The desorption of 2-mercaptoethanesulfonate (MES) spontaneously adsorbed on Au(111) has been studied by using both potential-step and voltammetric experiments. From the amount of gold oxide formed during the oxidation sweep in the fingerprint region it is shown that the adsorption process induces structural changes of the gold surface. It is also shown that together with the reductive desorption of MES ions a concomitant faradaic process occurs. The results suggest that this process is connected to the reduction of solvent on the structurally modified gold electrode. The reductive desorption process of MES undergone at more negative potentials is characterized by a single peak in the voltammetric response and the presence of a maximum in the chronoamperogram. It is shown that the logarithm of the maximum current, the time at which the maximum current appears, and the peak width at half height depend linearly on potential. An experimental protocol for the desorption/readsorption of MES based on a potential-step experiment followed by cyclic voltammetry is outlined as an appropriate tool to analyze simultaneously the desorption of adsorbed and readsorbed MES. A linear dependence between the two desorption peak potentials and the surface concentration of adsorbate was observed. Changes in the environment surrounding the adsorbed moities and in the potential of zero charge of the electrode are shown to be the factors ruling this dependence.

A new method is proposed for the determination of the potential of zero charge of gold electrodes modified with thiol monolayers. It makes use of the immersion technique, in combination with a vapor deposition protocol to build the thiol monolayers. As compared to previous methods, the present approach provides more accurate results, particularly in the case of long-chain alkanethiol monolayers, and it is applicable to any monolayer irrespective of its degree of hydrophilicity. Results are presented for a series of 12 alkanethiol monolayers and for 11-mercaptoundecanol and 11-mercaptoundecanoic acid monolayers. Good agreement is found between the variation of the potential of zero charge along the alkanethiol series with the corresponding change of the surface work function. The potential of zero charge of the 11-mercaptoundecanoic acid monolayer is shown to depend on the extent of dissociation of the acid, thus opening the possibility of applying this type of measurements to the study of surface ionization processes.

The formation and reductive desorption of self-assembled monolayers of 6-mercaptohexanol on mercury has been studied by using cathodic stripping voltammetry and capacitative transients, including the possibility of expanding or contracting the electrode area at the end of the preconcentration step. Experimental evidence shows the existence of three sequential stages during the formation of a thiol self-assembled monolayer. Each of these stages can be associated to the presence of (i) a low surface density state of oxidized thiol molecules, characterized by a single electrodimerization wave, (ii) a high surface density state, characterized by the emergence of a second voltammetric wave, and (iii) an ordered monolayer, which gives rise to a voltammetric spike. On the basis of electrode expansion experiments, a method is described to determine the surface concentrations of oxidized products, which does not require a baseline subtraction of the voltammograms to account for the nonfaradaic current. Quantitative voltammetric fits are consistent with the initial formation of a mixture of noninteracting monomers and dimers of oxidized thiol. The value of the maximum surface concentration and the ability to block the Ru(NH 3 ) 6 3+ electron transfer reveal that oxidized thiol molecules adopt a nearly perpendicular orientation in the high surface density state, which hampers ionic permeation. A theoretical model is proposed to account for the observed voltammetric behavior. The transition from the lower to the higher surface density states is modeled as a chemical step involving the exchange of metal free sites. Capacitative transients are also interpreted in terms of the three-stages model.

Packing restrictions and hydrophobic interactions are likely to lead to a spatial distribution of redox centers in electroactive monolayers. A mean field analysis of the electrochemical implications of spatial redox dispersion in SAMs, including the possibility of surface ion pair formation, has been carried out. The boundary value problem associated with a layered distribution of potential-induced charges has been solved by using the orthogonal collocation technique under equilibrium conditions. Spreading of the redox centers into a 3D dielectric slab results in broader and asymmetric voltammograms, reflecting a layer-by-layer redox conversion. It is also shown that the voltammetric shape is sensitive to the specific features of the spatial redox distribution, and theoretical requirements for the appearance of asymmetric broadening are examined in terms of the electrostatic properties of the monolayer. It is suggested that this type of spatial inhomogeneity may cause some of the broad and asymmetric voltammetric shapes that often characterize the electrochemical behavior of electroactive SAMs, and that some structural information can be gained from the analysis of these voltammograms, as long as electrolyte ions do not permeate the organic monolayer. The effect of surface ion association on the voltammetric features is also examined, and it is interpreted in terms of the distinct sensitivity of the potential at each redox plane with respect to the local counterion concentration. Comparison is made with the experimental results of Chidsey et al. (J. Am. Chem. Soc. 1990, 112, 4301) for the oxidation of FcCO 2 (CH 2 ) 11 SH/CH 3 (CH 2 ) 9 SH and Fc(CH 2 ) 16 SH/CH 3 (CH 2 ) 15 SH mixed monolayers.

The voltammetric response of an electrode covered with an electroactive self-assembled monolayer is modeled including discreteness of charge effects and interfacial ion association. Discreteness of charge potentials are estimated according to the hexagonal array model of Macdonald and Barlow, and results are compared with those obtained in previous work with the cutoff disk model. As a consequence of the slower variation of the discreteness of charge potential on the applied electrode potential, voltammetric waves are predicted to be asymmetrical and wider than those computed from the cutoff disk model. For relatively high redox coverage and/or small values of the integral capacities of the inner and outer part of the monolayer, a negative differential capacity is predicted. This is a consequence of the additional stabilization provided by the discreteness of charge effect, which allows the charge density at the redox plane to increase faster than the charge density on the electrode surface when the monolayer is being oxidized. Comparison with experimental results shows that inclusion of the discreteness of charge effects results in a variation of the absolute value of the interfacial parameters, while their qualitative trends are the same as those obtained on the basis of an average potential model. Therefore, only the physical significance of the fitting parameters may help to discriminate among the different models.

The influence of ion pairing on the potential distribution at an electrode modified with a redox-active selfassembled monolayer is considered. The voltammetric response of the film is modeled as a function of its dielectric properties and the extent of association between the redox centers and counterions in solution. On the basis of the scheme of squares, a general treatment for any number of association steps is introduced. In addition, the effect of the counterion selective permeation on the voltammetric features is presented. It is shown that ion pairing and double-layer effects are strongly coupled, and may lead in some situations to the same variation of the peak potential with electrolyte concentration. The possibility of obtaining association parameters from voltammetric experiments is discussed. Comparison is made with Rowe and Creager's data on the oxidation of mixed monolayers of ferrocene-n-hexanethiol and n-alkanethiols.

Changes have been demonstrated in the concentration of those volatile compounds responsible for green sensory notes in virgin olive oils obtained from four different varieties at three stages of ripeness. The information corresponding to each volatile compound for each stage of ripeness, after fuzzy filtering of the quantitative data, has been established and the relationship between volatile compounds and green sensory attributes has also been demonstrated by means of principal components analysis, correlation and stepwise linear regression analysis.KEY WORDS Oleu europeu L.; virgin olive oil; volatiles; green sensory note; ripeness; statistics INTRODUCI'IONVirgin olive oil is the oil extracted from the fruit of the olive tree, Olea europea L. Mediterranean countries are the main producers and consumers of this vegetable oil which is almost the only one consumed without a fuither refining process. It is characterized by a fragrant and delicate flavour which has been appreciated by consumers since ancient times.Different sensory notes describe the characteristic flavour of this fo~dstuff,'-~ the 'green' attribute being one of the most important! A study of the relationship between fruit maturity and aroma components showed that the characteristic flavour is obtained by the balance between green and fruity Various volatile compounds are responsible for the green sensory note, with aliphatic compounds and the corresponding hexyl esters5*'-* being the main contributors to the 'unripe' component of the fruit flavour. It has been demonstrated that these compounds are biosynthesized from C18-unsaturated fatty acids in plants following the lipoxygenase pathwayg*" and they are usually the major components in virgin olive oil aroma.*Author to whom correspondence should be addressed.The aims of this work were several. The first was to determine whether there are differences, related to fruit ripeness, in the volatile content responsible for the green odour perception in virgin olive oil, analysed by a dynamic headspace gas chromatographic method. The second was to characterize each stage of ripeness by means of fuzzy logic" and the third was to study the relationships between the compounds responsible for this perception and green sensory attributes evaluated by sensory panels. EXPERIMENTAL SamplesVirgin olive oil was properly obtained from fresh, healthy fruits of good quality collected at three stages of ripeness: at the beginning of the harvest period when fruits are of green colour, in the middle of the harvesting and at the end when fruits are of black co10ur.'~ Olive fruits are drupes which change colour during ripening depending on the variety. The four varieties selectedarbequina, picual on Marteiia (Spain), coratina or racioppa (Italy) and koroneiki (Greece)-were chosen as representing a substantial proportion of the bottled virgin olive oil tradeI3*l4 and being completely different from either sensory evaluation6 or chemical compo~ition.'~

The study of direct electron transfer between enzymes and electrodes is frequently hampered by the small fraction of adsorbed proteins that remains electrochemically active. Here, we outline a strategy to overcome this limitation, which is based on a hierarchical analysis of steady-state electrocatalytic currents and the adoption of the "binary activity" hypothesis. The procedure is illustrated by studying the electrocatalytic response of horseradish peroxidase (HRP) adsorbed on graphite electrodes as a function of substrate (hydrogen peroxide) concentration, electrode potential, and solution pH. Individual contributions of the rates of substrate/enzyme reaction and of the electrode/enzyme electron exchange to the observed catalytic currents were disentangled by taking advantage of their distinct dependence on substrate concentration and electrode potential. In the absence of nonturnover currents, adoption of the "binary activity" hypothesis provided values of the standard electron-transfer rate constant for reduction of HRP Compound II that are similar to those reported previously for reduction of cytochrome c peroxidase Compound II. The variation of the catalytic currents with applied potential was analyzed in terms of the non-adiabatic Marcus-DOS electron transfer theory. The availability of a broad potential window, where catalytic currents could be recorded, facilitates an accurate determination of both the reorganization energy and the maximum electron-transfer rate for HRP Compound II reduction. The variation of these two kinetic parameters with solution pH provides some indication of the nature and location of the acid/base groups that control the electronic exchange between enzyme and electrode.