Yukio Yoneda

Reduction-oxidation and catalytic properties of La1 − xSrxCoO3

Catalytic oxidations of carbon monoxide, propane, and methanol have been investigated over perovskite-type mixed oxides (La1 − xSrxCoO3 − δ, x = 0, 0.1, 0.2, 0.4, and 0.6) by the use of flow and pulse methods. The reduction-oxidation properties as well as nonstoichiometry (δ) and desorptivity of oxygen were also measured. These properties were correlated with the catalytic activity with emphasis on the effect of the Sr substitution. Reducibility (or oxidizing ability) of these catalysts greatly increased with the extent of Sr substitution (x), while the reoxidation became much slower with increasing x. The catalytic activities increased with x when x was low, but decreased at higher x values. This variation of activity was well explained on the basis of a redox-like mechanism, taking account of the reduction-oxidation properties of catalyst. The oxidation state of the catalysts during the steady-state catalytic oxidation of carbon monoxide varied with x and COO2 ratio also in accordance with the above explanation.

A new dual parameter scale for the strength of lewis acids and bases with the evaluation of their softness

Abstract A dual parameter scale has been proposed to represent the acidity of metal ions as Lewis acids. One parameter, X , is related to the electronegativity of the ion and is derived from the equation: (10 X) 1 2 = x i = x M 0 + (ΣI n ) sol1 2 , where x i and x M 0 are the electronegativities of the metal ion and the neutral metal atom, respectively, and I n is the n -th ionization potential ( M ( n −1) + → M n + ). Another parameter, Y , is calculated by Y = 10( I n / I n +1 )( r i /√ n ), where r i is the ionic radius of the metal ion and n is its formal charge. This parameter, Y , can be considered to express the tendency of a metal ion to form a dative pi-bond and consequently should correspond to its “softness”. The instability constant of a metal ion complex, K , can be described by these parameters as follows, p K = log K = αX + βY + γ , where α and β dual basicity parameters of the ligand corresponding to X and Y , respectively, and γ is a constant determined for each ligand. Some examples of the applications of these parameters are given.

Base-strength distribution studies of solid-base surfaces

Abstract A method has been developed for the determination of the base-strength distribution of solid surfaces. The present method consists of the titration of solids suspended in cyclohexane with benzoic acid, using a series of H − indicators. The base-strength distribution of solid surfaces has been measured by this method for alkaline-earth oxides and some other solid bases. Alkaline-earth oxides increased remarkably in base strength upon heat treatment in vacuum. The base strength of these oxides proved to decrease in the order SrO( H − ⩾ 26.5) ≈ CaO( H − ⩾ 26.5) > MgO( H − = 18.4 to 26.5), in accordance with the observation by Krylov et al. and also with the order expected from the partial negative charge of combined oxygen anions in these oxide surfaces. A distinct distribution of base strength was observed on these oxide surfaces. No basic sites having an H − equal to or higher than 12.2 were observed on the surfaces of ZnO, ZrO 2 , ThO 2 , or Na 2 CO 3 . The catalytic activity and selectivity of alkaline-earth oxides in the elimination of hydrogen chloride from 1,1,2-trichloroethane were quantitatively interpreted from the observed base-strength distribution.

Catalysis by heteropoly compounds. VI. The role of the bulk acid sites in catalytic reactions over NaxH3 − xPW12O40

Abstract Several acid-catalyzed reactions were investigated over partial Na salts of 12-tungstophosphoric acid (Na x H 3 − x PW 12 O 40 ). Thermal desorption of pyridine absorbed in the bulk combined with in-frared measurement demonstrated that the Na salts were purely protonic acids and their “bulk” acidity (strength and amount) decreased as the Na content increased. On the basis of the absorptivity of reactants into the bulk of catalysts and the activity patterns for several acid-catalyzed reactions, it has been proposed that catalytic reactions are divided into “bulk-type” and “surface-type” reactions. In the case of the “bulk-type” reactions such as dehydration of 2-propanol or conversion of methanol to hydrocarbons, the reactants were readily absorbed in the bulk and the catalytic activities for these reactions were well correlated with the bulk acidity of the Na salts measured by the thermal desorption of pyridine. It was very probable that these reactions proceeded mainly in the bulk. On the other hand, reactions of hydrocarbons like butene and cumene, which were adsorbed only on the surface, take place only on the surface (“surface-type” reactions).

Catalysis by heteropoly compounds: IV. Oxidation of methacrolein to methacrylic acid over 12-molybdophosphoric acid

Abstract Oxidation of methacrolein over heteropoly acid catalysts (mainly 12-molybdophosphoric acid) at 300 °C was investigated by use of a flow method. Primary (Keggin) and secondary structures of the catalyst were rather stable upon heat treatment at 250–350 °C, and reproducible data of the oxidation reaction were obtained by the pretreatment at 350 °C in a N 2 + O 2 + H 2 O stream. Effects of oxygen in the feed gas demonstrated that oxygen atoms of polyanions or steam were directly involved in the reaction and the reduced polyanions were reoxidized by gaseous oxygen. Presence of steam had a remarkable effect both on the rate and selectivity. Rate equation, ESR study, and the comparison between the catalytic oxidation (flow method) and noncatalytic oxidation (pulse method) indicated that the reduction of catalysts was rate-determining in a redox mechanism and the catalysts were in a highly oxidized state under the present reaction conditions. Effects of metal salt formation and silica support were also examined. On the basis of these results, a possible reaction mechanism with an ester- or diol-type intermediate has been proposed.

Contrast between nickel and platinum catalysts in hydrogenolysis of saturated hydrocarbons

Abstract In order to contrast the reaction mechanism of hydrogenolysis on nickel catalysts with that on platinum catalysts, the detailed analysis of initial reaction products in hydrogenolysis of five hexane isomers and methylcyclopentane was ensured by means of a pulse technique, using hydrogen as carrier gas for a gas Chromatographic microreactor. The hydrogenolytic products from reactants on nickel and platinum catalysts showed a very interesting contrast with each other at low conversion. Analogous to other reactions in hydrogen atmosphere, the reaction intermediates in hydrogenolysis are presumed to be normal alkyls -CH 2 (CH 2 2) n CH 3 on nickel catalysts, which are selectively hydrocracked owing to the successive α-scission to give methane as a main product. On the contrary, a carbonium ion mechanism has been proposed for platinum catalysts to interpret both the characteristic distribution of the initial hydrogenolytic products and the considerable skeletal isomerization during hydrogenolysis. Since heterolytic splitting of a carbon-hydrogen bond of saturated hydrocarbon gives a carbonium ion, the carbon-hydrogen heterolysis at the stage of adsorption on platinum catalysts may be due to the large stability of platinum-hydride coordination, which cannot be expected for the nickel-hydride case. The contrast between nickel and platinum catalysts in hydrogenolysis of saturated hydrocarbons is thus correlated with the nature of the carbon-hydrogen splitting, homolytic or heterolytic, at the stage of adsorption, which is understandable only in terms of the softness of the catalyst metals.

Catalysis by heteropoly compounds. V: The reduction mechanism of H3PMo12O40

Abstract The reduction mechanism of heteropoly compounds of molybdenum and tungsten as well as the behavior of water contained in them was investigated. Stoichiometry of reduction by H 2 and reoxidation by O 2 , together with the variation of ir and ESR spectra upon reduction, confirmed that the reduction of H 3 PMO 12 O 40 proceeded in two steps. In the first step, a hydrogen molecule dissociates into protons and electrons, the latter being trapped by molybdenum atoms. Both protons and electrons migrate rapidly throughout the whole bulk. In the second step, protons react with bridging oxygen atoms of heteropoly anion and are evolved as water. The anion structure is maintained in the first step, while it is partially destroyed in the second step.

The classification of metal catalysts in hydrogenolysis of hexane isomers

Abstract The hydrogenolysis of hexane isomers has been studied on various supported metal catalysts (Fe, Co, Ni, Pd, and Pt) in the presence of hydrogen. The previous conclusion that there are two different mechanisms in the hydrogenolysis of saturated hydrocarbons on nickel and platinum catalysts has been found to hold generally on the above metal catalysts. From the product patterns in catalytic hydrogenolysis of hexane isomers, the catalysts could be classified into two groups, one containing Fe, Co, and Ni and the other containing Pd and Pt. This distinct difference cannot be explained by an intensive factor, but must be attributed to some more intrinsic property of these metals, because this contrast between the two groups is generally observed among several metal-catalyzed reactions. The effect of the support on the hydrogenolysis of hexane isomers was also investigated in the case of platinum and nickel catalysts. It was found that the differences observed between nickel and platinum were preserved on all supports used.

Catalysis by heteropoly compounds x. Synthesis of lower olefins by conversion of dimethyl ether over 12-tungstophosphates

Conversion of dimethyl ether and methanol into hydrocarbons was studied over several salts of 12-tungstophosphoric acid, H3PW12O40, at 563 K by a flow method. For acid salts of Na, the catalytic activity decreased in parallel with the acidity as the Na content increased (NaxH3−xPW12O40, x = 0–3). However, the carbon number distribution and the olefin to alkane ratio of the product hydrocarbons were almost independent of the Na content. These results indicate that the acidity determines the activity but its effect on the selectivity is not significant. On the other hand, when the salts of organic bases, Cs and NH3, which showed a low ability to absorb dimethyl ether, were used, the selectivity towards lower olefins was much improved. It was concluded that the olefin to alkane ratios were not correlated with the acidity, but with the ability to absorb dimethyl ether; the ratios increased markedly as this ability decreased. This was explained on the basis of “pseudo-liquid” behaviour of heteropoly compounds.

Linear free energy relationships in heterogeneous catalysis: I. Dealkylation of alkylbenzenes on cracking catalysts

Abstract It was attempted to survey the applicability of LFER (Linear Free Energy Relationships) in heterogeneous catalysis to dealkylation of alkylbenzenes. The principle of LFER in heterogeneous catalysis is to find linear relationships between the rate data and some numerical variables which represent either reactivity of reactants or catalytic activity of heterogeneous catalysts. The microcatalytic gas chromatographic technique was used to measure the reaction rates at 400 °C on cracking catalysts. As for monoalkyl-benzenes, the logarithms of the rate constants hold a linear relationship with ΔHc+(R1) for about two and one-half orders of magnitude in the rate constants. This relationship can be formulated by the following equation: log k i ( R 1 ) = log k i (0) − γ′ i Δ H c + ( R 1 ) 2.303RT where ki(0) is a characteristic value dependent upon both the catalyst i and the kind of reaction (dealkylation in this case), but not upon the reactants, and ΔHc+(R1) is the enthalpy change for the hydride abstraction from corresponding paraffins. By this equation it is meant that LFER is applicable to solid acid catalysis. Therefore, the reaction rate constants can be estimated from the values of ΔHc+(R1) and a few measured values such as ki(0) and γ′i. The effects of the second substituent group on dealkylation rate are also discussed by the Hammett law. Linear relationships between the logarithms of the rate constants and ΔHc+(R1) or σ(R2) are applied to the work previously published. In these cases, fine linear relationships are also obtained.