In the design and operation of bioreactors, hydrodynamic effects are consistently critical factors. Depending on the application area, their impacts can be divided into two categories: the first is tissue engineering and regenerative medicine, which focuses on sublethal mechanical stimulation and its effects on cell physiology and differentiation; the second is biopharmaceutical manufacturing, which emphasizes the influence of hydrodynamic stress on cell growth and product expression during process scale-up, integrating these effects into process design.
Extensive experience has been accumulated in studies of production cell lines such as CHO cells. Beyond model systems, stirred-tank reactors (STRs), as the industrial mainstream equipment, have also been widely investigated for their fluid dynamic characteristics. While agitation and sparging facilitate mixing and mass transfer, they also introduce shear and turbulent stresses that may damage cells or alter cellular metabolism. Therefore, the potential impacts of these stresses must be fully considered when setting parameters such as agitation speed and aeration rate. Modern CHO cell lines have become well-adapted to suspension culture, with relatively mature processes. However, as cell density and product yield continue to rise, so do the demands for mixing and oxygen supply, intensifying hydrodynamic stress and increasing the risk of cell damage.
Although shear sensitivity has received attention, systematic mitigation of the adverse effects of hydrodynamic stress on cells remains challenging. Current process development largely relies on empirical rules, employing only global parameters such as average power input or mixing time, which fail to reflect the heterogeneity within bioreactors. Computational Fluid Dynamics (CFD) enables relatively accurate simulation of these complex phenomena, yet its practical application in process development remains limited.
This paper focuses on hydrodynamic effects induced by impellers, as they are considered one of the primary sources of cell damage. While factors such as bubble rupture also generate stress, studies indicate their influence diminishes with scale-up. Subsequent sections will briefly introduce relevant principles, discuss mainstream design approaches, and propose optimization strategies to enhance bioreactor design efficiency and cell culture performance.
Impacts of Hydrodynamics on Mammalian Cell Culture
In recent years, research on how hydrodynamics influence mammalian cell culture has advanced significantly. Study targets have shifted from early adherent cells relying on microcarriers to suspension cells of greater industrial relevance, including CHO and hybridoma cells. Research is broadly categorized into two types: one conducted in model systems with controllable flow conditions, which aids in elucidating fundamental mechanisms such as shear stress; the other performed in actual stirred-tank reactors, which, despite structural complexity, better reflects industrial applications. Model systems support mechanistic investigations, whereas reactor systems are more suitable for developing process optimization and scale-up strategies.
Shear stress exerted on cells can be calculated relatively accurately in traditional laminar flow systems. However, in complex flow fields, additional factors—including normal stress, spatial variations, and temporal dynamics—must be accounted for, making it difficult to describe the overall impact on cells using a single value. Consequently, the broader term “hydrodynamic effects” is commonly employed.
In stirred-tank reactors, Energy Dissipation Rate (EDR) serves as a key parameter characterizing mixing intensity. It is divided into local volumetric energy dissipation rate (vEDR) and average volumetric energy dissipation rate (vEDR_av), with the latter widely used for process evaluation and scale-up design. In practice, vEDR_av typically ranges from 10 to several hundred W/m³, and deviations outside this range may be detrimental to cells.
Cellular responses to hydrodynamic stress depend not only on EDR itself but also on impeller type, reactor configuration, and cell line tolerance. Different cell lines exhibit markedly distinct responses to identical EDR conditions, with CHO cells demonstrating strong adaptability. Furthermore, multi-stage agitation systems are more favorable than single-stage systems for reducing local high-stress zones and improving the cellular growth environment. For instance, at equivalent EDR, single-stage agitation tends to create regions of extremely high local dissipation, whereas three-stage impeller arrangements effectively mitigate this issue.
Although much data originates from small-scale studies, they still provide valuable guidance for large-scale reactor design. Most industrial large-scale STRs currently adopt multi-stage agitation designs to lower local shear intensity. Nevertheless, even within a safe EDR range, insufficient mixing can lead to heterogeneous nutrient and oxygen supply, impairing cell proliferation and protein expression. Studies have noted that certain configurations result in reduced cell growth under low power input or inadequate agitation, typically stemming from poor mixing.
Thus, establishing rational scale-up models and accurately simulating fluid environments at industrial scale are critical for achieving stable culture and high productivity. Further systematic research is needed to define operating windows, optimize reactor design, and improve process scalability.
Evaluation of Hydrodynamic Effects in Process Design
In industrial bioprocess development, empirical correlations and heuristic approaches are commonly used for process design and scale-up. A prevalent strategy is to maintain consistent reactor geometry or key parameters—such as impeller tip speed, mixing time, or energy dissipation rate—across different scales. Among these, average volumetric energy dissipation rate is widely adopted for scale-up due to its ease of calculation via power input and its ability to sustain cell stability within a defined range.
Nonetheless, the direct correlation between average energy dissipation rate and cellular response remains unclear, leading to its empirical reliance in application. To more accurately assess the impact of agitation on cells, Kolmogorov length scale theory has been proposed. This theory posits that cell damage may occur when turbulent eddy scales are smaller than cell or microcarrier dimensions. While applicable to microcarrier cultures, this approach has limited utility for suspension cells, owing to their small and broadly distributed sizes and the theory’s failure to account for local extreme conditions and cell size distributions.
Accordingly, some studies have simulated the volumetric distribution of cells under various flow field conditions, attempting to establish relationships between cellular responses and local energy dissipation environments. Although the concept of “critical volume fraction” has been introduced—wherein the size of regions exceeding a specific threshold may affect cell growth and productivity—its predictive performance remains limited.
An alternative approach involves evaluating the coefficient of variation of hydrodynamic parameters (e.g., flow velocity, shear rate, and dissipation rate) to reflect flow uniformity within the reactor. Multi-impeller or axial-radial agitation systems generate more uniform flow fields and stable processes, rendering them widely adopted at industrial scale.
In summary, hydrodynamic heterogeneity can serve as a reference for reactor suitability but cannot independently predict cell growth or productivity variations. Consequently, process design should employ a multi-parameter comprehensive evaluation framework, integrating flow characteristics, cell physiology, and process control for systematic optimization.
Recommendations for Process Development
Clear correlations exist between hydrodynamics and cell culture behavior as described above; however, no universal parameter can independently and accurately predict cellular responses. While certain hydrodynamic parameters correlate consistently with bioprocess performance in specific reactor designs, such correlations often fail to translate upon reactor type changes or scale-up. Therefore, assessing the impact of fluid dynamics on cell culture must account for multiple parameters and their spatiotemporal distributions.
Although shear sensitivity is routinely incorporated into reactor design considerations, internal fluid heterogeneity is frequently overlooked. Industrial reactor process design is predominantly based on average parameters, including average energy dissipation rate, mixing time, volumetric mass transfer coefficient, impeller tip speed, and geometric ratios. Yet local high-stress regions and overall flow uniformity are typically only qualitatively evaluated via CFD. In recent years, CFD has become an essential tool for reactor design, with increasingly powerful simulation capabilities enabling detailed characterization of local fluid behavior, particularly in complex gradient environments.
As cellular responses to hydrodynamic stress cannot yet be fully predicted, current process development remains highly experiment-dependent, especially using scaled-down model systems. Accordingly, a “semi-predictive” approach combining CFD and Design of Experiments (DoE) is recommended to reduce trial-and-error and enhance process development efficiency and mechanistic understanding.
For process development involving new reactors or cell lines, comprehensive CFD simulations of reactor systems under varying agitation, tilt, or shaking conditions are advised. Beyond average energy dissipation rate, evaluations should include local extrema, distribution ranges, and potential anomalous zones. Such regions—characterized by high local dissipation rates or shear rates—may indicate underlying risks. Frequency distributions of energy dissipation can further be compared with critical stress profiles of the cell line to identify overlapping high-risk zones, enabling selection of appropriate agitation speeds or operating conditions.
Preliminary culture tests should cover the entire feasible operating range to validate simulation results and narrow the optimal operating window. Nested experimental design strategies can be employed to reduce experimental load while improving data utilization efficiency. Monitoring of key indicators (e.g., cell growth, titer, and product quality) facilitates identification of optimal or suboptimal operating conditions, with replicate experiments performed near critical operating points to meet statistical validation requirements under current regulatory frameworks.
If existing data on cellular responses to specific hydrodynamic conditions are available, such information can be used to retrospectively analyze CFD results and predefine potential suitable operating ranges. Integration with DoE methodologies further reduces experimental workload, shortens development timelines, and enhances the scientific rigor and controllability of process design.
Conclusion
The relationship between hydrodynamics and cell culture processes is not yet fully understood, with numerous unknowns remaining particularly at the process design level. Currently used process design parameters are largely average values derived from single systems, neglecting the effects of local fluid gradients—an issue especially pronounced during process scale-up. There is therefore an urgent need to challenge the applicability of traditional parameters and incorporate local-scale fluid analysis. Studies indicate that the Kolmogorov length scale hypothesis, without adequate consideration of flow field heterogeneity, may be insufficient to accurately describe cellular responses to hydrodynamic stress. Given the unpredictability of local gradients, cell culture process design still relies on experimental validation, underscoring the importance of in-depth flow field understanding.
To date, most studies focus on small-scale reactors, with limited data applicable to industrial scale-up. While some scale-down investigations exist, further research is required to establish direct correlations between hydrodynamic effects and cellular behavior, particularly in large-scale culture environments. Beyond agitation, hydrodynamic stress induced by bubbles may also influence cell behavior. Although early studies suggested minimal impact, recent literature reveals significant differences in bubble behavior between large- and small-scale reactors; for example, altered bubble migration under wake effects may affect cell growth.
In conclusion, despite improved adaptability of industrial cell lines to hydrodynamics, the influence of hydrodynamic effects must remain a key consideration during process optimization and parameter setting. The integration of experimental methods with modeling tools such as CFD promises to deepen understanding of bioprocesses, enabling more efficient and controllable process development.
