M-Star President John Thomas and ABEC Product Engineer Daniel Miller co-present in this installment. John gives an overview on M-Star’s modeling capacity while Daniel walks through a customer case study which uses computational modeling to provide practical insight for single-use bioreactor selection.

The post Single-Use Bioreactor Design and Peformance Comparison appeared first on M-Star.

M-Star President John Thomas shares best practices for bioreactor CFD simulation, covering the entire process from model setup to post-processing. Special emphasis is placed on process scale-up and scale-down, as well as bioprocess intensification.

The post Best Practices for Bioreactor CFD appeared first on M-Star.

M-Star President John Thomas discusses interfacial mass transfer within the context of free-surface and sparged-gas systems. The webinar covers the physics driving mass transfer, as well as approaches for modeling these physics in industrial engineering systems. Emphasis is given to dissolved oxygen and CO2 transfer, as related to cell growth and pH control. 

The post Interfacial Mass Transfer in Gas-Liquid Systems appeared first on M-Star.

M-Star President John Thomas discusses chemically reactive turbulent flows in the context of industrial scale CFD. The webinar covers explicit and implicit approaches for modeling fast competitive reactions, as well as the competition between mixing and reaction kinetics in both single and multiphase systems. It also includes approaches for handling equilibrium reactions. 

The post Modeling Chemical Reactions in Turbulent Flows appeared first on M-Star.

M-Star President John Thomas discusses direct numerical simulation (DNS) and large eddy simulation (LES) within the context of industrial scale computational fluid dynamics (CFD). The analysis includes a discussion of length scales relevant to turbulent flows, as well as the influence of LES filtering on the predicted velocity fields. Whereas previous investigations into DNS and LES modeling approaches have focused primarily on academic systems, these high-fidelity modeling applications are discussed within the context of industrial scale.  

The post DNS and LES CFD on GPUs appeared first on M-Star.

M-Star President John Thomas discusses advanced convective heat transfer in turbulent flows. This webinar is a longer form addition to a talk given at the AICHE Meeting in San Diego 2024. This work introduces a generalized, first-principles approach to predict convective heat transfer coefficients in turbulent systems using large eddy simulation combined with high-resolution lattice-Boltzmann methods (LBM). Moving beyond traditional empirical correlations, this approach directly connects local heat transfer rates to energy dissipation rates near boundaries, offering system-independent predictions. 

The post Advanced Convective Heat Transfer in Turbulent Flows appeared first on M-Star.

M-Star President John Thomas presents on analyzing flow in HVAC systems using real-time CFD. The analysis includes computing velocity and temperature distributions, as well as predicting particular counts and path lines. This discussion is particularly relevant to engineers and scientists working in the areas of human comfort, architectural design, facility design, and ventilation engineering. 

The post Modeling HVAC Systems using Real-time CFD appeared first on M-Star.

The team from M-Star Simulations will join with major industry partners at the 2024 AIChE Annual Meeting in San Diego, California. AIChE is the world’s leading organization for chemical engineering professionals, with more than 60,000 members from more than 110 countries. M-Star’s team will be involved in six presentations with partners that include Amgen, Sartorius, Corteva, SPX Flow, R.E. Mason, University of Dayton, and NOV Chemineer.

“The work that M-Star is presenting at the 2024 AIChE meeting is important in demonstrating both the utility of the platform and the accuracy of the predictions,” says Senior Support Engineer Eric Janz, who will be presenting at the conference. “In one presentation we are working with NOV Mixing Technologies (Chemineer) to validate gas induction through a hollow shaft and impeller into an agitated vessel. In another presentation we are working with SPX Flow to model transient heating of a jacketed vessel. In both instances we used their experimental setup and data to successfully validate the simulation results.”

The post M-Star Prepares to Make Impactful Contribution at the 2024 AIChE Annual Meeting appeared first on M-Star.

The post Residence Time Distributions, Mean-Age Analysis, and Flow Characterization in Open Systems appeared first on M-Star.

For pharmaceutical companies like Bristol Myers Squibb, the biomanufacturing process of biologic drugs— substances produced by living organisms within stirred tank bioreactors—presents challenges to process scale-up and intensification. But the complex fluid mechanics at play make traditional predictive mathematical modeling slow and difficult.

That’s why Bristol Myers Squibb turned to modern CFD software.

With M-Star CFD, Bristol Myers Squibb successfully developed a framework for building time dependent, bubble-resolved, two-phase models to investigate the real-time blending and mass and oxygen transfer in stirred tank bioreactors.

The following case study follows findings from research featured in Chemical Engineering Science, 2021.

Problem

Bioreactors are designed for specific scales and conditions. So, Bristol Myers Squibb needed to find a way to predict and understand how a process they developed and optimized at a small, tabletop scale will translate to the production scale— while still maintaining optimal bioreactor operation.

The problem is that dissolved gas concentrations—a key environmental parameter in a biologic process with living organisms—are informed by complex fluid mechanics that make a single-phase fluid model inapplicable. To properly capture the complexity of cell culture processes, Bristol Myers Squibb needed to develop a two-phase fluid model that is capable of not only handling agitation and gassing but also supporting species transport across the bubble and liquid interface.

What about RANS?

Time-average flow fields provide little value to turbulent bioprocess simulation. Fully transient simulations, which run efficiently on GPUs, present a far superior paradigm for model fluid mixing and mass transfer processes.

Solution

To predict the complex, multi-fluid mixing process—and successfully scale-up production and ultimately bring compounds to market faster—Bristol Myers Squibb turned to modern CFD software that solves lattice-Boltzmann-based transport algorithms with GPU resources.

“For many cases, this fully resolved implementation can generate engineering predictions faster and with fewer modeling assumptions than RANS/population balance approaches in a multi-CPU environment.”

Physics Investigated

• Fluid Mechanics
• Energy Input & Dissipation
• Bubble Dynamics
• Mass Transfer
• Fluid Species Transport

Systems Modeled

• Blend time at 500 L and 2000 L
• Power input and dissipation at 500 L and 2000 L
• Velocity field and local energy dissipation rates at 5 L
• Mass transfer coefficient at 5 L, 200 L, 500 L and 2000 L
• Sensitivity of output to coalescence and break-up models at 5 L

Computational Considerations

• 4 Nvidia Tesla V100 GPUs
• 1.5 billion lattice updates per second
• 0.5-1 minute of wall time per second of simulation time

With the support of M-Star CFD—including in-depth training with the M-Star team and consulting expertise in fluid mechanics and process simulation—Bristol Myers Squibb was able to model, solve and validate a mechanistic approach for predicting mass and oxygen transfer.

Results

Across the range of operating conditions, the predicted values from the model agreed with measured conditions. The results arm Bristol Myers Squibb with the model and information they need to understand how a process will translate from small scale to production scale.

With M-Star CFD, there were multiple order-of-magnitude improvements in computational speed when solving the fluid transport equations, further bolstered by the use of GPUs over a CPU cluster. The results were in good agreement with experimental data with no model reparameterization between scales or operating conditions.

“By running the algorithm on graphics processing units (GPUs), the approach is shown to solve at timescales practical for industrial application.”

All in all, this approach helped Bristol Myers Squibb speed up the production timeline and reduce the cost of biologics production, so they could get compounds to market faster.

About Bristol Myers Squibb

Bristol Myers Squibb is a global biopharmaceutical company headquartered in New York, NY that manufactures innovative pharmaceuticals and biologics in several therapeutic areas, including oncology, cardiovascular, immunoscience and fibrosis.

The post Bristol Myers Squibb Predicts Mass & Oxygen Transfer in Bioreactors appeared first on M-Star.