Rates of oxidation of dilute, aqueous solutions of acetic acid were measured at 252O to 286OC and 67 atm in liquid full and trickle-bed reactors packed with ferric oxide catalyst particles. Predicted global rates and conversions indicated that localized vaporization and liquid channeling could affect trickle-bed performance.
JANEZ LEVEC and J. M. SMITH University of CaliforniaDavis, California
SCOPEOxidation of dilute aqueous solutions of organic pollutants by using solid catalysts offers an alternate process to biological oxidation as a means of purifying liquid water. Hamilton et al. (1969) described such a process using a magnesia catalyst at 100OC. Sadana and Katzer (1974) have reported conversions of phenol to carbon dioxide of as high as using a copper oxide catalyst at 145OC and 17 atm pressure. Not enough oxygen can be dissolved in water to give high conversions and still operate the reactor liquid full. Therefore, a gas-liquidsolid system, such as a trickle-bed reactor where gas (air or oxygen) and liquid flow concurrently downwards through the catalyst bed, is needed. Alternately, a slurry reactor with gas bubbling through the slurry could be used. In these three-phase systems, finite mass transfer rates between gas and liquid as well as between liquid and solid can affect reactor performance. Also, inefficient contact areas between the phases, due, for example, to channeling, can influence results. Goto and Smith (1975a, b) studied the oxidation of formic acid in water in a tricklebed reactor using a CuO.ZnO catalyst operated at 40 atm and 200° to 240OC. Elevated pressure was necessary to maintain water in the liquid phase. Based on a model which accounted for gas-to-liquid, liquid-to-particle, and intraparticle diffusion, they were able to predict observed behavior reasonably well from intrinsic kinetics and effectiveness factors previously established (Baldi et al., 1974) from data obtained in liquid full operation. Formic acid is easily oxidized. Hence, relatively low temperatures could be used so that the vapor pressure of the liquid was low. Because the intrinsic rate was large, short catalyst beds and high liquid rates were satisfactory, perhaps avoiding uncertainties due to inefficient contacting between phases. Another published study in trickle beds, which accounts for mass transfer effects, is that of Hartman and Coughlin (1972) on the oxidation of sulfur dioxide in a bed of activated carbon. These authors also found that liquid phase and intraparticle diffusion adversely affected the conversion to sulfur trioxide. Investigations such as these which attempt to account for all mass transport factors are rare. In a useful review, Satterfield (1975) has summarized current design concepts and indicated the need for additional kinetics and transport research. The effect of efficiency of liquid particle contacting on reactor performance has been investigated by Satterfield and May (1972) and by Sedricks and Kenney (1973).As a further study of trickle-bed performance, this paper reports results f...