The Development of Empirical Correlations to Understand the Frictional Behavior of Aqueous Biomass Slurry Flows in Vertical Pipes

Highlights The frictional pressure drop correlation of agricultural residue-water slurry flows in vertical pipes is developed. Multiple linear regression with the backward elimination method was used in RStudio to obtain the optimal model. Some regression coefficients differ for different types of biomass feedstocks. The predicted pressure drops agree well with the experimental data within a 95% CI. Empirical models for the onset velocity of drag reduction of different particle sizes of biomass are proposed. Abstract. Large-scale biofuel production at levels equivalent to conventional oil refineries using long-distance pipeline hydro-transport of biomass can be a cleaner alternative to fossil fuels when it comes to economics and traffic congestion associated with the overland transportation of biomass. The transport of aqueous slurries of several saturated mass concentrations (5%-40%) and four particle sizes (from <3.2-19.2 mm) of two types of agricultural residue biomass (ARB) feedstock (corn stover and wheat straw) was studied through a vertical test section of a 29 m long, 50 mm diameter closed circuit pipeline facility, and frictional pressure drops were recorded at different flow rates (0.5-4.3 m s-1). A framework was developed in RStudio (4.0.5) to analyze the experimentally obtained frictional pressure drops of biomass slurries through a multiple linear regression approach using a backward elimination method and Akaike information criterion. An empirical model was proposed to predict slurry frictional pressure drop in terms of slurry velocity, slurry solid mass concentration, particle aspect ratio, and feedstock type. The model satisfactorily predicted the frictional pressure drops of both feedstocks of biomass-water slurry flows through pipes within a 95% confidence interval. The correlations introduced for onset velocities of drag reduction in terms of slurry solid mass concentrations seemed helpful to interpret the transition points of the corresponding slurries in vertical upward flows through pipes. The empirical correlation developed in this research could help select biomass slurry pumps and pipe dimensions when designing a typical long distance pipeline network for biofuel production at the commercial level. Keywords: Agricultural biomass wastes, Frictional loss prediction, Numerical model, Onset velocity correlation, Regression coefficients, Upward pipe flow.