PhD viva voce of Mr. Bebhash S. Raj on Thursday, October 9, 2025
Venue:
ME Auditorium, Mechanical Engineering Department.
October 9, 2025
The PhD viva voce of Mr. Bebhash S. Raj as per the following schedule:
Date of Examination: Thursday, October 9, 2025
Time: 2.30 PM - 4.00 PM
Venue: ME Auditorium, Mechanical Engineering Department.
Title of the thesis: Numerical simulation of polymer in manufacturing processes: Pelletization and Devolatilization
Chairperson: Prof. Amit Sethi, Professor, Electrical Engineering Department, IIT Bombay
External Examiner: Prof. Shanmugam Dhinakaran, Professor, Department of Mechanical Engineering, IIT Indore.
Internal Examiner: Prof. Neeraj Kumbhakarna, Associate Professor, Mechanical Engineering Department, IIT Bombay
Supervisor: Prof. Abhilash J. Chandy, Professor, Mechanical Engineering Department, IIT Bombay
Abstract of Thesis:
The production of polymers at an industrial scale involves highly complex processes governed by coupled interactions between fluid dynamics, heat transfer, mass transfer, and phase change phenomena. Two of the most critical stages in polymer manufacturing—pelletization and devolatilization—directly influence the quality, performance, and commercial viability of polymer products. Despite their importance, both processes are prone to persistent operational challenges that lead to production inefficiencies, increased downtime, and compromised product quality. Defects such as die-hole freeze-off, pellet agglomeration, and incomplete removal of residual volatiles are frequently encountered, especially under conditions where process parameters are not optimally tuned. Experimental analysis of these defects is often limited by the opaque nature of polymer melts, the harsh thermal environment, and the complexity of multiphase flow and heat transfer interactions. As a result, there is a significant need for high-fidelity computational models that can simulate real-world operating conditions, predict defect onset, and provide actionable insights for process optimization. This research presents a comprehensive computational framework employing three-dimensional Computational Fluid Dynamics (CFD) to simulate, analyze, and optimize both the pelletization and devolatilization processes in polymer manufacturing.
Firstly, the pelletization component focuses on the underwater die-face pelletizer, where molten polymer is extruded through die holes and immediately cut and cooled in a water bath. The study identifies critical regions prone to die-hole freeze-off, where excessive heat loss to the cooling water causes premature solidification of the polymer within the die holes, leading to blockage and production shutdown. Furthemore, conditions that lead to pellet agglomeration, where pellets fail to solidify sufficiently and adhere to one another after exiting the die, are also characterized. A key contribution in this area is the development of the Pellet Agglomeration Number (PAN), a non-dimensional metric that combines the effects of flow inertia (Reynolds number), pressure forces (Euler number), and thermal gradients (dimensionless temperature ratio). PAN serves as a robust predictive tool for assessing agglomeration tendencies and is validated against experimental observations from industrial-scale operations.
Secondly, the devolatilization component addresses the removal of residual volatile compounds, such as cyclohexane, from polymer-cement mixtures using a superheated steam contactor. The CFD model for devolatilization adopts an Eulerian-Lagrangian approach, where the continuous phase consists of turbulent superheated steam, and the discrete phase consists of polymer droplets injected into the steam flow. A significant challenge addressed in this part of the work is the role of particle distribution and clustering in determining the efficiency of volatile removal. To quantify this effect, the study introduces two novel metrics—the Cluster Distribution Index (CDI) and the mass-based Cluster Distribution Index (mass-CDI)—which capture the degree of particle dispersion or aggregation within the steam contactor. These indices are shown to directly correlate with devolatilization efficiency; lower CDI values indicate well-dispersed particle fields that maximize interfacial area for heat and mass transfer, whereas higher CDI values indicate clustering, which impairs solvent removal.
Overall, this research provides a validated, high-fidelity computational framework that advances the mechanistic understanding of pelletization and devolatilization processes in polymer manufacturing, thereby representing a significant step toward enabling predictive control and real-time optimization in industrial settings.