@article{Ngamaramvaranggul_Thongjub_2020, title={Simulation of Die-swell Flow for Wet Powder Mass Extrusion in Pharmaceutical Process}, volume={13}, url={https://ph02.tci-thaijo.org/index.php/ijast/article/view/242292}, abstractNote={<p>The simulation of extrusion spheronization in pharmaceutical industry is constructed to develop drug product. This process has four steps: mixing, extrusion, spheronization and drying. The mixing combines water and powder together with high shear until it creates strong bonds to gather powder particles in liquid solution. The work is focused on extrusion of wet powder masses, which can be classified as non-Newtonian fluid. The continuous creeping flow motion is explained in terms of the Navier-stokes equation and the rheological behavior is represented by Oldroyd-B constitutive model. The solution is solved with numerical scheme through the semi-implicit Taylor-Galerkin/pressure-correction finite element method in two-dimensional axisymmetric system under the conditions of isothermal, incompressible, laminar flows. In addition, the velocity gradient recovery and the streamline-upwind/Petrov-Galerkin schemes are applied to improve the convergence of solution. Finally, the swelling ratio of extruded product is presented to compare with the experimental results in drug production. The extrudate size obtained from computation shows agreement with both the experiment and the analytical figures. The extrudate size for experiment matches well with computational method with a discrepancy to the analytical formula. The swelling ratio of free surface for experiment is slightly different from numerical prediction with an error of 1% whilst the error between analytical and experimental values reaches 6%. For the next simulation to duplicate the feasible products, the new setting can be checked by this simulation before the set-up of real experiment.</p>}, number={4}, journal={Applied Science and Engineering Progress}, author={Ngamaramvaranggul, Vimolrat and Thongjub, Nawalax}, year={2020}, month={Oct.}, pages={354–361} }