B.F.M. Kuster

The influence of the initial and catalyst concentrations on the dehydration of d-fructose

Abstract The dehydration at 95° of d -fructose (0.25 m -1.0 m ) to 5-hydroxymethyl-2-furaldehyde (HMF) and the rehydration of HMF (0.25-1.0 m ) to levulinic and formic acids in 0.5-2 m HCl has been studied. The conversion rate of d -fructose was proportional to the Hammett activity. The acidity had a smaller influence on the conversion rate of HMF, although it was not proportional to the catalyst concentration. The rehydration of HMF was faster in the presence of d -fructose. The yield of levulinic acid was independent of the catalyst concentration, but was lower at higher initial concentrations of d -fructose and HMF, and a kinetic model has been derived. The formation of humin was of an overall order 1.3 in an intermediate between d -fructose and HMF, and of an order 1.7 in an intermediate between HMF and levulinic acid.

The influence of pH and weak-acid anions on the dehydration of d-fructose☆

Abstract The influence of the pH (1–6) on the rates and yields in the dehydration of d -fructose to 5-hydroxymethyl-2-furaldehyde (HMF) and the rehydration of HMF to levulinic and formic acids at 175° has been studied by using a stirred tank-reactor. The conversion rate of d -fructose passes through a minimum at pH 3.1, whereas at pH τ 3.9 no formation of HMF occurred and at pH τ 2.7 no formation of levulinic acid occurred. Isomerisation to d -glucose is observed at pH τ4.5. When a weak-acid anion, which functions as a base catalyst, is present at pH 3, the yield of HMF is lowered and isomerisation occurs.

The influence of water concentration on the dehydration of d-fructose

Abstract The influence of water concentration on the rates and yields in the dehydration of d -fructose to 5-hydroxymethyl-2-furaldehyde (HMF) and in the rehydration of HMF to levulinic and formic acids at 95° has been studied by the addition of polyethylene glycol-600 (PG-600) to the reaction mixtures. Decrease of the water concentration highly increased the conversion rate of d -fructose, whereas that of HMF was slightly decreased, resulting in higher maximum concentrations of HMF from d -fructose. As compared with reactions in water, yields of levulinic acid were only slightly decreased at 40 vol. % PG-600 and halved at 70% PG-600.

The effect of bismuth on the selective oxidation of lactose on supported palladium catalysts.

The selective oxidation of lactose by molecular oxygen has been studied in a batch reactor containing an aqueous slurry of 0.5 kmol m-1 reactant and 1.0 kg m-3 catalyst. The in situ Bi promotion of a commercial Pd-C catalyst resulted in 100% selectivity to sodium lactobionate up to conversions of 95% in the pH range 7-10 and at temperatures up to 333 K. Performing the reaction under such conditions that the oxygen transfer to the liquid phase was rate-controlling allowed the production of sodium lactobionate in high yields in approximately 1 h. A maximum initial reaction-rate of 0.47 mol kg-1 s-1 was found at a molar Bi to Pd ratio of 0.50-0.67. Fifteen batches of lactose were oxidized with the same charge of catalyst without significant loss in initial activity or selectivity. Such other aldoses as maltose, glucose, and galactose could be oxidized analogously with similar selectivities.

Analytical Procedures for Studying the Dehydration of D-Fructose

Abstract In order to study the kinetics of the dehydration of D-fructose, procedures for the qnantitation of fructose and its dehydration products, S-hydroxymethyI-2W furaldehyde (HMF), Ievulinic acid, and a “humin”, were developed. For many reaction conditions, these compounds, together with soluble polymers (up to 15%) that are humin precursors, account for at least 98% of the amount of initial D-fructose. Fructose, HMF, and levulinic acid were determined by g.l.c. of their O -trimethylsilyl derivatives. U.v. absorption and titration could also be used for the determination of HMF and levulinic acid. Humin was determined gravimetrically.

The alkaline anthraquinone-2-sulfonate-H2O2-catalyzed oxidative degradation of lactose : an improved Spengler-Pfannenstiel oxidation

The alkali-catalyzed oxidative degradation of lactose (1) to potassium O-beta-D-galactopyranosyl-(1----3)-D-arabinonate (2) has been studied and compared with that of D-glucose to D-arabinonate and D-galactose to D-lyxonate. A mechanism for the degradation of 1 catalyzed by alkali only is presented and discussed, taking into consideration the main reaction products. Increasing the reaction temperature from 293 to 318 K resulted in a drastic decrease of the selectivity for 2. Increasing the oxygen pressure from 1 to 5 bar did not significantly influence the selectivity. The overall reaction kinetics followed first-order behavior with respect to lactose, D-glucose, or D-galactose. The simultaneous addition of catalytic, equimolar amounts of sodium 2-anthraquinonemonosulfonate and H2O2 showed a pronounced effect on the selectivity. A reaction mechanism for this type of alkali-catalyzed oxidative degradation of carbohydrates is presented and discussed. Lactose could be oxidized up to almost complete conversion with a selectivity of 90-95% (mol/mol), whereas D-glucose was oxidized to D-arabinonate with a selectivity of 98%. This increased selectivity was maintained at temperatures from 293 up to 323 K, allowing a reduction of the batch time necessary for almost complete conversion from 50 to 1.5 h. The overall reaction kinetics still followed first-order behavior with respect to lactose, D-glucose or D-galactose. The apparent activation energy amounted to 114 +/- 2 kJ mol-1 for lactose, to 109 +/- 2 kJ mol-1 for D-glucose, and to 104 +/- 9 kJ mol-1 for D-galactose.

Deactivation of carbon-supported platinum catalysts during oxidations in aqueous media

The platinum catalyzed selective oxidation reactions of ethanol en methyl-α-D glucopyranoside (MGP) were studied in combination with catalyst characterization using Transmission Electron Microscopy (TEM), CO chemisorpticn, cyclic voltammetry and open-circuit potential measurements. Some evidence for Catalyst Deactivation by over-oxidation, leaching, platinum particle growth and site coverage is presented. The absence of deactivation observed for ethanol can be attributed to the higher reducing ability of ethanol.

Oxidation of methyl and n-octyl alpha-D-glucopyranoside over graphite-supported platinum catalysts: effect of the alkyl substituent on activity and selectivity

The oxidation of methyl and n-octyl Ol-D-glucopyranoside to methyl and n-octyl Ce-D-glu- copyranosiduronate with molecular oxygen over a graphite-supported platinum catalyst was investigated. An increase of the length of the n-alkyl substituent from methyl to n-octyl resulted in a ten-fold decrease of the catalyst activity and an increase of the selectivity at pH 8.0 and 323 K. The selectivity decreased with increasing pH. The lower activity for a longer n-alkyl substituent is attributed to steric effects upon adsorption on the platinum surface and not to internal diffusion limitations. A tentative reaction scheme is presented, which describes the formation of side products through oxidation of secondary hydroxyl groups, ring cleavage and hydrolysis. Major side products are mono- and di-carboxylates with 2, 4, and 6 carbon atoms and mono-carboxylates, resulting from the oxidation of the alkyl substituent. C-C-Bond cleavage mainly occurs between @2 and C-3 or C-4 and C-5, the former being less important for a longer alkyl substituent. The higher selectivity for a longer alkyl substituent is attributed to its protecting ability against hydrolysis and the exposition of neighboring hydroxyl groups to the platinum surface.

The Selective Oxidation Of Methyl-alpha;-D-Glucoside On A Carbon Supported Pt Catalyst

Abstract The selective oxidation of methyl-α-D-glucoside to methyl-α-D-glucosiduronate by oxygen using active carbon supported Pt was studied in a three phase stirred tank reactor. The temperature was varied from 293-313 K, the pH from 6-10, the oxygen partial pressure from 5.0 10 3 to 1.0 10 5 Pa at a constant pressure of 1 10 5 Pa, the methyl-α-d-glucoside concentration from 50 to 1000 mol m −3 and the catalyst concentration from 1-5 kg m −3 . The conversion ranged from 0.02 to 0.10. At this conversion range methyl-α-d-glucosiduronate was obtained with a selectivity of 100%. At the investigated reaction conditions the initial reaction rates were free from mass and heat transfer limitations. Methyl-α-d-glucosedialdehyde was found to be a reactive intermediate. The data could be described by a reaction rate equation of the Langmuir-Hinshelwood type, corresponding to a rate-determining step on the Pt surface involving two active sites. No distinction was made between the sites of chemisorption for oxygen and methyl-α-d-glucoside. The rate equation was based on a reaction sequence with two reaction paths, the first involving the adsorbed methyl-α-d-glucoside and dominating at low pH, the second involving the methyl-α-d-glucoside anion and dominating at high pH. The adsorption of methyl-α-d-glucosedialdehyde and of methyl-α-d-glucosiduronate was found to be negligible. The surface was mainly covered with oxygen at low pH and with both oxygen and the methyl-α-d-glucoside anion at high pH.

The isomerisation of lactose to lactulose catalysed by alkaline ion-exchangers

Summary Lactulose is industrially produced by the homogeneous alkali catalysed isomerisation of lactose. The use of heterogeneous alkaline ion-exchangers as catalyst can result in a simpler overall process and in higher selectivities. The effect of the type of ion-exchanger, particle diameter, temperature and sugar concentration on the kinetics has been studied. A kinetic model is given which can be used to describe the experimental results. Loss of selectivity can occur due to a limiting diffusion rate. High sugar concentrations are beneficial for a high yield. Some proposals are given for process improvements.