Christine Roizard

Absorption of sulfur dioxide in N-formylmorpholine: investigations of the kinetics of the liquid phase reaction

Abstract Absorption of sulfur dioxide in N -formylmorpholine results in the reversible formation of 1:1 complex, as shown previously. This paper presents a kinetic investigation of the liquid-phase reaction. This reaction was assumed to be a fast process and this work was aimed at the verification of this hypothesis. Absorption of sulfur dioxide either as pure gas or diluted into an inert, was carried out at 25°C in a stirred vessel with a flat interface. The experimental data were modelled under the assumption of a fast reversible formation of a SO 2 –NFM complex. The comparison between experiment and model calculations indicated that the liquid-phase reaction was instantaneous or at least very fast, with Ha number over 100.

Mass transfer with chemical reaction: the slow reaction regime revisited

A simple calculation procedure for the case of absorption with chemical reaction in case of the slow reaction regimes is presented. It is based on film theory and concerns reactors open to the liquid and to the gas. The influence of the reactor type and of reaction orders is shown.

Behavior of fine particles in the vicinity of a gas bubble in a stagnant and a moving fluid

Abstract In order to explain the mass transfer enhancement factor obtained in slurry reactors, a static experimental system and a dynamic one are used to observe the behavior of fine activated carbon particles in the vicinity of a gas bubble. We first investigate the possible adhesion of fine particles in a static system. A gas bubble formed at a needle point, is maintained in a beaker filled with the liquid phase, while injections of concentrated activated carbon particles solutions are made with a syringe over the gas bubble. Visual observation of the bubble show that there is adhesion of particles on the gas bubble, which can be totally covered by activated carbon. This adhesion depends on the ionic strength and on the pH of the liquid solution. Further, we investigated the case of a static gas bubble in a hydraulic tunnel. Use of a movie camera allows to visualize the particles trajectories around the gas bubble. It has been found that there is no adhesion of the fine particles used in this study on the gas bubble when the liquid–solid solution is flowing, the particles follow the streamlines of the fluid; for the highest values of the fluid velocity investigated, we observed a small helicoidal vortex of the carbon particles behind the gas bubble. No influence of the ionic strength, the pH nor of the particle diameter can be pointed out. The activated carbon particles behavior is obviously controlled by hydrodynamic forces in a dynamic liquid system. Observed phenomena are fundamentally different from the static case. Therefore, the enhancement factors obtained using these particles cannot be explained by particle adhesion at the interface.

Kinetics of sulfur dioxide oxidation in slurries of activated carbon and concentrated sulfuric acid

The oxidation of sulfur dioxide (SO2) in the presence of activated carbon particles has been studied experimentally in slurries of both pure water and sulfuric acid solutions. In the water slurries, the oxidation of absorbed molecular SO2 is accompanied by parallel oxidation of HSO3− ions produced by SO2 dissociation. In acid slurries of pH less than one, the dissociation of molecular SO2 is inhibited and only direct oxidation of molecular SO2 is observed. Experiments in up to 35 wt% sulfuric acid solutions were carried out in slurries of fine coconut activated carbon particles. The rate of molecular SO2 oxidation has been correlated both to an empirical power-law model and to a heterogeneous Langmuir-Hinshelwood type model: Rpl = 0.21 m10.69m20.19m30.72, Rin = 3.81 108m1m22(1 + 8.46 103m1 + 4.53 103m2 + 1.20m3)3 The Langmuir-Hinshelwood model corresponds to a kinetic mechanism limited by the surface reaction between adsorbed, non-dissociated O2 and two adsorbed molecular SO2 molecules.

Gas absorption in a centrifugal reactor: hydrodynamics

Abstract For the removal of SO 2 and/or NO x using an oxidizing liquid solution, absorbers with a high mass transfer efficiency are required. Here, a centrifugal gas—liquid absorber has been designed and studied; the chosen device is a rotor with blades which induce the formation of a thin liquid film on the reactor wall. The operating parameters that can be varied are the rotation speed, the back pressure at the liquid outlet and the gas and liquid flow rates. Three distinct flow regimes can be visualized, depending on the values of the parameters. The hydrodynamic regimes vary from thin liquid film containing dispersed gas bubbles with large gas holdup to thick films containing virtually no gas. Residence time distributions of the liquid phase allow to measure the liquid film volume (liquid holdup) in the reactor and the dispersion coefficient. The liquid volume increases with back pressure but decreases with increasing rotation speed. The Peclet number decreases with increasing back pressure. In the investigated range, no influence of the liquid and gas flow rates was found on either of the above two factors. Furthermore, the liquid volume and the Peclet number can easily be correlated to the Euler number, which is the ratio of the relative back pressure to the centrifugal pressure. A hydrodynamic model based on a pressure balance in the reactor leads to the expression of the liquid film thickness along the height of the blade as a function of Euler number and Δ p *, the dimensionless ratio of the relative back pressure to the gravity. These two dimensionless numbers depend only on the operating parameters. The angular liquid velocity was supposed to be proportional to the rotor angular velocity: the shift coefficient k is determined from the experimental values of the liquid holdup. Results show that k can be correlated to the Euler number. Using these results, calculated liquid film thickness profiles are presented and discussed.

Screening method for solvent selection used in tar removal by the absorption process

The aim of this paper is the study of the treatment of flue gas issued from a process of biomass gasification in fluidized bed. The flue gas contains tar which should be selectively removed from the fuel components of interest (e.g. H2, CO and light hydrocarbons) to avoid condensation and deposits in internal combustion engine. The chosen flue gas treatment is the gas–liquid absorption using solvents, which present specific physicochemical properties (e.g. solubility, viscosity, volatility and chemical and thermal stability) in order to optimize the unit on energetic, technico-economic and environmental criteria. The rational choice of the proper solvent is essential for solving the tar issue. The preselection of the solvents is made using a Hansen parameter in order to evaluate the tar solubility and the saturation vapour pressure of the solvent is obtained using Antoine law. Among the nine families of screened solvents (alcohols, amines, ketones, halogenates, ethers, esters, hydrocarbons, sulphured and chlorinates), acids methyl esters arise as solvents of interest. Methyl oleate has then been selected and studied furthermore. Experimental liquid–vapour equilibrium data using bubbling point and absorption cell measurements and theoretical results obtained by the UNIFAC-Dortmund model confirm the high potential of this solvent and the good agreement between experimental and theoretical results.