Search

Photocatalytic degradation of the natural oestrogens 17β-oestradiol and oestriol and the synthetic oestrogen 17α-ethynyloestradiol in water were investigated. The reactions were carried out in a quartz coil reactor coated internally with titanium dioxide (Degussa P-25). Degradation by UV light alone was also investigated. Reactions were monitored using fluorescence spectroscopy. The effect of initial concentration and the effect of light intensity on the photocatalysis and photolysis of 17β-oestradiol in water were also investigated. The results showed that photocatalysis and photolysis are capable of degrading all three oestrogens in water. Photocatalysis is much more effective than UV light alone and all reactions follow pseudo first order kinetics. The rate was found to be proportional to the square root of light intensity for photocatalysis of 17β-oestradiol. The relationship between light intensity and the rate of degradation by photolysis is linear up to a point when it starts to level off.

The presence of low levels of natural and synthetic steroid estrogens in the aquatic environment, and their biological effects on aquatic organisms, are presently issues of concern. In this study, we investigated the temporal removal of estrogenic activity of several potent and environmentally relevant steroid estrogens by photocatalysis over an immobilised titanium dioxide (TiO2) catalyst. We used a recombinant yeast assay to measure estrogenic activity, which provided detection limits within the reactor of 53 ng/l for 17beta-estradiol and 17alpha-ethinylestradiol, and 100 ng/l for estrone. Pseudo-first-order kinetic data showed that photocatalysis over titanium dioxide was equally effective at removing the estrogenic activity of all three steroid substrates in aqueous solutions (initial concentrations of 10 microg/l) with a 50% reduction in estrogenicity within 10 min. In control experiments without TiO2 catalyst, the rate of UVA photolysis of the steroid substrates varied, but was most effective with 17alpha-ethinylestradiol followed by estrone, and was least effective with 17beta-estradiol (0.42, 0.2 and0.1 times the rate achieved with photocatalysis, respectively). The application of photocatalysis for the removal of steroid compounds within STW effluent released into the aquatic environment is discussed.

This contribution describes new composite carbon paste electrodes (CPEs) for the determination of phenolic compounds. The composite CPEs were prepared by in situ generation of polypyrrole (PPy) within a paste containing the enzyme polyphenol oxidase (PPO). The best paste composition (enzyme/pyrrole monomer/carbon particles/Nujol) was determined for a model enzyme, glucose oxidase (GOx) according to the enzymatic activity of the resulting electrodes and to the enzyme leakage from the paste during storage in phosphate buffer. The in situ electrogenerated PPy improves the enzyme immobilisation within the paste since practically no enzyme was lost in solution after 72 h of immersion. Moreover, the enzyme activity remains particularly stable under storage since the biocomposite structure conserves 80% of its activity after 1 month of storage. Following the optimisation of the paste composition, PPO-based carbon paste biosensors were prepared and presented excellent analytical properties toward catechol detection with a sensitivity of 4.7 A M(-1) cm(-2) and a response time lower than 20 s. The resulting biosensors were applied to the determination of polyphenolic compounds (e.g., epicatechin and ferulic acid).

Titanium dioxide (TiO2) photocatalysis is a possible alternative/complementary technology to conventional water treatment methods. The TiO2 catalyst may be used as slurry or it may be immobilised onto a supporting substrate. With immobilised TiO2 films mass transfer problems occur in most photocatalytic reactors, which results in a reduction of reactor efficiency and in the accuracy of measured catalyst efficiency and kinetics. In order to determine the real intrinsic kinetics of photocatalytic reactions on immobilised TiO2 films a stirred tank reactor (STR) was used. The reactor incorporated a propeller and a baffle, thus providing good mixing and efficient mass transfer to the TiO2 film. Degussa P25 was immobilised onto borosilicate glass by a dip coating method and the kinetics of the photocatalytic degradation of the model pollutants, formic acid and oxalic acid were investigated as a function of catalyst loading, initial pollutant concentration and propeller rotation speed. The rate of degradation. of either acid. was not mass transfer limited at propeller speeds greater than 1000 rpm. The rate of formic acid degradation was dependent upon catalyst loading up to a maximum loading above which a decrease in the degradation rate was observed. The apparent quantum yield for the photocatalytic degradation was 5% for oxalic acid and 10% for formic acid. This compares very well with usual reported apparent quantum efficiencies for photocatalysis which are similar to1%. The photocatalytic oxidation of both acids could be described using a Langmuir-Hinshelwood kinetic model. (C) 2003 Elsevier B.V. All rights reserved.

The photocatalytic and electrochemically assisted photocatalytic degradation of formic acid was investigated in an one-compartment photoelectrochemical cell incorporating nano-crystalline TiO2 electrodes prepared by the immobilisation of Degussa P25 on tin oxide coated glass. The application of +1.0 V (SCE) to the TiO2 electrode resulted in a marked increase in the rate of formic acid degradation under anaerobic and air sparged conditions as compared to the open circuit electrode. When the solution was sparged with O2 the application of a positive bias did not result in a major increase in the rate. A second cathode compartment was added to the cell for the simultaneous photocatalytic oxidation of formic acid and the recovery of copper ions from solution. Formic acid was degraded at the TiO2 photoanode and copper metal was recovered at the copper mesh cathode with a high efficiency.

This work investigated the potential of semiconductor photocatalysis for the treatment of drinking water. More specifically it determined the effect of the following operational parameters on the adsorption, rate of initial degradation, and mineralisation of Aldrich humic acid solutions: concentration, temperature, oxygen, light intensity, and pH. The kinetics of humic acid degradation were found to be complex due to the heterogeneity and high molecular weight of these compounds. Models which are often applied to semiconductor photocatalysis did not fit; the high adsorption of humic acid on the titanium dioxide surface needed to be taken into account.

Pathogens in drinking water supplies can be removed by sand filtration followed by chlorine or ozone disinfection. These processes reduce the possibility of any pathogens entering the drinking water distribution network. However, there is doubt about the ability of these methods to remove chlorine resistant microorganisms including protozoan oocysts. Concern has also been raised about the production of disinfection by-products following the chlorination process. Titanium dioxide (TiO 2 ) photocatalysis is a possible alternative/complementary drinking water treatment method. TiO 2 electrodes were prepared by the electrophoretic immobilisation of TiO 2 powder (Aldrich and Degussa P25). These electrodes were tested for their photocatalytic bactericidal efficiency. E. coli K12 was used as a model test organism. The rate of disinfection was greater for the P25 electrode compared to the Aldrich electrode under open circuit conditions. The application of an electrical bias to the working electrode increased the rate of disinfection by ∼40% for the P25 electrode and ∼80% for the Aldrich electrode. The effect of applied potential was more pronounced under conditions of high initial bacterial cell loading and high light intensities. Bacterial recovery did not occur up to 48 h after disinfection.

In this paper a comparison between three commercially-available, screen-printable graphite inks for the construction of phenolic biosensors is made. The enzyme tyrosinase was immobilised within a polymer matrix and the substrate catechol was used to characterise the bio-electroanalytical response of each electrode. Biosensors fabricated from Gwent graphite inks exhibited the greatest sensitivity (5740 mA mol cm −2 ) compared to Dupont and Acheson graphite-based inks. This difference in sensitivity was attributed to a combination of a larger electroactive surface area, and thus a greater number of immobilised enzyme molecules. However, the dynamic range was considerably smaller (0.025–14 μM) indicating that the enzyme molecules were easily accessible to the substrate catechol. The surface properties of the biosensors were characterised using ac impedance, which indicated that the presence of the polymer on the electrode surface not only increased the charge-transfer kinetics of the three biosensors, but also increased the surface roughness of biosensors fabricated from Gwent inks. On the basis of these results Gwent graphite-based inks were used for analysis of phenolic compounds in lager beers by flow-injection analysis. The biosensor displayed favourable response characteristics, but cannot differentiate between the various phenolic compounds present in the samples. Nevertheless, the biosensor maybe suitable for indicating the phenolic status of beer or brew samples compared to time-consuming traditional methods, e.g. colorimetric or chromatographic methods.

Micro-molar concentrations of aqueous 17-β-oestradiol were 98% destroyed in 3.5 h by photocatalysis over the titanium dioxide powder immobilised on Ti-6Al-4V alloy. The concentration of oestradiol was determined by HPLC with fluorescence detection. The degradation kinetics were fitted to a Langmuir–Hinshelwood model with k (S) = 4.4 × 10 −2 μmol dm −3 min −1 and K (S) = 0.347 dm 3 μmol −1 . The pseudo-first-order rate constant (1.57 × 10 −2 min −1 ) was in line with the 50% degradation time of 40 min. The apparent quantum yield per electron was φ e ′ = 0.41%. The effect of pH on the initial rate of degradation was similar to that reported for phenol.

Photoanodes were prepared by immobilising TiO2 powder on the surface of Ti-6Al-4V alloy. The TiO2 electrode was incorporated into the anode compartment of a two compartment photoelectrochemical cell and short circuited to a copper wire counter electrode in the cathode compartment. The anode compartment contained oxalate solution and the cathode compartment contained copper sulphate solution. On illumination of the TiO2 anode oxalate was photooxidised at the surface of the TiO2 and the photogenerated electrons moved via the external circuit to the counter electrode where copper ions were reduced. The total charge passed during the experiments was consistent with the amount of copper plated out at the counter electrode. Ninety-seven per cent of the copper in solution was recovered after 120 min of irradiation under unoptimised conditions. (C) 1999 Elsevier Science B.V. All rights reserved.

Carbon dioxide was reduced photocatalytically using aqueous CdS or ZnS colloids containing tetramethylammonium chloride to give the dimeric and tetrameric products namely, oxalate, glyoxylate, glycolate and tartrate. A model is presented to explain the role of the tetramethylammonium ions. Studies were also performed using ZnO, SiC, BaTiO 3 and Sr TiO 3 , which in the absence of tetramethylammonium ions produced formate and formaldehyde. The relative quantum efficiencies of the six semiconductors were related to their band gaps and conduction band potentials. The role and effectiveness of several `hole acceptor' (electron donor) compounds in this process is shown to be related to their redox potentials.

TiO2 electrodes were prepared by the electrophoretic immobilisation of Degussa P25 TiO2 powder on titanium alloy (Ti–6Al–4V). These electrodes were used as photoanodes in a two compartment photoelectrochemical cell to investigate the photocatalytic oxidation of oxalate. The effects of electron acceptors, i.e. oxygen and Fe3+, on the rate of mineralisation of oxalate were determined. Significant increases in the short-circuit photocurrent densities were observed upon the addition of oxalate to the anode compartment when oxygen was present only in the cathode compartment. When oxygen was present in both the anode and cathode compartments, little photocurrent was observed. Oxygen scavenges both conduction band electrons from the TiO2 surface and primary radicals formed in the initial one electron oxidation of oxalate. An increase in the anodic photocurrent and rate of degradation of oxalate was observed when oxygen was replaced with Fe3+ as the electron acceptor in the cathode compartment. The TiO2 coated anodes behaved as microporous nanocrystalline film electrodes where the anodic photocurrent is dependent on the diffusion of majority charge carriers through the film. Oxalate acts as an efficient hole acceptor, inhibiting the photoanodic production of oxygen.

TiO2 powder was immobilised on solid support substrates (stainless steel, titanium alloy, titanium metal, and tin oxide coated glass) using electrophoretic coating and spray coating. Electrochemical anodisation of titanium metal was also carried out to give a thin film of TiO2 on the surface. The coated substrates were annealed in air at elevated temperatures to improve the adhesion of the catalyst to the supporting substrates. The photocatalytic efficiency of the TiO2 coatings was compared using the degradation of phenol in aqueous solution as a standard test system. In the case of the powder derived films, the photocatalytic efficiency was found not to be markedly dependent upon either the substrate used or the annealing temperature employed in the coating process. Surface analysis of the immobilised TiO2 showed no significant differences in the elemental composition or in band gap energies. The electrophoretic coating method was found to be the most reproducible resulting in highly fractured thick films with high photocatalytic activity. (C) 1998 Elsevier Science B.V.

The effect of different hole acceptors on the photocurrent response of particulate TiO2 anodes was investigated. In all cases there was a significant increase in the photocurrent over that observed for the photooxidation of water. The photocurrent efficiency observed in the presence of hole acceptors followed the order: oxalate > formate > acetate > methanol. The order of efficiency of the hole acceptors does not correlate with reported free hydroxyl radical reactivity constants or with reported oxidation potentials for the substrates studied. Adsorption measurements made in the dark showed that oxalate was much more strongly adsorbed on the electrode surface than either acetate or formate. Impedance spectroscopy indicated an increase in the capacitance of the system due to the presence of oxalate, both in the dark and under illumination.

Apple tissue containing polyphenol oxidase was incorporated in carbon paste matrix to make a biosensor for the analysis of flavanols in beers. The electrochemical and structural properties of different carbon paste compositions were assessed using ac impedance, other electrochemical techniques and scanning electron microscopy. Results are discussed regarding electroactive surface, mechanical stability, impedance parameters and surface roughness of the carbon pastes. From these results, an optimum graphite/Nujol ratio of 4 was chosen and applied to the construction of apple-based biosensors. The electrochemical, morphological and analytical properties of these sensors were investigated with respect to biomolecule loading. The analytical properties (sensitivity, Imax) of these biosensors increased with plant tissue loading. This result was correlated with ac impedance parameters, electroactive surface and carbon paste morphology. A biosensor using a plant tissue/carbon paste ratio of 0.086 was used to assess sensitivity and selectivity to various flavanols and phenolic compounds.

Various plant tissues containing polyphenol oxidase were incorporated in carbon paste electrodes to make biosensors for the determination of catechol related components in beers, principally flavanols such as catechins and their dimers and trimers. Calibration graphs were produced for each plant tissue with both catechol and (−) epicatechin as standards. The response of the banana based sensor was rather erratic and showed a large zero error, probably due to flavanols in the banana. Potato, wet apple and dried apple were all satisfactory. The response with dried apple was 60% that of wet apple, but showed greater stability. The response of (−)epicatechin was 1.4 to 1.5 times that of catechol with wet or dried apple. Banana and potato based sensors were used for the determination of total flavanols in a range of commercial beers and lagers. Results with the banana were rather high due to the zero error. Good analytical data were obtained with potato, comparable to those obtained using colorimetric or liquid Chromatographic analyses.

Currently drinking water from the Mourne Mountains in Northern Ireland is treated by the traditional process of coagulation and filtration. Semiconductor photocatalysis was investigated as an alternative. Aldrich humic acid solutions (100 mg dm−3, filtered through 0.22 μm Millipore filters) were photolysed in a 0.1% suspension of Degussa P25 titanium dioxide using a mercury lamp. Degradation of humic acid was followed over time by measuring colour in Hazen units, 254/400 nm absorbance, fluorescence (excitation 237 nm; emission 400 nm) and size exclusion high performance chromatography. The extent of complete oxidation was determined by measuring dissolved organic carbon, chemical oxygen demand and carbon dioxide released. It took approximately 12 min to reduce the humic acid concentration by half; however 50% complete mineralization took 60 min. Some of the intermediates of reaction were highly fluorescent. In the absence of oxygen, light or titanium dioxide no significant degradation occurred.

The yields of oxalate and especially of glyoxylate and glycolate were enhanced by the use of methanol containing hydroxylamine and tetramethylammonium chloride in the controlled potential electrochemical reduction of carbon dioxide at a lead electrode. An eightfold increase in the solubility of CO2 was found compared with water so that the initial concentration was 4.48 × 10−3m. Products were mainly glyoxylate (up to 9.7 × 10−3m at −2.0 V) and glycolate (up to 6.6 × 10−3m at −2.3 V) with some oxalate (up to 1.3 × 10−4m at −1.7 V) depending on the time and electrode potential. Very little formate was produced. Current densities were about 30 A m−2.

Variable sweep rate voltammetry of carbon dioxide at a glassy carbon electrode in aqueous solutions containing tetraethylammonium perchlorate gave two successive one-electron kinetically controlled waves. The apparently low peak potentials (−0.74 V vs. SCE at pH 10) were rationalised by consideration of the ionic radius of the tetraethylammonium cation relative to other cations. The rate limiting process appeared to be release of carbon dioxide molecules from aqueous bicarbonate. Equilibrium constants for this process were measured by rotating disc electrode voltammetry and found to be 0.069 at pH 9.1 and 0.076 at pH 10. The rate constant for the formation of carbon dioxide molecules was calculated from Kl 1 2 and found to be kf = 1.6 s−1 at pH 10. However, these data conflicted with published data for this process from homogeneous kinetics. An adsorption model was applied to rationalise the data. The surface coverage was found to be 0.019 at pH 10, in line with published data.

Photoexcited electrochemically generated quinone radical anions reduced 1,2-dibromobenzene to bromobenzene, 1,4-dibromobenzene to bromobenzene and 4-chlorobenzonitrile to benzonitrile. In the presence of anthracene, 2-bromophenyl-, 4-bromophenyl- and 4-cyanophenyl-anthracenes were formed. With acetaldehyde, acetone, acetophenone, benzaldehyde and benzophenone, the major products were the corresponding pinacols, with small amounts of the two-electron secondary alcohols. In acetonitrile as solvent, cinnamonitriles, hydrocinnamonitriles and phenylglutaronitriles were formed in addition to the alcohols. Glyoxylic acid was reduced to tartaric, glycolic and malic acids. The reduction of CO2 was unsuccessful.

Aqueous solutions of CO2 containing tetramethylammonium chloride were photolysed with visible light in the presence of colloidal ZnS to yield tartaric acid, glyoxylic acid, oxalic acid, formic acid and formaldehyde.

The photoelectrochemistry of quinone radical anions has been demonstrated qualitatively by the photoassisted reduction of methyl viologen with benzoquinone and of neutral red with chloranil. Data were then collected for the estimation of quenching rate constants using Marcus–Weller theory. Reduction potentials of seven quinones were obtained in four solvents (and two aqueous mixtures) by cyclic voltammetry. The solvent effects on these potentials were studied by fitting them to the Taft relationship. The effects of proton donors were also noted. Absorption spectra of the radical anions were measured and the solvent effects noted and commented upon. From the molar absorption coefficients of the radical anions, the mean lifetimes of the excited states were estimated. Fluorescence spectra were obtained for anthraquinone and naphthaquinone radical anions and excitation energies were calculated. These values were estimated for the other quinones. Values of redox potentials for the excited radical anions were thence obtained. The Gibbs energies of the electron transfers between the excited quinone radical anions and the various substrates were obtained and hence the Gibbs energies of activation were calculated using the Marcus equation. The quenching rate constants were calculated using the Rehm–Weller equation and plotted vs. ΔG giving a characteristic Marcus plot including some data in the inverted region. The significance of the inverted region is discussed.

Photoexcited electrochemically generated quinone radical anions reduced 1,2-dibromobenzene to bromobenzene, 1,4-dibromobenzene to bromobenzene and 4-chlorobenzonitrile to benzonitrile. In the presence of anthracene, 2-bromophenyl-, 4-bromophenyl- and 4-cyanophenyl-anthracenes were formed. With acetaldehyde, acetone, acetophenone, benzaldehyde and benzophenone, the major products were the corresponding pinacols, with small amounts of the two-electron secondary alcohols. In acetonitrile as solvent, cinnamonitriles, hydrocinnamonitriles and phenylglutaronitriles were formed in addition to the alcohols. Glyoxylic acid was reduced to tartaric, glycolic and malic acids. The reduction of CO2 was unsuccessful.

Aqueous carbon dioxide is photochemically reduced in the presence of semiconductor suspensions and colloids. Experiments have been performed using CdS, ZnO, SiC, Ba TiO3, and SrTiO3, dispersions. Product analysis showed the formation of formic acid formaldehyde, sometimes methanol and in the presence of tetramethylammonium chloride, two-carbon products such as glyoxylic and acetic acids and sometimes acetaldehyde. Product yields and photochemical yields were studied as a function of pH, solution composition, added sacrifical electron donors and catalysts, irradiation intensity and irradiation time. A demonstration experiment with a flat-bed solar collector showed the formation of formic acid and formaldehyde using direct sunlight.

Most bioassays were originally of the photometric type. They involved changes in a species which involved a strong change in photometric properties. Probably the best known is the involvement of NAD+/NADH in biochemical reactions.

There are three basic electrochemical processes which are useful in transducers for biosensors: (i) potentiometry, the measurement of a cell potential at zero current; (ii) voltammetry (amperometry), in which an oxidising (or reducing) potential is applied between the cell electrodes and the cell current is measured; and (iii) conductimetry, where the conductance (reciprocal of resistance) of the cell is measured by an alternating current bridge method.

In order to make a viable biosensor, the biological component has to be properly attached to the transducer. This process is known as immobilisation. There are five regular methods of doing this, as follows.

Crystals have always had a fascination for people. Are not diamonds a girl’s best friend? Their perfect shapes, their colours and the way they can be grown from a clear solution have long had a fascination. They have been (and are) involved in magic and sorcery and ‘new age’ religion. But they also have considerable scientific importance. The crystalline state is the one with the highest order, and therefore minimum entropy. Electrical properties of crystals have been of interest since the discovery of electricity. Crystal sets were the earliest form of radio transmitters and receivers. After many years in which thermionic valves dominated the radio industry, it was realised that crystals gave much more accurate frequency control than the capacitor-inductance coil combination. Now, of course, there has been a switch to transistors, but these sold-state devices are all crystals, as are all the ‘chips’ in computers.

There has been considerable controversy about these reductions. Bewick and Greener [7, 8] claimed to have produced glycolic acid from CO2 on lead and malic acid from CO2 on mercury but they did not detect glyoxylic acid. Similar results were reported by Wolf and Rollin [9] but others were unable to repeat these results, or obtained different products [3]. A number of workers have studied the voltammetric reduction of glyoxylic acid [10, 11]. But there have been no direct preparative studies since Bauer [12] in 1931. Some preparative studies are presented here. It has been generally assumed that the electrochemical reduction of glyoxylate goes by a two electron process to give glycolate (perhaps explaining Bewick's results [7]). However another possibility exists: the formation of tartaric acid by a one-electron reduction followed by a dimerisation (pinacol formation). Further reduction of tartaric acid could lead to malic acid (as claimed by Bewick [8] and Wolf [9]). We observed an HPLC peak from the reduction products of the reduction of oxalic acid which could be attributable to malic acid [4].

Skin contact electrodes require electrolyte gels between the skin and the electrode in order to ensure good electrical contact. The effect of different types of electrolyte gel on skin impedance was studied. The main types of gels used were wet gels, karaya-gum based hydrogels and synthetic copolymer-based hydrogels [2-acrylamido-2-methylpropanesulfonic acid–N, N′-methylenebis(acrylamide) copolymers]. The effect of variation in gel composition on the impedance of the skin was investigated.

Cyclic voltammograms of quinones were recorded in acetonitrile in the presence of various substrates: carbonyl compounds, halobenzenes, Methyl Viologen and Neutral Red. When illuminated with light of λ >410 nm, catalytic waves were observed. From the ratio of the catalysed to uncatalysed peak current, electron transfer rate constants were calculated using the working curves of Saveant and coworkers. The values of these rate constants were compared with the values obtained by Shukla and Rusling for different systems using a similar method and with quenching rate constants calculated using Rehm–Weller–Marcus theory.

The photoelectrochemistry of quinone radical anions has been demonstrated qualitatively by the photoassisted reduction of methyl viologen with benzoquinone and of neutral red with chloranil. Data were then collected for the estimation of quenching rate constants using Marcus–Weller theory. Reduction potentials of seven quinones were obtained in four solvents (and two aqueous mixtures) by cyclic voltammetry. The solvent effects on these potentials were studied by fitting them to the Taft relationship. The effects of proton donors were also noted. Absorption spectra of the radical anions were measured and the solvent effects noted and commented upon. From the molar absorption coefficients of the radical anions, the mean lifetimes of the excited states were estimated. Fluorescence spectra were obtained for anthraquinone and naphthaquinone radical anions and excitation energies were calculated. These values were estimated for the other quinones. Values of redox potentials for the excited radical anions were thence obtained. The Gibbs energies of the electron transfers between the excited quinone radical anions and the various substrates were obtained and hence the Gibbs energies of activation were calculated using the Marcus equation. The quenching rate constants were calculated using the Rehm–Weller equation and plotted vs. ΔG giving a characteristic Marcus plot including some data in the inverted region. The significance of the inverted region is discussed.

Aqueous solutions of CO2 containing tetramethylammonium chloride were photolysed with visible light in the presence of colloidal ZnS to yield tartaric acid, glyoxylic acid, oxalic acid, formic acid and formaldehyde.

Linear sweep voltammetry of glyoxylate on a stationary mercury electrode showed strong adsorption effects in addition to kinetic control because of the hydration of the aldehyde groups. The methods of Shain were used to minimise the adsorption and to determine the kinetic function at pH 9.0. The dehydration rate constant was found to be 1.14 ± 0.03 s-1.

In order to look for new catalytic routes in the electroreduction of carbon dioxide, the voltammetry of mixed solutions of carbon dioxide and sulphonated phthalocyanines in DMSO has been investigated. An additional reduction wave was observed at more negative potentials than the second (−1.3 V SCE ) phthalcyanine reduction wave, for both copper and nickel phthalocyanines. No additional waves were observed with cobalt phthalocyanine, though shifts were observed in the second phthalocyanine reduction wave for all three phthalocyanines. The additional waves are thought to indicate the formation of a complex between CO 2 and the reduced phthalocyanine ion.

We have studied voltammetric reduction waves for carbon dioxide in water, dimethylsulphoxide, acetonitrile and propylene carbonate using electrodes of glassy carbon, mercury, platinum, gold and lead as applicable. The apparent number of electrons involved was determined from the current function and in some cases by chronoamperometric analysis of the diffusion controlled tail of the wave. Results were compared with those from other workers.

The aqueous voltammetry of Co, Ni, Cu and Zn sulphonated metallophthalocyanines has been examined and compared to the voltammetry in DMSO. In DMSO, Ni, Cu and Zn phthalocyanines exhibit two quasi-reversible, phthalocyanine ligand based, one-electron transfers. Two reduction waves were also observed for Co phthalocyanine. However, the first reduction wave appeared to involve both a metal based electron transfer and a change in the coordination of the metal. The aqueous voltammetry of the cobalt phthalocyanine was similar to its non-aqueous behaviour. The other three phthalocyanines did not show the quasi-reversible behaviour in aqueous solution. Reversibility was much diminished and the forms of the waves changed greatly, indicating the involvement of protons in the redox processes.

Carbon dioxide, in aqueous solution containing tetramethylammonium ions at pH 9 was reduced in two steps to yield oxalic acid at −0.9 V and glyoxylic acid at −1.8 V on carbon and on mercury. Products were analysed by HPLC and confirmed by specific colorimetric measurements.