The electrochemical oxidation of the adenine moiety in different adenine derivatives, including adenosine, adenosinemonophosphate (AMP), adenosinediphosphate (ADP) and adenosinetriphosphate (ATP), has been studied at pyrolytic graphite electrodes in aqueous solution in the pH range 5–11. All these compounds yield strongly adsorbed electroactive oxidation products with reversible behavior and formal potentials close to 0 V at pH 9.0. These oxidation products show strong electrocatalytic activity for the oxidation of the reduced form of β-nicotinamide adenine dinucleotide (NADH). The electrochemical properties of the electrocatalyst obtained by oxidation of the corresponding parent nucleotide were investigated by cyclic voltammetry. The influence of the structure of the parent compound on the electrocatalytic activity of the corresponding product toward NADH oxidation was studied also. The kinetics of the electrocatalytic reactions were evaluated from rotating disk electrode measurements. These modified graphite electrodes can be used as NADH transducers in biosensors for substrates of dehydrogenase enzymes and also for the measurement of enzymatic activity.

A biosensor based on modified carbon paste electrodes for the amperometric detection of d-sorbitol is described. Carbon paste electrodes were modified by d-sorbitol dehydrogenase (SDH) and nicotinamide adenine dinucleotide (NAD+) and coated with a non-conducting poly(o-phenylenediamine) (PPD) film. After partial oxidation of the immobilised NAD+ by applying a potential step from 0.0 to 1.2V (SCE), the modified electrode allowed the amperometric detection of the NADH enzymatically obtained at an applied potential of 0.0V. The resulting biosensor responded rapidly to sorbitol up to 8×10−4M with a detection limit of 4×10−5M.

An enzymatic biosensor for hydrogen peroxide based on a horseradish peroxidase (HRP)-ferrocene carbon paste modified electrode and coated with a layer of electrochemically generated poly(o-aminophenol) is reported. A linear calibration curve is obtained over the range 1×10−8M to 1×10−5M. The biosensor responds to hydrogen peroxide in a few seconds and has a detection limit of 8.5×10−9M. The response of the biosensor is diffusion controlled at low substrate concentrations. Flow injection assays of hydrogen peroxide at a sampling rate of 150 injections per hour with a relative standard deviation of 0.8% (50 samples) are possible. Applicability of the sensor for measurement of hydrogen peroxide in real samples (milk) was demonstrated.

Poly-(o-phenylenediamine was electropolymerized on the surface of an alcohol dehydrogenase-nicotimanide adenine dinucleotide-modified carbon paste electrode to yield an electrochemical biosensor for ethanol. Efficient electrocatalytic oxidation, at low applied potentials, of the enzymatically produced NADH by the conducting film provided a current that was dependent on ethanol concentration. The influence of various experimental variables was explored to optimize the analytical performance. Linear responses for ethanol in the range 3 × 10−8 − 3 × 10−6 M and a response time of 20 s were obtained. The trend in sensitivity towards different alcohols is in agreement with the known biospecificity of yeast ADH. Ethanol sensor was applied to the analysis of different alcohol beverages, resulting in excellent accuracy and precision.

A sensitive immobilized enzyme electrode for l-lactate is described. Toluidine blue O (TBO) and lactate dehydrogenase (LDH) were co-immobilized onto the surface of a graphite electrode by means of a dialysis membrane. l-lactate is oxidized to pyruvate by LDH in the presence of nicotinamide adenine dinucleotide NAD+ which is then reduced to NADH. The NADH formed is electrocatalytically oxidized by TBO and the reduced mediator is detected at 0 V vs. a AgAgCl reference electrode. Different dialysis membranes were tested in order to improve stability. The most stable enzyme electrode was fabricated with a benzoylated membrane, showing a half-life of eight days. This amperometric electrode responds fast and linearly to l-lactate in a concentration range 10−6 − 6 × 10−5 M. The limit of detection is 4 × 10−7M. The enzyme electrode was applied to the determination of l-lactate in food samples. Results were compared to those obtained using a spectrophotometric method, showing a good agreement.

A reagentless uric acid selective biosensor constructed by immobilising uricase and horseradish peroxidase (HRP) in carbon paste without the addition of an electron transfer mediator is described. The response of the electrode is based on the enzymatic reduction of hydrogen peroxide in the presence of uric acid. Uricase and HRP were dispersed in the carbon paste and the optimum paste mixture was determined. Poly(o-aminophenol) was electropolymerised at the working surface area of the electrode acting as a conducting polymer layer. Cyclic voltammetry was used to characterise the permselective characteristics of the polymer layer. At an applied potential of 50 mV vs. Ag/AgCl, a linear response was obtained up to 1 x 10(-4) M, with a limit of detection of 3 x 10(-6) M. The sensor had a response time of 37 s. a calibration precision of 2.2% (n = 4) and an estimated sample frequency of 20 h(-1). Responses to the analyte of interest were pH dependent. The sensor was incorporated into a flow injection system for the qualification of uric acid in human serum. Results compared favourably with a standard spectrophotometric method.

A reversed-phase liquid chromatographic method involving precolumn derivatization with 9-fluorenyl-methylchloroformate (FMOC) has been developed for the determination of dimethylamine (DMA) in groundwater. FMOC reacts rapidly with DMA under mild conditions, and the derivative is separated from matrix components and FMOC degradation products on a C18 column using isocratic elution with a mobile phase of 0.03 M acetate buffer (pH 4.0)-acetonitrile (30:70, v/v). Mean recovery was 98.5±4% and a detection limit of 4.5·10−7 M, corresponding to an injected amount of 3.6 pmol, was estimated using fluorimetric detection with excitation and emission wavelengths of 265 and 310 nm, respectively. No matrix interferences were found even when highly coloured and/or cloudy samples were examined. Several mixtures containing eleven amines, diamines and amino acids can be resolved using this method.

A detailed study of the electrochemistry of clenbuterol at bare carbon-paste electrodes (CPEs) has been carried out. Results showed that clenbuterol undergoes an ECE process. This compound is irreversibly oxidised at high potentials, resulting in the formation of a product which demonstrates quasi-reversible electrochemical behaviour at less positive potentials. The amount of this chemical product formed is very pH-dependent. Investigations into the electrochemical behaviour of clenbuterol at Nafion-modified CPEs were also made. The use of a thin Nafion film cast over the CPE resulted in a large increase in peak current over bare electrodes. Linear accumulation occurred with time, the linear range increasing with decreasing concentration. This allowed the detection of low concentrations of clenbuterol. Diffusion proved to be the rate-controlling process of clenbuterol through the Nafion membrane.

An electrochemical study of three important beta-agonist drugs at unmodified and Nafion-modified carbon paste electrodes was carried out. All the compounds are oxidized irreversibly st high positive potentials at a bare carbon paste electrode, giving rise to sharp, well-defined peaks. A secondary oxidation process was observed at pH values above 6.0. The rate-determining step was investigated for each compound at two concentration levels. Electrochemical activation procedures were optimized to ensure reproducible signals for the construction of calibration graphs. The modification of the carbon paste surface with a Nafion film allowed a preconcentration process to take place for all compounds, such that higher sensitivities were achieved compared with the bare surface. Such modifications resulted in limits of detection for the compounds down to 2.5 X 10(-8) mol dm(-3). Optimum accumulation was obtained at low pH values (2-3). Cyclic voltammetry at two Nafion film thicknesses demonstrated that a diffusion-controlled process exists within the Nafion layer. Salbutamol showed a higher affinity for Nafion than the other two compounds, both in terms of longer linear accumulation and linear accumulation at higher concentrations.

The preconcentration of fenoterol on a Nafion-modified carbon paste electrode and its subsequent determination using differential pulse voltammetry is described. The effect of pH and percentage Nafion concentration on the accumulation behaviour of fenoterol was studied, and accumulation curves, calibration graphs and reproducibility studies at two different Nafion concentrations have been carried out in the range 2.5 × 10−8−5.0 × 10−7 M fenoterol. A limit of detection in aqueous solutions, calculated using a signal-to-noise ratio (S/N) of 3, was 9.0 × 10−9 M. Application of the electrode to pharmaceutical preparations, without sample pretreatment, resulted in acceptable deviation from the stated concentration (RSD = ± 3.81%, n = 4). For more complex matrices, a suitable extraction procedure was developed, resulting in recoveries of >90% (urine) and >75% (serum).

The adsorptive stripping voltammetric behaviour of folic acid and riboflavin was studied at a hanging mercury drop electrode by phase-selective a.c. voltammetry. In 0.1 M sodium acetate buffer (pH 5.0) a cathodic scan gave peaks at −0.29 and −0.55 V vs. Ag/AgCl for riboflavin and folic acid, respectively. The adsorptive stripping response was evaluated with respect to concentration dependence and preconcentration time. Both compounds can be simultaneous determined with a relative standard deviation of 1.4% at 5 × 10−8 M riboflavin and 0.74% at 10−8 M folic acid. The method compared favourably with liquid chromatography with UV detection and was succesfully applied to the simultaneous determination of both compounds in multivitamin preparations. The average contents of riboflavin and folic acid were found to be 15.19 mg ± 2.1% and 1.8 mg ± 2.6%, respectively.

The conditions for the electrodeposition of mercury on the surface of carbon fibres to permit the sensitive and reproducible voltammetry of folic acid and mitoxantrone (MXT) were investigated. The accumulation behaviour of both molecules on a mercury film ultramicroelectrode was studied and the possibility of carrying out a.c. adsorptive stripping analysis using mercury-coated carbon ultramicroelectrodes was demonstrated. Optimum conditions for both the deposition of mercury and a.c. monitoring are described. This procedure gave rise to a linear calibration graph from 5.0 x 10−10 to 2.0 x 10−8 M, a detection limit of 5.0 x 10−10 M and a relative standard deviation (R.S.D.) of 5.05% at 5.0 x 10−9 M (n = 10 ) for the determination of MXT. For folic acid a linear calibration graph from 1.0 x 10−9 to 5.0 x 10−8 M, a detection limit of 9.0 x 10−10 M and an R.S.D. of 1.44% at 4.0 x 10−8 M (n = 10) were found. Both compounds can be determined directly in biological samples at physiological levels without any separation or clean-up procedure.

Aminopterin was studied as a model compound for its analogues which maintain the pteridine ring in their structure. Its adsorptive behaviour on mercury was studied and the DC adsorptive stripping and phase-selective AC adsorptive stripping conditions were optimized. 10-Edam, an aminopterin analogue, was studied and shown to behave similarly to aminopterin. Phase-selective AC voltammetry provided the best signal and gave a detection limit of 4 x 10(-12) M aminopterin in aqueous solution employing an accumulation time of 10 min. The optimized method was applied to the analysis of both aminopterin and 10-Edam respectively in human serum. After extraction with a C18 reversed-phase cartridge the detection limit of the method was 1 x 10(-8) M aminopterin and the overall assay percentage recovery was 73.5% (n = 5) at a concentration of 5 X 10(-7) M aminopterin in serum. The analysis of 10-Edam at the same concentration in serum yielded the higher percentage recovery of 94.46% (n = 5) following the same procedure.

The conditions for the deposition of mercury on the surface of carbon fibers to permit the sensitive and reproducible voltammetry of Se(IV) have been investigated. Several procedures employing both predeposition and in situ formation of a mercury film on the carbon fiber were evaluated, and the effect of the mercury concentration and plating time upon the sensitivity of the stripping response for Se(IV) was determined. Due to the steady-state diffusional mass transport characteristics exhibited by these electrodes, the need for stirred solutions, a specific deposition potential, and a deposition time period in the stripping voltammetric procedure was eliminated. This procedure gave rise to a linear response up to 3 ng/ml, a detection limit of 0.11 ng/ml, and a relative standard deviation (RSD) of 4.4% at 5 ng/ml (n = 5) for determination of Se(IV).

The electrooxidation of mitoxantrone (MXT) on carbon paste electrodes was studied using voltammetric techniques in adsorption conditions. The analyte was accumulated at the working electrode (a carbon paste electrode) under precisely controlled mass-transport conditions and stripped electrochemically in the same solution. An electrode pretreatment is proposed which shows good reproducibility of the analytical signal (0.81%). The stripping step was studied with alternating current voltammetry providing a linear response in the concentration range 5 x 10(-11) to 7 x 10(-10)M in aqueous samples. Finally, a method using the medium exchange and AC phase-selective adsorptive stripping voltammetry technique was proposed for MXT analysis in urine samples.

The electrodeposition of mercury thin films onto carbon fibres for the determination of aminopterin and its analogues has been optimized following an investigation of the electrochemical reduction processes of aminopterin obtained at a static mercury drop electrode. The advantageous characteristics of ultramicroelectrodes combined with adsorptive preconcentration and phase-selective a.c. stripping voltammetry were found to yield a very sensitive and reproducible method. By using this electrode, accumulation was performed at five different concentrations of aminopterin ranging from 5 x 10(-10) to 5 x 10(-8) mol dm-3. The electrode yielded a calibration graph from 2 x 10(-10) to 8 x 10(-9) mol dm-3 (r = 0.994) with a limit of detection [signal-to-noise ratio (S/N) = 31 of 1 x 10(-10) mol dm-3 aminopterin in aqueous solutions. The reproducibility of the signal was evaluated at three different concentrations of aminopterin producing relative standard deviations ranging from 3.57% at the 5 x 10(-10) mol dm-3 level to 2.49% at the 1 X 10(-8) mol dm-3 level (n = 10). The electrode was applied to the determination of aminopterin in urine resulting in a limit of detection (S/N = 3) of 2.5 x 10(-7) mol dm-3 without the employment of any pre-treatment of the urine.

The electrochemical behaviour of mitoxantrone (MXT), an important antineoplastic agent, has been studied at mercury electrodes. The nature of the process taking place at the hanging mercury drop electrode (HMDE) was clarified. The electrochemical behaviour observed was in close agreement with theoretical predictions for an adsorbed molecule which is reversibly reduced. Both the molecule and its reduced product appeared to be adsorbed at the surface of the electrode. Adsorptive stripping voltammetry has been proven to be advantageous over any other assay technique, allowing 5 x 10(-11)M MXT to be detected. The interference arising from surfactants competing for the adsorption sites at the electrode have been studied and the possibility of MXT determination in dilute urine samples has been shown. Some interesting data, such as the MXT adsorbing surface area and the kinetic constant of the associated coupled chemistry reaction were also determined.

A flow cell incorporating a stationary carbon paste electrode coupled with a.c. voltammetry was used. Preconcentration and determination of the cancer chemotherapy drug mitoxantrone (MXT) were achieved via a flow-injection approach utilizing adsorption of the drug on the electrode, followed by medium exchange and a.c. voltammetry on the adsorbed surface. A linear response was obtained in the concentration range 5 × 10−9–1.5 × 10−7 M with aqueous samples. The method was used for the determination of MXT in urine without any sample pretreatment; the relative standard deviation obtained was 1.1% (n = 5) with a concentration of 1 × 10−6 M. The proposed method is more selective than other methods suggested for the determination of MXT in urine and plasma samples, which usually require some kind of sample pretreatment.

Mitoxantrone was determined by flow injection analysis using a flow cell modified in the laboratory and fitted with carbon paste as an amperometric detector. The sample solution (100 μl, 5 × 10−8−1 × 10−5 M) was injected into the carrier stream of 0.1 M perchloric acid (pH 1.12). Mitoxantrone was determined by oxidation at the carbon paste electrode (CPE) at +0.90 V. A 60-cm delay coil (0.5 mm i.d.) was incorporated just before the detector (a canal thin layer) and a flow rate of about 4 ml min−1 was used. The system was successfully applied to the determination of mitoxantrone in a pharmaceutical preparation; the method was fast and reproducible.

Optimum conditions for the mercury electrodeposition onto glassy carbon electrodes and the optimum alternating current (AC) adsorptive stripping signals of folic acid (vitamin B-12) were found. The film was characterized by electron microscopy and the results obtained in the analysis of the vitamin were compared with those previously reported using conventional mercury electrodes (static or hanging). The mercury thin film electrode (MTFE) has shown more sensitive signals for folic acid with good performance and reproducibility. A detection limit down to 10(-12) M could be reached by using 300 s accumulation time.