università degli Studi di MODENA e REGGIO EMILIA

Research Objectives

The Project of Research collects a significant number of Units of Research working in the field of Electroanalytical Chemistry in Italy. In particular, it involves groups that devoted in the past great attention to the characterisation of the analytical system considered from all points of view, not only as regards development of analytical methodologies in the strict meaning of the term. This means studying the way in which a system works, not only defining conditions and limits within which it works. Furthermore, attention has always been paid to developing of new systems, based on new materials and new technologies. To this purpose, the groups have arranged suitable interfaces on the one side with inorganic and organic chemists and, on the other side, with physicists and engineers.
The present Project aims at developing new electrochemical sensors, potentiometric and, most of all, amperometric sensors, as well as at characterising them from a structural and a morphological point of view, at understanding physical and chemical principles on which they actually work, at evaluating their performances. Design and development of such devices will be directed to use them with respect to analytes of interest in food chemistry, and directly on food matrices. Liquid, but also semi-liquid food materials will be considered.
Two different types of sensors will be taken into account. More precisely, both the way they are made and the way they work will point to two different goals. In a first group non-specificity in the response will be sought: materials will be studied and conditions will be adopted in order that responses selective with respect to given analytes or classes of analytes are obtained. On the other hand, in a second group of sensors, developed for performing 'blind analyses', specificity will not be the characteristic of choice. The whole procedure will be finalised to collect best information characterising and discriminating the matrix under study considered as a whole, rather than to define its detailed analytical composition. The results of suitable chemometric treatments of the signal from the sensor will suggest modifications of the experimental conditions and even of the nature of the sensor chosen.
A suitable number of similar sensors will build up a so-called 'artificial' or 'electronic tongue', capable of identifying the matrix as to its taste. Micro or nanoelectrodes will be necessarily used, in order to minimise the 'damage' of the measuring system to the matrix, in order not to have to alter it in any way also when working on matrices very hard to work on by using conventional electrodes. The final step will hence consist in developing an array or ensemble of electrochemically hybrid nano- or microelectrodes. The different responses, considered as a whole, will be elaborated through suitable signal processing and regression/classification techniques, leading to a sort of 'digital printing' of the matrix under examination, suitable to characterise it efficiently. The research will also deal with the 'olfactory' characteristics of the matrix. They will be defined by an 'artificial' or 'electronic nose' that is currently used by a Group of Research involved in the Project. Depending on the adequacy of the responses of the artificial tongue-nose system, it is however possible that the available electronic nose is substituted by an ensemble of electrochemical sensors, or by an hybrid new electronic nose possibly involving electrochemical, microgravimetric, and spectrophotometric sensors, developed in the frame of the present Project.
Inside the groups of research involved in the Project the necessary expertise is present to carry out the tests necessary for possibly validating procedures that, during the work to the different goals of the Project, could be defined up to the point to be optimised as to accuracy, reproducibility, and robustness.
Bringing different groups together towards common goals inside a common Project of Research should supposedly imply a significant 'added value' with respect to simply adding the single contributions of the different groups. Many aims are pursued. It is a Project of Research dedicated to universities, financially supported by Ministry for Education, University, and Research. According to the meaning of 'fundamental research', which can be also found in the National Planning of Research, it has to be intended as 'oriented fundamental research'. Hence, most attention is devoted to the processes on which the way a new device or a new procedure work, is based, but particular care is also devoted to applications and even technological aspects. Most attention is also devoted to transfer of knowledge: research has also to be intended as training to a methodology of research for young researchers, through a direct involvement of them in advanced research themes. Many young researchers are hence present inside the different groups, and a considerable portion of the financial support expected is just devoted to recruit young researchers through grants for Doctorate, PostDoc positions, etc..
In planning the composition of the ensemble of groups involved in the Project one of the criteria has been that of achieving efficient complementary and relatively heterogeneous expertise, within the limits of a strong common scientific interest that can guarantee the goal or, better, the goals of the research to be reached. Some aspects of great importance in the frame of scientific project supported by EU are also well considered in the Project, though being it a national Project. Internalisation, of course, is not present under the form of a direct participation, but it is actually well inside the Project when considering the numerous international collaborations that every Unit of Research has active; they will play a very important role in the course of the execution of the Project. The issue of equal opportunity for both genders is not actually a problem inside the composition of the groups: the fact that only in one case a woman is the Local Responsible is due to historical reasons. The Project gives a contribution to overcome such a situation.
In our intention, it was not profitable to envision a very limited goal, excluding a priori some other contiguous ones. The title given to the Project gives reason of the strong common scientific basis, which will allow a scientific goal and a strong interaction among the different groups, including exchange of expertise and of young researchers, to be achieved. The expertise of each group does not overlap that of any other group, neither is it too much restricted. This will made exchanges from one group to another most profitable.
The field on which the developed devices will work, will be that of products of interest in food, as well as food matrices themselves. They constitute quite a wide field, however bound to one another by a very strong common link. This will allow us to test the versatility of the devices and, at the same time, to once more exchange different experiences, which holds for both young and less young researchers involved. The already existing collaborations inside the Group are quite a trustworthy starting point for planning and finally carrying out a necessarily wide-team work. <<<

First Results

The first phase has to lead to precise formulation of systems suitable to specific determinations. This will introduce, in turn, to complete characterisation of the developed systems and to application of them to the study of synthetic and natural matrices. Since now on, it will be carefully considered that the final objective will consist of different food matrices, under liquid or semiliquid phase: from milk to fruit juice, from alcoholic beverages to fresh cheeses, to honey. The analytes considered, in view of the foodstuffs chosen, will be additives like dyes, anti-oxidants, preservatives, and some toxins, particularly mycotoxins.
The tests performed with single electrodes on synthetic matrices will lead to a first, though precise enough indication of the electrode systems that can be proposed in the frame of an ensemble for building up an artificial tongue. It is important to notice that it can be hypothesised that different ensembles will be identified as the most suitable ones in different situations. Only in a final stage the best compromise will be chosen.
From a chemometric point of view, the best experimental conditions will be identified, in order to obtain maximum information content from the employed sensors and, at the same time, the most promising algorithms will be tested and submitted to a final choice, in order to develop multivariate models, on the basis of the responses of the single sensors. The collected information will give elements for drawing a sort of classification for the efficiency of the different sensors in the different situations.When arriving at the final act of the Project, a number of new electrode systems are expected to have been developed. Some of them will possess specificity, some other non-specificity characteristics. The most part will be at a miniaturised level. As to the selective sensors, the most part of those that have shown to deserve maximum interest will have been exhaustively characterised and the relevant performances will have been defined on simulated matrices. Some devices will have also been tested on real matrices. Finally, it is faithfully believed that it will have been possible to fully establish a few analytical procedures, so that we could activate the required further tests for validation of the method.
As regards non-specific sensors and sensor ensembles, one or more realistic hypothesis of electronic tongue will have been realised, together with the relevant experimental and statistic tests supporting the composition in terms of sensors. The package will also contain the software based on the algorithms that will be identified as the most effective ones in the elaboration of the collected signals. It does not seem realistic, just now, to forecast the precise point at which the research will be arrived as to a satisfactory testing of the proposed electronic tongue in the different situations. It is only possible to think at some assay tests, useful in order to plan the subsequent conclusive act. <<<

Timescale

24 months

National and international background

In the last years electroanalytical techniques have re-attracted most part of the great attention that was devoted to them in the first two decades after the Second World War, and that had then progressively faded. This happened for many reasons, which originated essentially from the strong increase of the potentialities of old and new spectroscopic and chromatographic analytical techniques. This led to a virtuous circle, which has led the producers to computerise efficiently the management of the measurements and to develop suitable software for treating the relevant signals. The use of these techniques became rapidly appealing and well user-friendly. Furthermore, the widespread diffusion of the spectroscopic and chromatographic instrumentation allowed the relevant costs to decrease. On the other hand, also due to the instrumental development of these techniques, the approach to electroanalytical techniques became comparably more difficult and almost suspicious: the supplier of electrochemical instrumentation could not invest enough money in research and development to allow them to compete successfully. As a conclusion, who needed analytical determinations barely considered the possibility to use electrochemical methodologies. In recent times, the availability of electrochemical instrumentation equipped with suitable software allowed the use of electrochemical techniques to those who estimated their execution and the evaluation of the results obtained too cumbersome. At the end, we may now conclude that a virtuous circle has begun for these techniques, in which the request of the market attracts the interest of the producers and vice versa.
At the same time, new amperometric sensors have assumed an importance comparable to that of the potentiometric ones. In turn, definite improvements have involved potentiometric sensors thanks to the development of new polymeric membranes for Ion Selective Electrodes, which allowed the limit of detection to be improved considerably [1-3].
A first winning card to play by the controlled potential electrochemical techniques in the two or three last decades consisted in the development of systems in which it is possible to couple speed of analysis with possibility of working in situ, without the necessity of significantly altering the matrix to work on. Microelectrodes constitute soft probes for the matrix, which can operate at an ionic strength not higher than that naturally present in real matrices [4-7]. The electrochemical responses obtained are not dramatically affected by the high electrical resistance of the medium, so much so that it is even possible to work in semi-liquid matrices [4,8].
An additional winning card assuming growing and growing importance consists of the modifications of the electrode surfaces, which permit the modification of the physico-chemical characteristics of the surface, allowing an increase in selectivity and sensitivity of the analytical responses to be gained [9,10]. However, the recent progresses in the field of new materials, particularly as regards nanostructured and functionalised, as well as composite materials, have not been yet, for the very most part, exploited in electrochemical applications. As a consequence, together with the relatively consolidated modifications of electrode surfaces such as those consisting of conventional ion-exchanger [10-13] and redox [10] polymers, whose potentialities are however still far from being fully defined, the field of conducting polymers [12-17], of metal oxides [18, 19], of metal nanoparticles [20-26], of self assembled monolayer [27-29], of Langmuir-Blodgett films [30], of polyelectrolytes [31], of permselective membranes [32-36], of (functionalised) carbon nanotubes - SWCTT and MWCT [37-39], suffer of a heavy delay with respect to the progress in the development of such materials, both as such and with respect to other different applications. On the other hand, it is easy to predict that the delay is going to increase more and more, and the number of 'lost occasions' to grow as well, in view of the low number of researchers involved in the application of new electrochemical systems to electroanalysis. As an example, it is common feeling that biosensors [40] are far from satisfying common expectations. However, it should be taken into account that a relatively low number of researchers work in the field, and that the research along this line necessitates of high costs, with respect to the amount of the supporting funds. This notwithstanding, in the last years new, quite interesting systems, based on redox mediators, are proposed as support for enzymes: some of the Units of Research involved in the present Project work along this direction.
Hyphenated techniques have recently involved also electrochemical techniques. Less common couplings have recently joined established ones, such as that of the controlled potential techniques with UV-visible spectroscopy [9,41]. We not only refer to infrared spectroscopy, especially in reflactance mode [42,43], but also to Surface Plasmon Resonance (SPR) [44-48], which is at the beginning of its story, though commercial instruments, coupling the spectroscopic to the electrochemical measurement, are already available on the market, and to Electrogenerated ChemiLuminescence (ECL) [49-52]. Furthermore, surface techniques like AFM end STM are coupled to electrochemical techniques [53-58] in surface morphological studies that can be considered of great analytical interest. Electromicrogravimetric techniques (EQCM) are used in measurements of mass variations of electrodes in order to monitor different possible phenomena connected with polarisation at a given potential [59-63]; furthermore, electrochemical techniques have also been coupled to mass spectrometry [61,64]. Magnetic fields have been found to affect the state of the electrode surface (e.g. crystallisation processes) and of the diffusion layer. The phenomenon is studied with respect to preparative electrochemistry [65-67]; it is however easy to envision interesting applications also in the field of electroanalysis. We also want to notice that an electroanalytical technique, i.e. the Scanning Electrochemical Microscopy - SECM [63,68-75] - is used in the study of the morphology of both conducting and insulating surfaces, as well as in the characterisation of surfaces separating two different phases. It is important to evidence that the possibility of coupling conventional electrochemical techniques to the long list of above reported techniques offers the possibility to study the same electrochemical phenomenon from two different points of view that are often complementary to each other. This is of fundamental importance when the aim is to achieve full comprehension of the principles on which a sensor works, in general and with specific reference to the analyte of interest.
The application of sensors, in particular of electrochemical sensors, to the in situ study of real matrices, which implies only minor, if any, manipulation of the matrix, is a more and more urgent necessity. It may happen that attention is devoted to one of the different analytes present in a matrix, or rather to the characterisation of the matrix as a whole, aiming at defining the overall characteristics of quality, on the basis of the different sensorial properties: 'solidity' (measured since 1940 [76]), 'colour' (in the sense of the well known colourimetric measurements, using parameters such as CIE [77] and CIELab [78], based on quantities drawn out from absorption spectra in the visible region) or, even more general, 'visible aspect' (by the modern multivariate image analysis [79]), 'odour' (olfactometry = 'artificial' or 'electronic nose', proposed in the last twenty years [80-82]), 'taste' ('artificial' or 'electronic tongue', which has been studied in quite recent times [83-92]). The goals of such analytical methods vary over a wide range, i.e. from the necessity of characterising the origin of the matrix (e.g. typicisation), to the definition of its use, to the on-line monitoring of a process, and to the identification of adulterations of different possible origin and nature. The notable complexity of the problems involved in the application of sensors to the analysis of real matrices should be ascribed to many causes: the matrix under examination, which is often composed by a high number of components strongly interacting with one another, the analytical technique chosen, which gives signals that can be only rarely 'directly' interpreted, as well as the property under study, which in many cases is not connected only to a single physical quantity, being rather often linked to complex multivariate relationships among variables of different nature. For this reason, the use of the whole of information content coming from different non-specific sensors is often more efficient; the information should be elaborated by chemometric techniques based on 'blind analysis' methods. Such an approach, in which no a priori assumption is made about the most significant portions of the collected signals, may be employed for analysing the responses from the sensors in a relatively simple and relatively fast way. Blind analysis techniques may also be employed for the analysis of micro- or nanoelectrode array or ensemble [93-97].
Chemometric techniques constitute a fundamental tool for i) planning, developing, and optimising the experimental methodologies for the acquisition of instrumental signals according to experimental design techniques; ii) evaluating the distribution of the analysed samples, evidencing similarities, clustering, or anomalous behaviour (clustering, PCA, etc.); iii) evidencing the regions of the signal bearing the most significant information; iv) building up classification models, in order to estimate if the information from the sensors allows the samples to be discriminated into suitable classes; v) building up models for calibration, in order to quantitatively evaluate, on the basis of the electrochemical measures as a whole, one or more properties of interest. To this purpose, mainly in the field of electrochemical sensors to use as artificial tongue, the techniques most used nowadays are Principal Component Analysis (PCA) as a Pattern Recognition technique and Principal Component Regression (PCR) and/or Partial Least Squares (PLS) and Artificial Neural Networks (ANN) as to classification and calibration problems [83,98,99]. The relatively low number of techniques usually employed is mainly due to the fact that only few chemometric techniques are actually present in commercial software packages that are possibly coupled to the analytical instrument. Although the methods cited have proven to work satisfactorily in most of the proposed applications, coupling them with suitable information compression /variables selection methods could be profitable tool from many viewpoints. Among these, the achievement of better predictive capability of the obtained models, of higher robustness with respect to ageing/fouling of the electrode and, in some cases, of better comprehension of the obtained results. To this purpose, several methods that are not used in commercial applications of sensors are actually very promising ones. Among these, let's cite the calibration methods based on orthogonalisation or stepwise decorrelation of predictors [100] or classification/class modelling methods such as the potential functions [101,102]. In this context, algorithms based on Wavelet Transform (WT) [103], which allow one to contemporary consider both local aspects and 'overall shape' of the signal, have already been proven to be very efficient in the few examples of application to electrochemical signals, in particular to those coming from electronic tongues [104-111]. It seems very important to go further along this direction when dealing with applications of electrochemical sensoristics to real matrices.
The arguments reported above do not want at all to induce concluding for scarce utility of sensors with high specificity in the determination of given analytes, in the context of a study of the matrix. The application of sensors to food matrices should actually be directed towards two directions: i) nonspecific analysis, bearing high content of information about the matrix as a whole; ii) quali- and quantitative definition of specific analytes of interest, such as 'precious' or characterising substances, as well as analytes that, for different reasons, are dangerous to health.
All Units of Research involved in the Project possess acknowledged competence in one or another of the fields that convey to define the required overall expertise. Starting from the development and characterisation of new electrode modifiers, from ion exchangers (UNIVE, UNIUD, UNIMI), to redox polymers (UNIMORE), to conducting polymers (UNIMORE, UNIPR), to metal oxides and metal inclusions (UNIPZ, UNIUD, UNIMI), to functionalised nanotubes (UNIPD), up to blind analysis of signals (UNIGE, UNIMI, UNIMORE) and to validation procedures of new devices and methodologies (UNIGE, UNIMI). The specific competency of each single Unit is evidenced in the individual subproject. <<<