- 1:57 PM

Effect of Scale on the Hydrodynamics of Low L/D Internal Gaslift Loop Reactor for Anaerobic Digester Applications

Mehul Vesvikar and Muthanna H. Al-Dahhan. Chemical Reaction Engineering Laboratory, Washington University, One Brookings Drive, St Louis, MO 63130

Internal gaslift loop reactors (IGLR) are extensively used in the industry for chemical and biochemical operations. These reactors are equipped with a gas sparger for distributing gas and a concentric draft tube to create liquid circulation. The liquid level to reactor diameter ratio (L/D) is normally greater than two in these reactors. For anaerobic digestion applications of IGLRs, the L/D ratio is close to one. The L/D ratio is one of the important variables affecting the hydrodynamics of the reactor. But not enough information about the hydrodynamics of low L/D ratio gaslift digesters is available in the literature. Thus main objective of this study was to investigate the hydrodynamics of these low L/D ratio gaslift digesters. Advanced non-invasive Computer Automated Radioactive Particle Tracking (CARPT) technique was used for hydrodynamic studies. The process of interest was anaerobic digestion (AD). Anaerobic digestion is a process where biomass (plant or animal waste) is treated by microorganism in absence of oxygen to produce biogas, a mixture of methane and carbon dioxide. AD is important because it reduces land, water, and air pollution and also produces energy in form of biogas. Anaerobic digester (ADr) is a three phase system; the slurry contains a variety of solids in water, which can be lighter or denser than the water. Biogas is produced during the process. In case of gaslift digester, biogas is recirculated to create mixing. Gaslift digester used for CARPT experiments was a laboratory-scale unit, 6 inches in diameter, containing cow manure. The slurry contained 10% solids. Since biogas conatins 60-70% methane and remaining carbon dioxide, its density is similar to that of air at standard conditions. Thus, air was sparged in the digester. The biological activity was stopped to inhibit production of biogas, such that effect of mixing by gas recirculation only can be studied. The gas sparging rate was varied from 1 to 3 lpm; 1 lpm gas flow corresponds to energy input density of 8W/m3. The hydrodynamic variables that were studied using CARPT includes, flow pattern, velocity profiles, volume of dead zones, circulation time distribution, and turbulent parameters such as turbulent stress and kinetic energy and turbulent diffusivity. The effect of geometric variables (size of draft tube and geometry of sparger) and operating variables (gas sparging rate and solids content in slurry) on the above hydrodynamic parameters was also studied. The results showed that the gas sparging rate does not have an appreciable effect on the hydrodynamics, but the effect of sparger geometry and draft tube diameter is significant. Considering the effect of geometry and scale on the hydrodynamics, the CARPT experiments were repeated on a similar pilot-scale digester unit that was 18 inches in diameter. A volumetric scale-up ratio of 25 was applied. The pilot-scale digester was geometrically similar and contained the same slurry that was used in laboratory-scale configurations. The gas sparging rate was adjusted such that the energy input per unit volume was same at both scales. Similar hydrodynamic quantities were studied at this scale. The effect of scale on hydrodynamics was evaluated by comparing the flow patterns, liquid velocities, circulation times and turbulence quantities. The scale has significant effect on the hydrodynamics of gaslift digester. The flow patterns, velocity profiles, and other hydrodynamic parameters are affected considerably with the scale. Thus, same energy input per unit volume cannot be the correct scale-up criteria to obtain same hydrodynamic behavior at different scales. The mixing was quantified in terms of dead zones volume and circulation time distribution. Same mixing intensity was not obtained at different scales with same power input per unit volume.