EFFECT OF LIQUID FLOW RATE AND BED VOID FRACTION ON THE HYDRODYNAMICS AND PERFORMANCE OF A TRICKLE BED BIOREACTOR
Advancing the chemical engineering fundamentals
Transport Phenomena in Porous/Granular Media - II (T2-7b)
Keywords: trickle-bed biorreactor, hydrodynamics, performance, RTD
Trickle-bed bioreactors (TBB) have shown to be capable of oxidizing various low-concentration volatile pollutants. Elimination capacity in TBB is determined by the microorganisms kinetic properties and the hydrodynamic characteristics of the flowing phases. In this paper we report the effects of superficial liquid mass flow rate and bed void fraction (e) on the hydrodynamic parameters gas-liquid pressure drop (dP/dz), total and dynamic liquid hold-up, biofilm wetting efficiency, and residence time distribution (RTD), in an isopropyl alcohol (IPA)-removing TBB. Although these variables are known to influence TBB behaviour(1) , they have been scarcely studied.
Experiments were carried out in an acrylic TBB (0.143 m id.×1.6 m of total packed length of 316 SS 1.0 in Pall rings with e=0.95). An acclimatized microbial consortium was used to remove IPA. The superficial liquid mass flow rates used were typical of the trickle flow regime: 6.8, 9.8, 11.8, y 13.8 kg/m2-s, while superficial gas mass flow rate was kept at 0.063 kg/m2-s, with an IPA average inlet load of 180 g/m3-h. Sets of experiments were performed at bed void fractions of 0.95, 0.86, 0.7, 0.6, and 0.44, corresponding to different stages of biofilm growth. During each set of experiments bed void fraction was kept constant by reducing the nitrogen source in the recirculating liquid.
Results show that (dP/dz), total and dynamic liquid hold-up,and biofilm wetting efficiency increased with increasing liquid flow rates and with diminishing bed void fraction. At constant e, increased liquid flow rate had a small effect on the IPA degradation rate, although it increased its conversion to CO2, the other by- product being acetone. The best TBB performance was obtained at e=0.6, with an 88 % of IPA removal and 67.3 % mineralization. On the other hand, at e=0.44, the TBB turned anaerobic and there was a dramatic reduction in IPA removal rate and mineralization, with methane being the main product. Results may be explained in terms of the amount of biomass present, wetting efficiency and the available area for liquid-biofilm contact. A sufficient amount of biomass is necessary for efficient IPA removal, but larger amounts result in reduced liquid-biofilm contact and poor performance. A high biomass content in the TBB causes a high pressure drop, and a considerable reduction of liquid-biofilm and gas-biofilm contact, thus reducing IPA and oxygen transfer.
RTD curves show main peaks around the observed mean residence time main, several smaller peaks, and tailing; these features are more pronounced as bed void fraction diminishes. The RTD curves imply that there are stagnant regions, internal recirculation and channeling within the TBB. The dispersion model gives a fair fit for the experiments at e=0.6, but fails for the other bed void fractions. This suggests that the usual plug-flow assumption used in TBB modelling if far from being fulfilled and that a two-parameter model may represent better the liquid flow.
---------------------------
(1) Trejo-Aguilar, G, Revah, S. Lobo, R.2005. Chem. Eng. J. 113, 145-152
See the full pdf manuscript of the abstract.
Presented Tuesday 18, 16:20 to 16:40, in session Transport Phenomena in Porous/Granular Media - II (T2-7b).
