Equilibrium adsorption and hysteresis effect of water vapor onto a series of activated carbon adsorbents

Adsorption technology has been playing a major role in separation process in industries. Adsorption process has proven a very efficient removal and recovery of pollutants from effluent streams. However, studies have shown that the presence of water vapor in gas streams affects the removal efficiency of organic vapors in adsorption processes.;The experiments associated with this research mainly focuses on the assessment of equilibrium adsorption and desorption of water vapor on a series of activated carbon (AC) samples with varied morphology. The objective is to determine water vapor adsorption and the hysteresis effect when using ACs at different temperatures. A gravimetric technique was used to determine the adsorption, desorption and hysteresis of water vapor on microporous adsorbents. Equilibrium adsorption and desorption capacities of water vapor were measured using a symmetrical gravimetric analyzer. AC samples were in the form of activated carbon monolith (ACM), activated carbon fiber cloth (ACFC), and granular activated carbon (GAC).;Equilibrium adsorption and desorption tests were conducted in a controlled environment with water vapor relative pressures ranging from 10% to 90% and at temperatures of 25 (298.15 K), 35°C (303.15 K). and 50°C (323.15 K). Resulting adsorption-desorption isotherm data consisted of relative change of weight of the activated carbon samples with respect to the water vapor relative pressure. Experimental results showed a gradual water uptake with an increase in relative humidity. The adsorption-desorption hysteresis loops at higher temperature (50°C) are constricted compared to the ones obtained at lower temperature (25°C and 35°C). ACFC showed higher adsorption capacities of water vapor than ACM and GAC samples at relative pressures > 0.5. The Duhinin-Radushkevich (DR) and the Qi-Hay-Rood (QHR) equations were used to model the experimental data. The models showed a good agreement with the experimental data. The data however, showed an unsatisfactory tit with Langmuir and Freundlich models.