Voltammetry belongs to a category of electro-analytical methods, through which information about an analyte is obtained by varying a potential and then measuring the resulting current. It is, therefore, an amperometric technique. Since there are many ways to vary a potential, there are also many forms of voltammetry, such as: polarography (DC Voltage) [43], linear sweep, differential staircase, normal pulse, reverse pulse, differential pulse and more [3, 44]. Cyclic voltammetry is one of the most widely used forms and it is useful to obtain information about the redox potential and electrochemical reaction rates (e.g. the chemical rate constant) of analyte solutions. In this case, the voltage is swept between two values at a fixed rate, however, when the voltage reaches V₂ the scan is reversed and the voltage is swept back to V₁, as is illustrated in Figure 4i). The scan rate, (V₂ − V₁)/(t₂ − t₁), is a critical factor, since the duration of a scan must provide sufficient time to allow for a meaningful chemical reaction to occur. Varying the scan rate, therefore, yields correspondingly varied results [3, 45].

The voltage is measured between the reference electrode and the working electrode, while the current is measured between the working electrode and the counter electrode. The obtained measurements are plotted as current vs. voltage, also known as a voltammogram. As the voltage is increased toward the electrochemical reduction potential of the analyte, the current will also increase. With increasing voltage toward V₂ past this reduction potential, the current decreases, having formed a peak as the analyte concentration near the electrode surface diminishes, since the oxidation potential has been exceeded. As the voltage is reversed to complete the scan toward V₁, the reaction will begin to reoxidize the product from the initial reaction. This produces an increase in current of opposite polarity as compared to the forward scan, but again decreases, having formed a second peak as the voltage scan continues toward V₁. The reverse scan also provides information about the reversibility of a reaction at a given scan rate [46].

The shape of the voltammogram for a given compound depends not only on the scan rate and the electrode surface, which is different after each adsorption step, but can also depend on the catalyst concentration. For example, increasing the concentration of reaction specific enzymes at a given scan rate will result in a higher current compared to the non-catalyzed reaction [45, 48, 49]. Figure 4 provides an overview of selected cyclic voltammograms from the work of Li et al. [47]. They describe a cholesterol biosensor based on the entrapment of cholesterol oxidase in a silicic sol-gel matrix (please refer to Section 3.5.5.) at a Prussian Blue-modified electrode. The electrode modification is aimed to improve the sensitivity and the selectivity of the biosensor. By means of CV they were able to determine a 3σ detection limit of nearly 1.2×10⁻⁷ M compared to other reported detection limits on the order of 10⁻⁵ M from literature at the time. They further characterized their system using chronoamperometry (see Section 2.1.2.) and obtained a 3σ detection limit of 5.4×10⁻⁷ M. CV is not only useful for sensing purposes, but also as a tool to characterize the processes that take place on the sensing electrode. Liu et al. have for example studied the electrochemical characteristics and characterized the modified process of a biosensor for the determination of the H₂O₂ concentration with reportedly high sensitivity. Horseradish peroxidase (HRP) was immobilized on a platinum disk electrode modified with a gold colloid-cysteine-nafion film in an approach to enhance the direct electron transfer between enzymes and electrodes without chemical mediators [48].

Another amperometric technique is known as chronoamperometry, where a square-wave potential is applied to the working electrode and a steady state current is measured as a function of time [3, 46]. Alterations in the current arise from the expansion or reduction of the diffusion layer at the electrode. The concept of a diffusion layer was introduced by Nernst and states that there is a stationary thin layer of solution in contact with the electrode surface. The local analyte concentration drops to zero at the electrode surface and diffusion controls the transfer of analyte from the bulk solution of higher concentration to the electrode. This results in a concentration gradient away from electrode surface. In the bulk solution the concentration of analyte is maintained at a value of c₀ by convective transfer.

This technique of chronoamperometry is, therefore, closely related to the Cottrell equation [3, 50], which is shown in Equation 2. It defines the current-time dependence for linear diffusion control at a planar electrode. In the Cottrell equation, the current I is dependent on F Faraday's constant, n the number of transferred electrons per molecule, A the electrode area, c₀ the analyte concentration, D the diffusion coefficient and time t. From the Cottrell equation the current depends on the rate at which the analyte diffuses to the electrode.

(2)I=nFAc0Dπt

Figure 5 shows complementary chronoamperometric data on the same system analyzed by CV in Figure 4 [47]. Often used in addition to other techniques, such as cyclic voltammetry, for time-dependent system characterization, chronoamperometry is a sensitive technique which does not require the labelling of reactants. One interesting application of chronoamperometry is the real-time monitoring of electrochemical neurotransmission by Michael et al. [51]. They provide a technical description and comparison of the temporal and chemical resolution of amperometry, fast-scan cyclic voltammetry and high-speed chronoamperometry in studies with single cells, brain slices and intact animals. In chronopotentiometry the potential is measured as a function of time in response to a constant or square-wave current. This technique was used by Martins et al. to investigate the adsorption of human serum albumin (HSA) to SAMs [52].

The first publication of electrochemical impedance spectroscopy dates to 1975 [53]. Through the application of a small sinusoidally varying potential U, one measures the resulting current response I [44, 54]. By varying the excitation frequency f of the applied potential over a range of frequencies, one can calculate the complex impedance, sum of the real and imaginary impedance components, of the system as a function of the frequency (i.e. angular frequency w). Therefore, EIS combines the analysis of both real and imaginary components of impedance, namely the electrical resistance and reactance, as shown in Equation 3 [49, 55].

(3)Z(jw)=U(jw)I(jw)=Zr(w)+jZi(w);w=2πf

EIS possesses the ability to study any intrinsic material property or specific processes that could influence the conductivity/resistivity or capacitivity of an electrochemical system. Therefore, EIS is a useful tool in the development and analysis of materials for biosensor transduction, such as the study of polymer degradation. For example, by coating conductive electrodes with insulating polymers and then introducing this to an enzymatic reaction, the direct or indirect degradation of the polymer coating can be studied. If during the reaction existing free ions are able to penetrate into the polymer, the insulating nature of the polymer would be compromised and result in modified impedance characteristics of the transducing element [56, 57]. A recent example of electrochemical EIS was its use to characterize the fabrication process of a hydrogen peroxide (HRP) biosensor [58].

For electrochemical sensing, impedance techniques are useful to monitor changes in electrical properties arising from biorecognition events at the surfaces of modified electrodes. For example, changes in the conductance of the electrode can be measured as a result of protein immobilization and antibody-antigen reactions on the electrode surface [44, 45, 49, 59]. To additionally measure changes in capacitance with EIS, sometimes referred to as Faradaic impedance spectroscopy, one scheme as proposed by Katz and Willner is to construct an array of interdigitated electrodes and to monitor antibody-antigen reactions in the gaps between the electrodes, as illustrated in Figure 6a. Binding events of complementary antibody-antigen components alter the electrical properties in the gap between two electrodes, where changes in gap conductivity correspond to changes in the real impedance component Z_r(w) and changes in the gap capacitance correspond to changes in the imaginary impedance component Z_i(w) [44].

Interesting is also the use of impedimetric sensing to detect the binding state of DNA. In an experiment reported by Hang et al. a microarray configuration of interdigitated electrodes was used [60]. This is illustrated in Figure 6b and shows their intent to demonstrate the potential of EIS for low-density DNA microarrays. Single-stranded DNA (ssDNA) was immobilized to a modified surface between the electrodes to act as the biorecognition element. The immobilized ssDNA is associated with counter cations, which initially support ionic conductivity. However, upon hybridization with the complementary DNA strand to the ssDNA there is a reduction in the density of these cations. This reduction further inhibits the free displacement of ions near the surface and leads to a corresponding increase in the overall electrical impedance between the interdigitated electrodes [16, 44, 60, 61, 62, 63].

Finally as a side note, Electrochemical Impedance Spectroscopy also shares the acronym EIS with the sensing system known as an Electrolyte-Insulator Semiconductor [64]. Interestingly and in relation to the topic of DNA biosensing, electrolyte-insulator-semiconductor systems have been used to measure changes in capacitive impedance as a result of DNA hybridization [65]. Detection occurs when the thickness of the dielectric layer increases as a result of target DNA hybridization at the electrode interface. Subsequently, the measured capacitance decreases according to the following equation, where C is the capacitance, ε is the dielectric constant in vacuum, ε₀ is the dielectric constant of the layer, A is the active area and d is the thickness of the dielectric layer: (4)C=ɛɛ0Ad

The FET is a type of transistor that uses an electric field to control the conductivity of a channel (i.e. a region depleted of charge carriers) between two electrodes (i.e. the source and drain) in a semiconducting material. Control of the conductivity is achieved by varying the electric field potential, relative to the source and drain electrode, at a third electrode, known as the gate. Depending on the configuration and doping of the semiconducting material, the presence of a sufficient positive or negative potential at the gate electrode would either attract charge carriers (e.g. electrons) or repel charge carriers in the conduction channel. This would either fill or empty the depletion region of charge carriers and thus form or deform the effective electrical dimensions of the conducting channel. This controls the conductance between the source and drain electrodes. In linear mode, when drain-to-source voltage is much less than the gate-to-source voltage, an FET operates much like a variable resistor to switch between conductive and non-conductive states. Alternatively in saturation mode, an FET operates as a constant-current source and is often used as a voltage amplifier. In this mode the level of constant current is determined by the gate-to-source voltage. FET devices are preferred for weak-signal and/or high impedance applications, hence their widespread use in the growing field of electrochemical biosensing [6, 66].

Although there are many different types of FET devices: MOSFET, JFET, MESFET, CHEMFET, etc., the current focus in biosensing applications is on ISFET (ion-selective field-effect transistor) and EnFET (enzyme field-effect transistor) devices [3, 6, 31, 67, 68]. The conversion of an FET into a sensing device normally involves the replacement of the metal gate electrode by a biochemically sensitive surface (e.g. an analyte-selective membrane or an ion-conductive solution), which is brought into contact with the analyte solution [69]. Also present in the analyte solution is a reference electrode, which completes the circuit via the gate-voltage bias [3, 11].

One of the most popular methods for the construction of FET-based biosensing devices is the immobilization of enzymes at the gate surface of pH-sensitive ISFET devices, creating an EnFET. The method of enzyme-immobilization is a critical aspect of the device's performance and sensitivity [11, 70]. Biocatalytic reactions influence the presence of accumulated charge carriers at the gate surface in proportion to the original analyte concentration. This produces an electrical signal in the form of a measurable drain current [67, 71]. The majority of current EnFET applications involve the analysis of penicillin, glucose and urea [3], but more applications are emerging for the analysis of an ever-expanding variety of analytes [68]. Figure 7 illustrates the sensing principle of a penicillin-sensitive EnFET.

Other than a few ISFET-based devices, mainly for pH testing, FET-based biosensors have not yet been successfully commercialized [72, 73, 74, 75]. Some of the typical problems include: the dependence of the resulting signal on the method of enzyme immobilization, the speed of response and the recovery times between measurements, the difficulty of establishing standard methods of enzyme-layer deposition and surface patterning techniques compatible with silicon-based device fabrication techniques [68]. However, research is continuing to overcome these problems. Many designs have been suggested for EnFET biosensors [76, 77, 78, 79, 80, 81, 82, 83], as well as other forms of FET-based biosensors (e.g. ImmunoFET, GenFET, cell-based FET, etc.), which have already been generally labelled as BioFET devices, or simply BioFETs [68]. Despite the mentioned problems, they remain strong candidates for reusable multiple-analyte sensing arrays or disposable electronic biosensors due to such factors as the ever-shrinking size of FET devices, their low-power operation, their high-density fabrication methods, and their interconnectivity with micro/nanoelectronic elements [34].

Another alternative in FET-based biosensing, which has generated an enormous amount of interest in the past years, is the use of nano-objects, such as nanowires or carbon nanotubes, in combination with FET technology. The use of nanowires in sensing applications is reviewed in detail in Section 2.2. Perhaps the most well-known and controversial work on the subject of nano-object based FET biosensing was presented by Cui et al. [85, 86]. In such sensing devices, the source and drain of a planar FET device is bridged by a nano-object. Binding to the surface of these nano-objects alters their ability to conduct, which serves as the detection mechanism. Although not functionalized with a biorecognition element, Figure 8 provides an illustration of a silicon nanowire FET as presented by Koo et al. [84]. Recently Gao et al. have built silicon nanowire arrays for the label-free detection of DNA [87]. A summarized schematic representation of their sensing principle and their measurements are shown in Figure 9.

In SPR configurations the surface between two mediums of differing refractive indexes is coated with a thin film of conducting material, which is often a noble metal, such as Ag, Au or Cu. If this thin film is irradiated by light of a certain wavelength at a specific incident angle, termed the SPR angle, the evanescent field wave from Total Internal Reflection (TIR) can excite an electron wave, known as a surface plasmon, which propagates along the metal surface [105, 106]. The coupled electromagnetic evanescent field extends typically ∼200 nm from the conductive surface (e.g. Au excited at red wavelengths) as its intensity decays exponentially. Therefore, the sensing mechanism of SPR spectroscopy is based on the measurement of small changes in the refractive index in the vicinity of the surface of a conductive thin film [105, 107, 108, 109].

For use as a biosensor, the changes in refractive index in the vicinity of the conductive thin film surface (i.e. within the evanescent field) can be induced by the presence of biomolecules. The surface concentration or mass coverage of which can then be calculated using the de Feijter formula [111, 112]. SPR biosensors can be used in the determination of a number of surface binding interactions, such as: small molecule adsorption [109], protein adsorption on self-assembled monolayers, antibody-antigen binding, DNA and RNA hybridization [113, 114], protein-DNA interactions [106], binding kinetics, affinity constants, equilibrium constants, as well as receptor-ligand interactions in immunosensing [105, 115] and many more.

In EC-SPR, the combination of electrochemistry and SPR, the thin metal film on the substrate serves not only to excite surface plasmons, but also acts as a working electrode for electrochemical detection or control [117]. This important dual role of the thin metal film can be seen in a typical EC-SPR schematic in Figure 14. One advantage of the EC-SPR configuration is the ability to simultaneously obtain information about the electrochemical and optical properties of films with thicknesses in the nanometer range. EC-SPR has often been applied to study the formation and the properties of thin films and mono-/multi-biolayers using, for example, self-assembly or electro-polymerization methods [110, 118, 119]. One can also use an electrochemical configuration in combination with SPR to optically measure e.g. reaction kinetics of biomolecules in the presence of electric fields. Such experiments were performed by Georgiadis et al. and Heaton et al. to monitor the in situ hybridization of DNA in the presence of different electrochemical fields [114, 120, 121]. A further and recent use of EC-SPR was the characterization of multilayer assembly processes on gold nanoparticles [122]. Kang et al. used EC-SPR to develop a biosensor to detect the enzymatic reaction between horseradish peroxidase and a film of polyaniline (PAn) on a gold electrode in the presence of H₂O₂. The combination of SPR and electrochemical oxidation/reduction enabled them to demonstrate large changes in SPR responses as a result of the transition in conductivity of a PAn film on a gold electrode surface. Their results are summarized in Figure 15.

Waveguide based optical evanescent sensors have equivalent sensitivity to small numbers of adsorbed analytes compared to most other biosensors. However, because of the need for both, an optically transparent substrate and interface (electrode), fewer combinations of waveguide spectroscopy and electrochemistry to SPR and electrochemistry have been made, since optically transparent electrodes are rare. Optical Waveguide Lightmode Spectroscopy(OWLS) is a label-free technique that allows for in situ measuring of adsorption, desorption, adhesion and biospecific binding processes [112, 123]. This technique is based on linearly polarized laser light that is coupled into a waveguide at two well-defined incident angles. These incoupling angles are sensitive to changes in the refractive index within the evanescent field above the surface of the waveguide. Monitoring of the changes in the incoupling angles enables determination of the adsorbed mass and the number of adsorbed molecules [123] using de Feijter's formula [111, 112].

Electrochemical OWLS (EC-OWLS) combines evanescent-field optical sensing with the electrochemical control of surface adsorption processes. It gives the possibility to directly observe mass adsorption as a function of an applied potential [124, 125]. The electrodes are connected according to the scheme shown in Figure 16. A transparent conductive electrode is made by coating the waveguide with a thin film of Indium Tin Oxide (ITO).

An example of such a combined experiment is given in Figure 17. First cyclic voltammetry was performed in order to get stable electrochemical conditions. Then a three step adsorption was performed. Poly(L-lysine)-grafted-poly(ethylene glycol) with 50 % of the side chains conjugated with biotin was adsorbed first, followed by neutravidin and biotin conjugated GOx. Subsequently, a potential of 1 Volt was applied several times in order to measure the current generated by the enzymes at different substrate concentrations.

Classic ellipsometry is focused on the external reflection of a light beam from a reflective surface, where linearly-polarized light of a known orientation is reflected as elliptically-polarized light. From the changes in the ellipsometric angles (Ψ, Δ), the optical properties, thickness, morphology or roughness of layers or films on the surface can be calculated [6, 102, 103] and used to determine e.g. the amount of adsorbed protein on a surface [103, 104].

Since ellipsometry can be performed on most reflective substrates it can be easily conducted on an electrode surface and combined with electrochemistry. Ying et al. have used ellipsometry and electrochemical methods to study protein adsorption on metal surfaces and specifically human serum albumin on gold surfaces [126]. Chronoamperometry and ellipsometry were combined for the study of immunosensor interfaces based on methods of Immunoglobulin G adsorption onto mixed self-assembled monolayers [127]. Yu et al. have combined imaging ellipsometry with electrochemistry to investigate the influence of electrostatic interaction on fibrinogen adsorption on gold surfaces [128].

The Quartz Crystal Microbalance with Dissipation monitoring measures changes in the frequency f and dissipation factor D of an oscillating quartz crystal upon adsorption of a viscoelastic layer [129]. The oscillation is based on the piezoelectric effect and the crystals typically have a fundamental resonance frequency of 5 MHz, which decreases upon mass adsorption. Modern versions of QCM, like the QCM-D, are insensitive to potentials applied at the solution interface and to local changes in ionic strength and can thus be readily combined with electrochemical sensing.

The measured mass includes hydrodynamically coupled water such as water associated with the hydration layer of e.g. proteins and/or water moved in cavities or by rough features in the film [130, 131]. The dissipation factor is a measure of energy loss and increases with e.g. higher viscosity of the solution or of adlayers. The main advantages compared to other common methods for measuring biomolecule adsorption, such as ellipsometry or surface plasmon resonance, are the flexibility to use virtually any surface coating and the information about the viscoelastic properties of the adlayer obtained through simultaneous measurements of both f and D. This includes any electrode material of interest. As with e.g. SPR or ellipsometry no labelling is required [129].

By using a conductive surface coating or the working electrode of the crystal itself, QCM allows for simultaneous electrochemical experiments [129, 132]. For electrochemical QCM a specially designed flowcell is required; typically the QCM-crystal acts as the working electrode, Ag/AgCl on the upper side of the flowcell as the reference electrode and a platinum wire in the outlet tube as the counter electrode [129, 133]. The same setup is also used for adsorption steps that are sensitive to an applied potential [129].

The main applications to date have been in the fields of polyelectrolyte multilayers [134], corrosion [135], enzymatic and nucleic acid based biosensors [129].

The invention of the scanning tunnelling microscope (STM) [136, 137] in the early 1980's has inspired various forms of SPM techniques. Despite the now diverse palette of SPM techniques, of which only the most common will be reviewed here, STM remains a promising technique for experiments combined with electrochemistry, such as Bae et al. [138].

Perhaps the most commonly employed techniques of SPM for imaging and characterization is Atomic Force Microscopy (AFM) [139]. AFM devices have evolved into versatile and powerful tools, which enable the investigation and imaging of surfaces with even molecular resolution. This is accomplished by monitoring the interaction force between the sample surface and a probe (e.g. a sharp tip), which is attached to the end of a force-sensing cantilever. The details and theory of AFM operation are not explained here, but can be read in a number of other sources [140, 141, 142]. Although AFM is often used ex situ in a separate characterization step before and/or after an electrochemical analysis [143, 144, 145, 146], the focus here will be placed on AFM's simultaneous combination with electrochemistry: EC-AFM. In other words, the imaging of the sample is performed simultaneously with the electrochemical modifications.

One area of investigation with generic EC-AFM methods is the study of electrode surface modification or performance under certain electrochemical conditions [110, 147, 148], which is important in application-specific electrode development for biosensors and chemosensors. Although not biosensing, but in the related field of biomechanics, Eliaz et al. have performed ex situ and in situ imaging, potentiostatic and potentiodynamic measurements by means of EC-AFM. Their studies were aimed at contributing to the improved osseointegration of titanium-based alloys in orthopedic devices and implants [149].

The development of ultramicroelectrodes (UMEs) and their combination with the SPM also paved the way for an SPM technique known as scanning electrochemical microscopy (SECM) [150, 151, 152]. SECM is used to locally investigate the electrochemical activity and/or the topography of a surface. Designed for either amperometric or potentiometric measurements, a UME probe, which is normally in the shape of a tip, is scanned along the surface of interest and is used to induce chemical changes and to collect electrochemical information. Two recent reviews on SECM from Sun et al. and Edwards et al. provide a very comprehensive overview of its general application and theory, as well as its use in current research [153, 154]. In short, SECM has proven useful in the detection and imaging of heterogeneous electron transfer reactions at metal/solution interfaces, electron transfer kinetics at solid/liquid interfaces, various electrocatalytic reactions, predominately the hydrogen oxidation reaction (HOR) and the oxygen reduction reaction (ORR), lateral charge and mass transfers and biological cells [153, 155, 156].

Numerous hybrid techniques have evolved around SECM [153], including its combination with surface plasmon resonance (SPR-SECM) [158], electrochemical quartz crystal microbalance (EC-QCM-SECM) [159, 160, 161], fluorescence spectroscopy (FS-SECM) [162], photoelectrochemical microscopy (PEM) or near-field scanning optical microscopy (NSOM-SECM) [163], electrogenerated chemiluminescence (ECL-SECM) [164] and atomic force microscopy (AFM-SECM) [153, 154, 165]. The most widely used of these hybrid techniques would be AFM-SECM. The general concept of the hybrid probe used in AFM-SECM measurements is illustrated in Figure 18.

AFM-SECM is capable of obtaining simultaneous electrochemical and topographical information at high spatial resolution through the integration of an electrode into the AFM probe. In other words, the structural information is simultaneously correlated with chemical activity of the substrate [167]. These hybrid probes are, therefore, quite effective for the identification and chemical mapping of active surface sites [168]. Critical to the resulting synergy of AFM-SECM is the proper modification of the probe. Please refer to the reference literature for information on the fabrication and application of various types of AFM-SECM probes [153, 154, 168]. Gardner et al. have studied the correlation of membrane structure and transport activity across a synthetic polyethylene terephthalate (PET) membrane [169].

Figure 19 shows the topographical imaging of the morphology of model neurons directly in the cell growth media without a redox mediator as performed by Kurulugama et al. [166]. Kreung et al. have mapped glucose oxidase (GOx) activity on a micropatterned surface and through a synthetic membrane [157, 170]. The same group has imaged immobilized horseradish peroxidase with the same techniques [171]. Simultaneous imaging topography and the methodology of using electrochemistry to agitate interfacial systems has been applied to crystal dissolution studies [172].

Among other recognition elements, antibodies are most widely used because of the high specificity of the antibody-antigen binding. However, they are also very sensitive to their environment. In order to detect antigens, antibodies are surface-immobilized, which can cause a severe loss of their biological activity. Since the activity of surface immobilized antibodies depends on their orientation, it is advantageous to assure that they are not randomly oriented. Other reasons for inactivity include steric hindrance in cases where the density is too high as well as denaturation due to non-specific interactions with the surface [216]. Immobilization techniques include microcontact printing [217], biotin-streptavidin binding [218, 219, 220, 221], direct spotting [218], adsorption to a conductive polymer matrix such as polypyrrole or polyaniline [214] as well as covalent binding [222].

Most commonly used is Immunoglobulin G (IgG) which, like typical antibodies, consists of one Fc and tow Fab' binding sites. It becomes inactive when the Fab' fragment binds to the surface [216]. Often antibodies help to discover new disease markers [223]. In typical applications, monoclonal antibodies are used as specific capture antibodies for e.g. prostate cancer marker (PSA). In such, so called sandwich assays, one antibody acts as a capture antibody while the second enzyme-labelled antibody is bound to the PSA [224].

Electrochemical enzymatic biosensors can be built up similar to the enzyme-linked immunosorbent assay (ELISA). After immobilizing antibodies to a surface, an analyte is introduced to which the antibodies bind specifically. In the most common detection scheme a secondary labelled antibody then binds to the analyte in order to detect its concentration (see Figure 21d). The detection antibodies are coupled to an enzyme, which allows quantitative measurements of the amount of bound antigens by monitoring the electrical signal generated by an enzymatic reaction [225].

A wide variety of antigens, usually in diagnostics, can be detected this way. It is e.g. possible to diagnose Hepatitis C already in an early stage [226]. More recently, capture antigens have also been used to detect cancer markers. Among the most important ones are carcinoembryonic antigen (CEA) [193], carbohydrate antigen 19-9 (CA19-9), carcinoma antigen 125 (CA125), alpha-fetoprotein (AFP), CA15-3, human chorionic gonadotropin (hCG) [192] and prostate specific antigen (PSA) [224], which are used to detect various types of cancer.

Nowadays, antibody fragments are emerging as credible alternatives to antibodies. They provide the same specificity as whole antibodies. Furthermore, they are smaller which can be an advantage for biosensor applications. In their main application field, biosensors and therapeutics, these small, highly specific reagents against target antigens are often used as bi- or trimers [227].

Most commonly used are IgG fragments [216, 217]. Such small, recombinant antibody fragments can be linked to other molecules, for example lipids, drugs and proteins [227]. An antibody that is naturally composed of only heavy chains is the camel antibody. Even though the light chains are missing, it binds very specifically [228]. For other applications biotinylated fragments are adsorbed to streptavidin coated surfaces [219]. As already mentioned for antibodies, the orientation is of great importance. Lu et al. [216] reported a threefold increase of activation upon proper/controlled orientation. In a more complex approach Vikholm et al. adsorbed Fab' fragments directly to gold but embedded them in a protein repellant polymer such that only the antigen binding site stuck out of the polymer [229].

Aptamers are folded single stranded DNA or RNA oligonucleotide sequences with the capacity to recognize various target molecules. They are generated in the SELEX (systematic evolution of ligands by exponential enrichment) process which was first reported by Ellington [230] and Tuerk [231] in 1990. In this approach suitable binding sequences are first isolated from large oligonucleotide libraries and subsequently amplified [6, 232].

The main application for aptamers is in biosensors. While antibodies are used in ELISA, the similar process for aptamers is called ELONA (enzyme linked oligonucleotide assay) [233]. They have many advantages over antibodies such as easier deposition on sensing surfaces, higher reproducibility, longer shelf life, easier regeneration and a higher resistance to denaturation. As antibodies, they are characterized by both, their high affinity and specificity to their targets [6, 232].

These microcapsules were first presented by Caruso and Donath in 1998 [234, 235]. They are fabricated by coating a microparticle with alternating layers of oppositely charged polyelectrolytes, such as the negatively charged poly(sodium styrenesulfonate) (PSS) and the positively charged poly(allylamine hydrochloride) (PAH). The shell thickness can be varied by adjusting the number of layers. Subsequently, the core is dissolved and the material diffuses through the shell into the surrounding solution. The capsules usually have a diameter of 1-10 micrometers [235, 236]. Optionally, coating of poly(L-lysine)-grafted-poly(ethylene glycol) can be applied to make them protein resistant [237].

For applications towards electrochemical biosensing, enzymes can be incorporated into PEM capsules. This could be accomplished by adsorbing the enzymes inside the pores of the CaCO₃ microparticles where their stability and catalytic activity was successfully maintained [238, 239].

Liposomes or lipid vesicles consist of bilayer forming amphiphilic molecules such as palmitoyloleoylphosphocholine (POPC), called lipids. Liposome sizes vary between 20 nm and several hundred micrometers. Applications are in e.g. drug delivery, cosmetic emulsions, and optical electrochemical biosensing. In the latter case they are loaded with enzymes. Many different encapsulation methods have been developed during the last decade. Among them, the extrusion technique and dehydration-rehydration have been the most successful ones. The entrapped enzymes catalyze the conversion of substrate molecules into products. Having stable vesicles and enzymes, as well as low enzyme permeability and high substrate permeability are prerequisites for their use in biosensing [240]. However, substrate permeability is typically low, which drastically limits the turnover rate of encapsulated enzymes. To overcome this drawback, membrane channels such as the outer membrane protein F (OmpF) from Escherichia coli were incorporated into the membrane [241]. They allow for diffusion of molecules with a molecular weight up to 400 Da [242]. Hill et al. incorporated glucose oxidase, horseradish per-oxidase and the combination of both into vesicles. There was no major loss in enzymatic activity [243]. In another application the enzyme beta-lactamase was incorporated into POPC vesicles [242].

Polymeric micelles are built from amphiphilic block copolymers. The inner, hydrophobic side usually consists of poly(acrylic acid) (PAA) or polyesters; for the outer, hydrophilic shell poly(ethylene glycol) (PEG) is most widely used. Their size can vary between 10 and 100 nm [244, 245, 246]. Thurmond et al. cross-linked the shells to increase stability [247]. Even though the main application is in drug delivery where drugs are enclosed, enzymes have also successfully been embedded in the core through electrostatic interaction with polyion complex (PIC) block copolymers [248]. Another way is to PEGy-late the enzymes, whereas the PEG is part of the PIC [249]. This allows for potential use in enzymatic biosensing.

Since encapsulated enzymes do not need to be mobile, hydrogels offer a suitable immobilization technique. Thus, cross-linked polymers form a biocompatible hydrogel that entraps the enzymes. The water soluble redox polymer should not cross-link too much in order to allow swelling in water. At the same time it mediates electron transfer between the enzymes and the electrode via electron diffusion through electron exchange between mobile polymer segments. In applications, such as electronic biosensors, it is important that the entrapped enzymes are not denaturated and remain in their active state. Otherwise, the sensitivity and selectivity of the biosensor is decreased drastically [250, 251, 252, 253]. Mao et al. introduced a new redox hydrogel, consisting of a polymer with 13-atom-long tethers between the redox centers and the polymer backbone. This led to an electron diffusion coefficient one order of magnitude higher than for previously known systems [254].

Like hydrogels, the sol-gel technique is applied directly on surfaces. In the first experiments porous SiO₂ glass was used. Braun et al. were the first ones to report about successful enzyme entrapment in a sol-gel matrix [255]. The material is formed via partial or complete hydrolysis of a suitable precursor, usually silanes, mixed with biomolecules. Subsequently, gelation and aging follows. Aging strengthens the network and the drying process simultaneously leads to shrinkage [256, 257]. Tunable pore sizes and pore size distribution, optical transparency, no need of a mediator, mechanical rigidity, chemical inertness, permeability and thermal stability gives them highly desirable properties [257, 258]. Typical applications include selective coatings for optical and electrochemical biosensors [257], for example carbon ceramic composite electrodes (CCE) for amperometric biosensors [258].