(110g) Computational Modeling of a Biomass Screw-Feeder with Compressible Non-Newtonian Rheology

Lignocellulosic biomass, such as agricultural waste and forestry byproducts, can be a carbon-neutral source of energy and materials, providing an alternative to fossil sources. Handling of such feedstocks with high variability is critical to efficient biochemical or thermochemical conversion. Conveying of biomass into high pressure pretreatment reactors using screw-feeders is one of the important feedstock handling steps. Mechanical failure and blockage in screw-feeders are among the difficulties that biorefineries face during operation. The complex properties of biomass feedstock, i.e. moisture content, insoluble solids concentration, and non-ideal physical properties, make the blockage prediction very challenging.

The focus of this study is to develop a computational-fluid-dynamics (CFD) model that can predict the torque requirements in these screw-feeders by including the complex rheological and compressive characteristics of biomass feedstock. As a result, we can improve the mechanical design and avoid operating conditions that lead to blockage and failure. The screw-feeder studied in this work is 1.7 m long and is enclosed inside a grooved housing with tight clearance ~ 1 mm. Both the screw and the housing are tapered with outer diameter decreasing from 0.53 m to 0.4 m. A conical back pressure damper at the housing exit assures that the biomass is compressed to the required pressure before moving from the feeder to the pretreatment reactor. Our model is validated against torque measurements from experiments that are performed for various screw speeds and feeding rates.

In order to model this screw-feeder, we have developed a customized solver in OpenFOAM [1], an open-source CFD library. The biomass feedstock is modeled as a single-phase compressible mixture whose density varies with pressure, based on a feedstock dependent equation-of-state, derived from a recent study by Duncan et al. [2]. This is a reasonable approximation because the biomass particles are densely packed in the screw-feeder during compression. The transport equations for mass and momentum are solved using a pressure-based low Mach algorithm. The rheology of the biomass is considered using two different non-Newtonian viscosity models; density-dependent Cross model [3] and a Bingham model [4] with density-dependent yield stress. The parameters for these models are obtained from rheology experiments that characterize yield-stress and strain-rates on a small-scale screw extruder. The screw-feeder torque is calculated for different rotation speeds using both of the proposed constitutive models. Finally, our modeling results are compared with our experimental torque data for model validation.

References:

[1] Weller, Henry G., Gavin Tabor, Hrvoje Jasak, and Christer Fureby. "A tensorial approach to computational continuum mechanics using object-oriented techniques." Computers in physics 12, no. 6 (1998): 620-631.

[2] Duncan, Joshua C., Anaram Shahravan, Joseph R. Samaniuk, Thatcher W. Root, Michael D. Graham, Daniel J. Klingenberg, C. Tim Scott, Keith J. Bourne, and Roland Gleisner. "Pressure-driven flow of lignocellulosic biomass: A compressible Bingham fluid." Journal of Rheology 62, no. 3 (2018): 801-815.

[3] Cross, M. M. (1965). Rheoogy of non-Newtonian fluids: A new flow equation for pseudoplastic systems. Journal of Colloid Science, 20(5), 417–437.

[4] Bingham, E.C. (1916). "An Investigation of the Laws of Plastic Flow". Bulletin of the Bureau of Standards. 13 (2): 309–353.