Search

The mechanism of the heterogeneously catalysed reaction between N,N-dimethyl-p-phenylenediamine and CoIII(NH3)5CI2+ at silver, gold and silver-on-gold (Ag/Au) discs has been studied by means of electrochemical methods. Both the mixed (or mixture) potentials and the mixture currents were determined by recording the current–potential curves of the reactants. Silver halide, formed during the reaction, was determined by subsequent galvanostatic reduction. The reaction at silver was strongly inhibited by the formation of silver halide whereas the reaction at gold was inhibited by adsorption of the organic compounds as well as by iodide. At Ag/Au discs almost no inhibition occurred. The explanation is that the reduction of the cobalt complex takes place predominantly at the silver surface whereas the simultaneous oxidation of the p-phenylenediamine occurs on the gold. In this way the formation of inhibiting silver halide is suppressed as well as the inhibition caused by the adsorption of p-phenylenediamine.

The peroxide bleaching of aqueous phenolphthalein, alizarin and crocetin in alkaline solution is not caused by free radical species, nor by singlet oxygen, but by hydrogen peroxide and the perhydroxyl anion, HOO–.

A mechanism of electron-transfer-through-the solid is proposed to account for the widespread catalysis by carbons of redox reactions in solution. Charcoal catalysis of the aqueous reaction between Fe3+ and I– ions was predicted, found, and studied kinetically. The activation energy of the heterogeneous process was only 24 kJ mol–1 compared with 100 kJ mol–1 for the homogeneous reaction. The [Fe(CN)6]3–+ I– reaction in water was found to be catalysed by carbons as well as by several phthalocyanines. Neither reaction was catalysed by solids such as SiO2 or Ba[SO4] which do not conduct electronically.

The cation and anion transference numbers of aqueous silber perchlorate at 25°C have been determined by the direct moving boundary method from 0.01 to 0.2 M. The cation constituent transference numbers rise in a monotone fashion with increasing concentration contrary to the expectations, of the limiting Debye-Huckel-Onsager theory. The results can be fitted by the revised 1963 Fuoss-Onsager theory only by the assumption of unrealistically large distances of closest approach. The behavior of aqueous silver perchlorate closely parallels that of aqueous silver nitrate and suggests that some property of the silver ion is responsible.

The anion transference numbers of aqueous perchloric acid at 25°C have been determined from 0.01 to 0.3M by the moving-boundary method. Confirmation was obtained by cation measurements at 0.1M which showed that low values resulted if the following solution contained traces of acid. Reasons for the variation of moving-boundary measurements with current have been discussed. The transference numbers gave physically more reasonable parameters with the revised 1963 Fuoss-Onsager theory but were fitted less well by it than by the earlier Fuoss-Onsagerand Pitts theories.

The rates of infusion of theaflavins, thearubigins, and caffeine from Koonsong Broken Pekoe into water were measured at 79.5 and 94.0°C. The kinetic data have been interpreted in terms of a simple two-phase model, and the rate constants of infusion and of re-absorption were calculated. At low infusion times thearubigins were extracted faster than theaflavins, with the caffeine rate intermediate. However, the rate constants for the overall infusion process were quite similar for the three constituents. The corresponding activation energies are small as are those calculated from Long's data on the rate of extraction of all soluble material.

The rates of extraction of theaflavins from sieved Kapchorua Pekoe Fannings (600–710 μm) and from a Ceylon Broken Orange Pekoe Fine Leaf Blend have been measured at 80°C with a set of aqueous salt and buffer solutions spanning the pH range 3·1 to 8·1. There was little effect on adding 1:1 electrolytes but considerable decreases in rate occurred in the presence of calcium chloride. The rate constants in acid buffers were the same as in a normal unbuffered infusion but in alkaline buffers of pH 8 they were over 50% larger. This phenomenon was quantitatively accounted for by the partial dissociation of theaflavins at higher pH and the greater diffusion coefficient of the resulting ionic salt. The relevance of these findings to tea brewing and to industrial theaflavins extraction is discussed.

The upper concentration limit for the direct moving boundary method for measuring transference numbers can be appreciably extended by introducing three new concepts. First, the closed side of the cell must contain an electrolyte whose apparent molar volume changes relatively little with concentration. Second, the closed electrode chamber must be stirred to avoid the build-up of regions of differing concentration. Third, the solution in this chamber should be pre-saturated with the insoluble salt involved in the electrode reaction. These three modifications allow the large volume correction for concentrated solutions to be calculated much more exactly. The new technique was tested with fair success for 3M KCl in water at 25°C using the KCl ← LiCl system. The high current dependence of the transference number was mainly attributed to the Soret effect at the boundary.

Concentrations of theaflavin and caffeine were determined in whole Kapchorua Pekoe Fannings and in six of its sieved fractions, as well as in Kapchorua Pekoe Dust, Betjan Flowery Broken Orange Pekoe and Rupai Flowery Orange Fannings and in one sieved fraction of each. The partition constants of theaflavin and caffeine between these leaves and water at 80°C were also determined. Neither property varied significantly with leaf size but did depend on leaf origin and manufacture. The concentration and partition constants for caffeine were greater than for theaflavin. As a corollary, the ratio of caffeine to theaflavin is larger in tea initially drawn from the teapot than in the beverage prepared after topping up with fresh hot water.

The rate of the isomerization of cis-diamminedichloroplatinum(II) to the trans form has been determined in aqueous KCl solutions from 65°C to 100°C. At 37°C the half-life is 1.8 years. Pt, AgCl, Si, SiO2, and graphite did not catalyse the isomerization while other chemical reactions occurred with Ag and Ketjenblack carbon.

A new kind of catalysis operative for certain reactions in the electrochemical double layer has recently been predicted and tested for by Despic et al. To investigate this phenomenon further we have determined the solvolysis rates of two tertiary butyl salts by a pH-stat technique. The rate of hydrolysis of t -butyl acetate was measured in the range 35–75°C in water and in 0.5 M Na 2 SO 4 solution, and in the presence of open-circuit and electrically pulsed gold and silver foil electrodes. Pulses were applied at 1 Hz over the same potential ranges as those used previously. The homogeneous rate constants were found to be much lower than those of Despic et al. and no significant catalytic effects were detected during various pulsing regimes. A similar set of experiments was carried out on the solvolysis of t -butyl bromide in 80 v/v % ethanol + water at 25°C where silver metal is known to catalyse the reaction heterogeneously. Catalysis by a floating silver electrode was confirmed, but applicaition of 1 Hz pulses actually decreased the rate. This could be explained by a simultaneous electrochemical reduction process at the more negative potentials. No evidence to substantiate non-faradaic electrocatalysis was found for either reaction.

Stable chlorine electrodes with low bias potentials have been developed by introducing 25% Ir+75% Pt electrodes and an improved gas line. With their use in cells with transference, cation constituent transference numbers have been measured at 25°C in NaCl solutions from 1.7 to 6 modal. These results agree well with four other sets of data in the literature but disagree with two further sets based on emf determinations with Ag/AgCl electrodes. A table of ‘best’ NaCl transference numbers is proposed. The conductances of the chloride ion-constituent in concentrated NaCl, KCl, and HCl solutions are compared.

The transference numbers of HCl in water at 25°C have been determined up to 8 mol-kg−1 by using cells with transference. The problem of the solubility of AgCl from Ag/AgCl electrodes was avoided by employing dilute chlorine gas/iridium electrodes for T H and hydrogen gas/platinum electrodes for T Cl . The sums of the independently measured T H and T Cl values never differed from unity by more than 0.9%. The cation constituent transference number of HCl was also measured at 1M by the recently modified moving boundary method, but at higher concentrations the Soret effect produced unacceptably large current dependences. Combination of these transference numbers with literature conductances, diffusion coefficients and activity coefficients led to a new set of phenomenological transport coefficients l ij . The resulting curve for l12/c vs. c behaves more normally than did the curve based on previous transference numbers.

Wagner and Traud postulated that the behaviour of a mixture of redox couples at an electrode can be quantitatively described by the addition of their individual current-potential curves. It is argued that the most satisfactory tests of this hypothesis are furnished by systems in which the metal electrode is not itself a chemical component of one of the couples, as with the couples Fe(CN) 6 3− /Fe(CN) 6 4− and I 3 − /I − at platinum. Electrochemical and kinetic experiments have therefore been carried out with this system. Agreement with the additivity hypothesis was found to be good for pre-oxidised platinum but poor for reduced platinum. The explanation lies in the fact that a reduced platinum surface preferentially adsorbs iodide ions, and this hinders the electron transfer of the other couple. Attention is drawn to other systems where the additivity hypothesis is likely to break down.

The addition of various semiconducting solids (silicon and metal sulphides) and of two ionically conducting solids [copper(I) and thallium(I) iodides] appeared either to catalyse or inhibit the rate of iodine formation by the reactions [Fe(CN)6]3–+ I– and Fe3++ I– in aqueous solution. The curious effect of CuI on the latter reaction was investigated in more detail. Adsorption experiments demonstrated that [Fe(CN)6]3– and I3– chemically attacked these solids. Their chemical instability in the presence of the oxidants or the reaction mixtures was confirmed by thermodynamic calculations and the observed kinetic effects could then be explained. Only silica and silver(I) iodide did not affect the rates of the reactions; adsorption tests and thermodynamic calculations showed that these two solids were stable under the experimental conditions. The catalytic effect produced by a solid is therefore genuine only if kinetic and/or thermodynamic evidence has established its chemical stability in the reaction mixture.

Hydrodynamic flow conditions were controlled by means of large horizontal rotating discs. These were coated with ground and sieved Kenyan Arabica coffee, and rotated at various speeds in distilled water at 25°C. The rate of extraction of caffeine was measured in each experiment. At a constant rotation speed, the first order rate constant rose steeply with decreasing size of coffee particle and was found to be inversely proportional to the square of the particle radius. For a given particle size, however, the rate constants were independent of the rotation speed. These two findings show that the rate-determining step for extraction from coffee-coated discs was the diffusion of caffeine through the bean particle and not its diffusion through the Nernst layer. The same mechanistic conclusion can be drawn for stirred coffee suspensions by applying the hydrodynamics of turbulent flow.

The rates of extraction of caffeine from sieved Kapchorua PF (600–710 μm) have been measured at 80°C with a range of aqueous salt and buffer solutions of ionic strength 0·11 mol dm −3 . The first-order rate constants and the half-times of infusion showed no trend with pH when buffers from pH 3·0 to pH 8·3 were employed. The rate constants decreased on the addition of common salts like NaCl, KCl and CaCl 2 but increased in the presence of electrolytes such as Bu 4 NCl that contain large ions. The results cannot be interpreted by changes in osmotic pressure although Donnan effects may be involved. Close parallels were found between the rate constants and the solubilities of caffeine in electrolyte solutions at 25°C. In particular, the values of both properties rise appreciably in the presence of species containing aromatic or other organic rings with which caffeine molecules associate.

The rates of infusion of theaflavin and caffeine into water at 80°C have been measured for Kapchorua Pekoe Fannings, Kapchorua Pekoe Dust, Betjan Flowery Broken Orange Pekoe and Rupai Flowery Orange Fannings whole teas and sieved fractions. For Kapchorua PF fractions the rates of theobromine infusion were also determined. Caffeine always infused faster than theobromine, and both infused faster than theaflavin. Rate constants were obtained from the slopes of first order kinetic plots which yielded straight lines with small intercepts. The rate constants for all three constituents increased by a factor of 2.2 as leaf size decreased from 850–1000μm to 500–600μm. The rates of extraction from whole teas varied more than fourfold. Kapchorua PD gave the most rapid rates and Betjan FBOP the slowest. Comparisons of similarly sized fractions of teas showed that leaf size distributions accounted for half these differences and the intrinsic properties of the leaf for the remaining factor of two. The Kenyan CTC teas gave higher rate constants than the Indian orthodox teas. From the rate constants, the effective diffusion coefficients for theaflavin and caffeine in Betjan FBOP leaf were calculated and found to be some 100 times smaller than in water. This shows the diffusion processes within the leaf to be complex ones.

Rates of extraction of caffeine into water at 80°C were measured for green and roasted Kenyan arabica coffee beans. The green beans were roasted to controlled extents in a Moda coffee roaster, ground, and sieved to a size range 0.85–1.18 mm. Certain physical properties (weight loss, colour, bean volume and infusion pH) were determined for each product. The half-time of caffeine infusion did not change significantly on light roasting but dropped by 40% on more severe treatment and by a further 30% when roasted to scorching. The activation energy also decreased on strong roasting. The kinetic parameters have been correlated with the physical properties of the coffees and with the chemical and physical changes of roasting.

The concentrations of theaflavins and of caffeine in sieved Kapchorua Pekoe Fannings (600–710 μm) have been determined at 80°C with a range of aqueous salt and buffer solutions of ionic strength 0·11 mol dm − 3 and of pH 1·9 to 8·3. The caffeine concentration in the infusions and in the leaf, and its partition constant between swollen leaf and solution, was little affected by the presence of electrolytes or pH changes. The concentration of theaflavins in the infusions was not changed by the addition of salts but was considerably greater in acid media and also, temporarily, in borate buffers. The acid effect was corroborated by experiments with a Ceylon Broken Orange Pekoe blend. The most significant finding was that the concentration of theaflavins in the Kapchorua leaf rose markedly with falling pH while its partition constant decreased. These results show that hydrogen ions liberate extra theaflavin by breaking down the leaf structure and/or the bonding of some theaflavin within the leaf. The effect has significant commercial implications since the market price of tea leaf correlates with its theaflavin content.

Koonsong Broken Pekoe tea leaf was equilibrated with water at 79.5 and 94.0°C using water: leaf ratios from 25:1 to 100:1. The concentrations of theaflavins, thearubigins, and caffeine were measured in the aqueous extracts. Analysis of the results by a simple two-phase model gave the concentrations of the three constituents in the original leaf without the necessity for the usual assumption of complete extraction. The theory also yielded the partition constants (distribution coefficients) of the three constituents between swollen leaf and aqueous solution, and the enthalpy changes for their extraction.

Green and medium-roast Kenyan arabica coffees were ground and sieved, and the 0·85–1·8 mm size fractions partially converted into two water-swollen forms by an appropriate series of treatments. The first form still contained a mix of coffee solubles, the second only caffeine. The rates of caffeine infusion into water at 80°C were then measured for the dry coffee and for the two water-swollen preparations. The caffeine was extracted two to three times faster from the solute-free water-swollen preparation than from the dry material Analysis of the results showed that counterflow of water the swelling of coffee particles, caffeine association with other solubes and physical restraints within the bean matrix all contribute to the low diffusion coefficient of caffeine inside the coffee particles. The behaviour of the green and the medium roast coffees was surprisingly similar.

Reactions in solution can be heterogeneously catalyzed not only by conductance or e. m. f. electrodes but also by precipitates formed during the titration of qliquots. Attempts to follow a Menschutkin reaction by Volhard titrations revealed serious errors caused by the catalysis of a side reaction by insoluble silver salts.

The first-order solvolysis of PhCH2CMe2Cl in 50% v/v EtOH–H2O at 40 °C was found to be inhibited by two types of carbon, Hopkin and Williams activated charcoal and Akzo Ketjenblack E.C. Competitive adsorption experiments indicated that the substrate was adsorbed at its phenyl end. The surface solvolysis was found by a combination of kinetic and adsorption measurements to have a rate constant 1/4 as large as the homogeneous solvolysis and a greater activation energy. The main reason for the slower solvolysis of the adsorbed molecules was preferential adsorption of ethanol on the carbon surface. The adsorption isotherm on charcoal of PhCH2CMe2Cl in 50% v/v EtOH–H2O was also derived from the experimental data.

The cation- and anion-constituent transference numbers of aqueous 0.02M potassium picrate have been measured at 25°C by the moving-boundary method. The large difference (0.014) between the results and the predictions of the Debye-Huckel-Onsager theory, and similar anomalies in the literature, are consistent with the existence of picrate ion dimers (Pi22−) with a formation constant of 4(±2) liter-mole−1.

Summary The rate of leaching of lead from platinized platinum electrodes has been studied, and was found to require the presence of both acid and oxygen. The results indicate that only the lead in the top 2 or 3 atomic layers of the deposit dissolves and that it is largely present there in the form of PbO; most of the dissolution occurs within an hour. The potentials of lead-containing platinized platinum electrodes behave in a non-Nernstian manner in the presence of lead nitrate.

Summary The variation of the mass specific area of platinized platinum with deposition potential from −56 to +594mV (NHE) has been studied at constant mass degrees of platinization (10 and 5 mg cm−2). Maximum mass specific area was obtained at a deposition potential of approximately, 150 mV (NHE) where the average current density of platinization was ca. 27 mA cm−2. At deposition potentials more cathodic than about 140 mV (NHE) a coupled reduction reaction led to the simultaneous evolution of hydrogen. This was accompanied by a sharp drop in mass specific area. The coulombic efficiency of the deposition process passed through a maximum, as predicted1, at a potential of about 120 mV (NHE).

The addition of various semiconducting solids (silicon and metal sulphides) and of two ionically conducting solids [copper(I) and thallium(I) iodides] appeared either to catalyse or inhibit the rate of iodine formation by the reactions [Fe(CN)6]3–+ I– and Fe3++ I– in aqueous solution. The curious effect of CuI on the latter reaction was investigated in more detail. Adsorption experiments demonstrated that [Fe(CN)6]3– and I3– chemically attacked these solids. Their chemical instability in the presence of the oxidants or the reaction mixtures was confirmed by thermodynamic calculations and the observed kinetic effects could then be explained. Only silica and silver(I) iodide did not affect the rates of the reactions; adsorption tests and thermodynamic calculations showed that these two solids were stable under the experimental conditions. The catalytic effect produced by a solid is therefore genuine only if kinetic and/or thermodynamic evidence has established its chemical stability in the reaction mixture.

Die Aktivierungsenergie der Redoxreaktion zwischen Fe" und I′ wird durch Anwesenheit von Holzkohle von 100 kJ/mol auf 24 kJ/mol gesenkt.

The extent of complexation between silver(I) nitrate and triethylamine and tri-n-butylamine in toluene solutions at 23 °C has been determined by infrared spectroscopy. Both amines produced a mixture of monoamine and diamine silver complexes, and the second stepwise formation constants (K2) are reported. Infrared and atomic absorption experiments on solutions saturated with silver nitrate also yielded formal values of the first formation constant (K1) and of the solubility product (Ks) of silver nitrate in toluene. The results showed that Et3N coordinates more strongly with silver(I) in toluene than does Bu3nN. The values of K1 in toluene were found to be 4–5 powers of ten higher than in protic and dipolar aprotic solvents, while the values of K2 were smaller. Structural and solvation reasons are given to explain these results.

Publisher Summary This chapter focuses on the heterogeneous catalysis of solution reactions and discusses various steps that can be rate-determining in heterogeneously catalyzed solution reactions. These mechanisms can be distinguished in practice by the resulting kinetic behavior and by other means that are described in the chapter. The chapter also discusses general stoichiometric and thermodynamic aspects and the specific types of catalyzed reaction (substitution, isomerization, and redox) that have been studied. It focuses on catalysis at the solid/liquid interface. In surface-controlled catalyses, the rate-determining step involves the reaction on the surface of an adsorbed reactant or of a derived species. It is convenient to divide surface reactions in solution into four main categories: unimolecular, racemisation and isotopic exchange, bimolecular, and electron transfer. The chapter also describes kinetic and mechanistic studies that have been carried out on the various types of catalyzed solution reactions.

Citrate-stabilized gold sols catalyse the redox reaction between ferricyanide and thiosulphate ions in aqueous solution. The initial rates of this catalysed reaction have been measured at 25 °C over a wide range of reactant concentrations, while keeping the pH and cation concentrations constant. The kinetic orders of ferricyanide and of thiosulphate were found to be fractions which added up to unity within experimental error. This is consistent with the predictions of the electrochemical theory of such catalysis and points to a catalytic mechanism of electron transfer through the gold particles. Further electrochemical experiments support this conclusion. The magnitude of the heterogeneous rate constant confirms that the catalysis is surface- and not diffusion-controlled.

Computer calculations have been carried out to test how well the variation with concentration of moving boundary transference numbers is described by 3 theories: the standard Fuoss–Onsager transference equation, a transference equation based on the revised 1963 Fuoss–Onsager electrophoretic expression, and the Pitts conductance equation arranged into its transference form. The fit of all three theories leaves something to be desired. Many aqueous transference data are less than adequately described by the theoretical equations, and several sets of non-aqueous results are fitted well only by use of distances of closet approach (a) that are physically unreasonable, even absurd. The revised Fuoss–Onsager theory often produces more sensible a values but at the price of poorer fitting to the experimental data. Suggestions are made as to why the theoretical equations apply less well to transference numbers than they do to conductances.

It has been shown that, contrary to common belief, the more accurate Fuoss and Onsager (1963) electrophoretic term often differs significantly from the term B2√c/(1 +κa) normally employed in the conductance equation. The 1963 term incorporates a complex function F(b) for which an extended and revised numerical table is given.

A theoretical analysis shows that the volume correction for transference numbers obtained by the direct moving boundary method should be calculated using apparent molar volumes and not, as has been the practice hitherto, partial molar volumes. Various other assumptions and formulae for the volume correction that have been proposed give rise to appreciable errors, particularly where high local concentrations develop as around a dissolving anode. One such case was examined experimentally. Application of the apparent molar volume treatment to the indirect moving boundary method reconciles different equations derived in the literature for the corrected Kohlrausch relation.