Desorption of Phenol from Activated Carbon by Hot Water Regeneration. Desorption Isotherms Gorazd Berc ˇ ic ˇ and Albin Pintar Laboratory for Catalysis & Chemical Reaction Engineering, National Institute of Chemistry, P.O. Box 3430, 1001 Ljubljana, Slovenia Janez Levec* Department of Chemical Engineering, University of Ljubljana, P.O. Box 537, 1001 Ljubljana, Slovenia Adsorption of phenol from aqueous solutions at high temperatures on activated carbon was studied. The adsorption capacities were determined by the semicontinuous desorption experi- ments carried out in a fixed bed adsorber, which was operated in liquid-full mode at a pressure of 25 bar and at temperatures up to 190 °C. It was found that in the range of phenol concentrations from 0.005 to 30 g/L the five-parameter Redlich-Petersen adsorption model is equivalent to the six-parameter Fritz-Schlu ¨ nder adsorption model when describing liquid- solid adsorption on the Filtrasorb (F-400) activated carbon. By hot water regeneration, 95% recovery of initial adsorption capacity of activated carbon was obtained. It is also demonstrated that the proposed experimental method for determination of adsorption capacities at high temperatures can be used as an alternative to the traditional method by taking breakthrough curves. Introduction Wastewaters containing relatively low concentrations of toxic or poorly biodegradable organic contaminants cannot be economically treated by a direct oxidation technique such as wet air oxidation. One can use an activated carbon (AC) adsorption step to remove the organics from a diluted wastewater. Activated carbon is capable of adsorbing a broad spectrum of organics from wastewaters; however, once the bed becomes saturated, the exhausted carbon must be regenerated before reused. Regeneration of AC adsorbers which operate in the liquid phase is mostly performed by use of different solvents and salt solutions and by pH swing (Radeke and Hartmann, 1992). When adsorber is used for purification of drinking water, the total reactivation of spent carbon at higher temperatures is carried out in multiple-hearth furnaces or rotary kilns (Jankovska et al., 1991). Regeneration of AC beds by temperature swing is mostly used for adsorbers which operate in the gas phase and is achieved by means of steam, hot inert gases, or direct heating (Boppart, 1995). A technique which combines adsorption and oxidation processes has been reported recently for the treatment of wastewaters containing relatively low concentrations of toxic organic contaminants (Levec and Pintar, 1995). In this process, the organics are removed from the wastewater by adsorption on the activated carbon. Once the carbon bed is saturated with organics, it is regenerated by the temperature swing method, which is routing hot water at temperatures up to 200 °C (and elevated pressure) through the adsorber and oxidation reactor. This procedure desorbs most of the organics, and subsequently accomplishes their destruction by oxidation either in a trickle bed or in a wet air oxidation reactor. Since this was not of great importance in the past, only a few studies of adsorption and desorption at elevated temperatures were made (Bhatia et al., 1990; Ling and Hsu, 1995; Radeke and Hartmann, 1992). It should be pointed out that all these studies were made at ambient pressure and at temperatures below the normal boiling point of water. At elevated temperatures isotherms for adsorption of an organic contaminant from aqueous solutions onto a particular adsorbent can be determined by measuring contaminant concentrations in an experimental setup which can operate in either batch or continuous mode (Smis ˇek and C ˇ erny, 1970). One of the main difficulties arising when carrying out batch experiments at tem- peratures above the normal boiling point of water is the filtration of the suspension. If the conditions during sampling/filtration are not the same as during the adsorption, a new equilibrium is established which consequently leads to wrong results. The adsorption capacity obtained by measuring the breakthrough curves can be affected by transport limitations, especially when larger particles are used. The apparatus dead volume is considerably higher compared to low temperature experiments due to employment of preheater, cooler, and pressure reducer; therefore it must be accounted for. At high temperatures and high concentrations its influence on the breakthrough time cannot be neglected, otherwise the calculated capacities would be too high. The aim of this work is to determine the phenol adsorption/desorption isotherms for activated carbon at temperatures up to 190 °C. The new experimental technique employed is simple and avoids the problems usually encountered in the conventional methods and is particularly advantageous at temperatures and pres- sures above the normal boiling point of water. Once these isotherms are at our disposal, it would be possible to effectively design the oxidation reactor in the ad- vanced adsorption/oxidation wastewater treatment unit. Experimental Section Apparatus. The adsorption isotherm of phenol from aqueous solutions on activated carbon at 25 °C was determined by a standard bottle-point method (Chat- zopoulos et al., 1993). The apparatus for measuring the adsorption isotherms by a semicontinuous technique above 100.0 °C is schematically shown in Figure 1. Properties of the column and the activated carbon as well as the operating conditions employed are sum- marized in Table 1. Before starting an experiment, the * Author to whom correspondence should be addressed. 4619 Ind. Eng. Chem. Res. 1996, 35, 4619-4625 S0888-5885(96)00208-4 CCC: $12.00 © 1996 American Chemical Society