Get your free personalized podcast brief

We scan new podcasts and send you the top 5 insights daily.

Scaling from shake flasks to stirred-tank reactors is notoriously difficult because flasks feature violent surface rotation and thin liquid wall films. The free-surface algorithms originally developed to track large rising gas bubbles in 160,000-liter reactors can accurately model oxygen transfer and shear in shake flasks, allowing direct mathematical translation into stirred-tank aeration and agitation parameters.

Related Insights

Scaling up a bioprocess from lab to production fundamentally alters physical properties like oxygen transfer (KLA). This change in physics, not necessarily a procedural mistake, is often the root cause of failure at scale, leading to different cell growth and product quality.

Because CFD operates strictly on physical parameters rather than biological mechanisms, a vessel needs to be fully characterized only once across 100 to 200 operational points. When switching to a new molecule, media, or cell line, the scale-up software only needs updated fluid properties such as viscosity, density, surface tension, and salt concentration, eliminating the need to re-run simulations from scratch.

By training on multi-scale data from lab, pilot, and production runs, AI can predict how parameters like mixing and oxygen transfer will change at larger volumes. This enables teams to proactively adjust processes, moving from 'hoping' a process scales to 'knowing' it will.

New modalities like CAR-T and cultivated meat are highly shear-sensitive. Shaken bioreactors are ideal because they are surface-aerated, avoiding bubble formation. The primary cause of cell damage is not the mixing itself, but the high local shear forces created when bubbles burst, a problem that surface aeration circumvents.

Conventional scale-up rules of thumb, such as constant power per volume, fail to accurately reproduce cellular microenvironments when working with shear-sensitive cell therapies or delicate organisms like Streptomyces. By simulating and matching comprehensive physical parameters across vessels using CFD, teams achieve far more comparable cell morphology and reliable cell growth without relying on protective additives like antifoam.

Massive CFD simulation datasets can be distilled into compact reduced-order surrogate models of just a few megabytes using machine learning. Rather than sifting through overwhelming data files, the model interpolates between simulated test points to predict key bioreactor parameters—including kLa, shear rate distribution, and mixing times—in milliseconds for any given operating condition.

Over 90% of scientific publications omit the shaking diameter for shake flask experiments. This single parameter can alter oxygen supply by up to 50%, making it as crucial as impeller type in a bioreactor and a primary reason for failed experiment replication.

Beyond oxygen transfer, the ventilation rate (VVM)—which removes volatile compounds like CO2—is a critical scale-up parameter. A process failed to scale until the bioreactor's aeration was reduced from a standard 1 VVM to 0.5 VVM to match the shake flask's implicit rate, restoring product yield.

Orbital shaken bioreactors, like shake flasks, are where fundamental decisions about production strains and media are made in industry. Despite their importance, the topic is often omitted from university education, leading to a knowledge gap that directly causes poor experimental design and reproducibility issues.

In early microbial cultivation R&D, focusing on whether a system is 'stirred or shaken' is a distraction. The most critical parameter for success is the amount of oxygen introduced (KLa and oxygen transfer rate), not the mechanical method of delivery.