7 results
Now showing1 - 7 of 7
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Data-driven transport modeling for tracer injection and mixing time screening in stirred tank reactors(Elsevier, 2026-10-30); ; ; ; Transient CFD of liquid mixing in stirred tanks is too expensive for systematic design studies, and mixing time predictions deviate from experiment even at scale-resolving resolution. Transport-based recurrence CFD (rCFD) decouples scalar transport from the flow solve, replaying a short pre-computed flow database through cell-to-cell displacement maps, so that transport level corrections can be applied without repeating the momentum solution. This is exploited here for the 95% mixing time in a 30 L dual-impeller stirred tank reactor, using a 4.5 s LB-LES flow database from M-Star as the sole input. The rCFD first results inherit the over-prediction of its input flow field (33.1 ± 1.4 s, against 30.1 ± 3.0 s for M-Star and 28.1 ± 2.7 s for Ansys Fluent at comparable spatial resolution, and 25.0 ± 3.0 s measured by conductivity probes in experiment part). Calibrating a single face-swap diffusion parameter in rCFD transport model against one probe measurement brings the prediction to 24.5 ± 1.0 s, recovering experimental agreement without re-running the flow simulation. The rCFD simulation requires approximately 3 s of wall-clock time per second of process time (423 × M-Star, 1200 × Ansys Fluent) and is almost insensitive to a fivefold increase in extrapolation time step. Ninety injection locations spanning radius, height, and azimuth are therefore screened from the single database in 2.25 h, against estimates of 40 and 113 days for the reference solvers, resolving a 3.6-fold spread in mixing time (13.6–48.8 s) in which injection height and baffle-relative azimuth dominate. The main objective of this work is a fast, spatially and temporally resolved description of scalar transport; the 95% mixing time is used as the quantitative metric for validating it against sparse experimental data.Publicationtype: Journal Article2 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , A method for optical quantification of local oxygen limitations in multiphase bioreactors(Wiley, 2026-06-23); ; ; ; Spatial oxygen limitation is a major source of physiological stress and metabolic, scale-dependent heterogeneity in aerobic bioprocesses. In gas-liquid stirred tank reactors, oxygen availability is governed by hydrodynamics and mass transfer and is therefore inherently non-uniform. Despite its importance, experimentally accessing spatial information on oxygen local supply remains challenging, as most established techniques rely on point-wise sensing or intrusive tracer-based methods. This work presents an optical method for visualizing local oxygen limitation based on the reversible redox chemistry of methylene blue in alkaline glucose solutions. After driving the reactor into a uniformly reduced, colorless state, the spatial reoxidation of the indicator during oxygen aeration is recorded and evaluated pixel-wise. The resulting two-dimensional fields of colorization time provide a spatially continuous indicator of local oxygen supply governed by hydrodynamic transport and gas-liquid mass transfer. Systematic experiments were conducted to assess the influence of glucose and sodium hydroxide concentration on the stability and reproducibility of the redox system, enabling the selection of a robust reaction medium. The method was then applied to a laboratory-scale stirred tank operated at different impeller speeds. The resulting colorization fields reveal characteristic spatial structures associated with oxygen-limiting regions, including insufficiently supplied zones under weak agitation, progressive homogenization with increasing turbulence, and residual transport limitations near reactor walls at high agitation rates. The presented approach requires no local tracer injection and operates under normal aeration conditions, providing qualitative yet physically meaningful insight into oxygen-supply and mixing related heterogeneities. It therefore offers a simple and cost-effective tool for identifying limiting operating points, supporting future scale-transfer studies, and complementing CFD-based analyses of oxygen transport in stirred tank bioreactors.Publicationtype: Journal Article34 9 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Reproducibility of GPU-based Large Eddy Simulations for mixing in stirred tank reactors(Elsevier, 2026-03-02); ; ; ; CFD simulations are widely used to quantify the mixing performance of stirred tanks for various applications in chemical engineering and biotechnology. Due to advances in GPU computing, these simulations increasingly employ Large Eddy Simulation (LES), which explicitly resolves the dynamics of large-scale turbulence. Although such simulations are fully deterministic and therefore theoretically reproducible, small numerical variations induced by round-off errors, floating-point arithmetic, and differences in the distribution and ordering of operations in parallel computing lead to separation of trajectories i.e., different flow-field evolutions and consequently to significant run-to-run variability in predicted mixing times, even on the same hardware architecture. This work investigates the impact of repeated simulations, in the form of a case study, on the mixing-time distribution observed in a (Formula presented) stirred tank reactor using two commercial CFD packages operating with representative, production-level solver configurations. The analysis does not aim to assess the general performance of numerical method classes, but rather to quantify run-to-run variability under fixed solver settings and to compare the resulting numerical distributions to experimental variability. The results demonstrate that numerical variability is of comparable magnitude to the experimental spread, highlighting the necessity to treat LES-derived metrics as statistical ensembles rather than deterministic values. It is concluded that the reporting of confidence intervals is essential for methodological rigour in LES-based mixing studies.Publicationtype: Journal Article36 61 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , 35 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Lagrangian sensor particles for detecting hydrodynamic heterogeneities in industrial bioreactors: experimental analysis and Lattice-Boltzmann simulations(Elsevier, 2025-04-19); ; ; ; This study analyzes trajectories of three particle types in an industrial-scale bioreactor, equipped with a Rushton turbine and a pitched blade turbine, to characterize hydrodynamic compartments. The trajectories obtained from measurements with Lagrangian Sensor Particles (LSP,exp) are compared to those generated by Lattice-Boltzmann large eddy simulations (LB LES). The latter method is used to reproduce analogous simulated LSPs (LSP,sim) as resolved particles. Additionally, for benchmarking purposes, massless tracer particles (tracer,sim) are incorporated to accurately represent fluid flow dynamics. Discrepancies in the axial probability of presence and velocity between LSP,exp and LSP,sim likely stem from differences in mass distribution, density, number of particles, and ratio of particle size to grid. A necessarily high LSP,sim volume fraction in LB LES leads to increased collisions and clustering, negatively impacting flow dynamics, and reducing turbulent kinetic energy by at least 3%. Circulation and residence time distributions for the three types of particles identify three hydrodynamic compartments within the bioreactor, validated by local mixing time distributions. The ratio of overall average circulation time to global mixing time is Θglob,95≈3.0⋅t¯circ for LSP,exp, which largely corresponds to literature results. A theoretical LSP size of dp,th≈1mm is estimated to be flow following on micro-scale in the bulk phase, if a Stokes number of St=0.1 is assumed. However, Stokes number estimations confirm that LSP,exp are capable to follow flow patterns on the meso-scale and macro-scale with St≈0.2 and St≈0.002, respectively. Hence, hydrodynamic structures at length scales greater than or equal to the size of the impeller can be investigated by current state-of-the-art LSPs, which proves their technological readiness for industrial bioreactors.Publicationtype: Journal Article53 26 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Dynamics of Lagrangian sensor particles : the effect of non-homogeneous mass distribution(Multidisciplinary Digital Publishing Institute (MDPI), 2024-08-01); ; ; ; The growing demand for bio-pharmaceuticals necessitates improved methods for the characterization of stirred tank reactors (STRs) and their mixing heterogeneities. Traditional Eulerian measurement approaches fall short, culminating in the use of Lagrangian Sensor Particles (LSPs) to map large-scale STRs and track the lifelines of microorganisms such as Chinese Hamster Ovary cells. This study investigates the hydrodynamic characteristics of LSPs, specifically examining the effects that the size and position of the Center of Mass (CoM) have on their flow-following capabilities. Two Lagrangian Particle (LP) designs are evaluated, one with the CoM and a Geometric Center aligned, and another with a shifted CoM. The experimental study is conducted in a rectangular vessel filled with deionized water featuring a stationary circular flow. Off-center LPs exhibit higher velocities, an increased number of floor contacts, and moreover, a less homogeneous particle probability of presence within the vessel compared to LPs with CoM and Geometric Center aligned. Lattice Boltzmann Large Eddy Simulations provide complementary undisturbed fluid velocity data for the calculation of the Stokes number St. Building upon these findings, differences in the Stokes number St between the two LP variants of ΔSt = 0.01 (25 mm LP) and ΔSt = 0.13 (40 mm LP) are calculated, highlighting the difference in flow behavior. Furthermore, this study offers a more representative calculation of particle response time approach, as the traditional Stokes number definition does not account for non-homogeneous particles, resulting in an alternative Stokes number (ΔStalt = 0.84 (25 mm LP) and ΔStalt = 2.72 (40 mm LP)). This study contributes to the improved characterization of STRs through the use of Lagrangian Sensor Particles. Results highlight the implications the internal mass distribution has on LSP design, offering crucial considerations for researchers in the field.Publicationtype: Journal Article32 16 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Experimental analysis of lifelines in a 15,000 L bioreactor by means of Lagrangian Sensor Particles(Elsevier, 2024-05-01); ; ; ; This study employs Lagrangian Sensor Particles (LSPs) with a diameter of 40 mm equipped with a pressure sensor to investigate cell lifelines in a 15,000 L stirred tank reactor (STR) with three Elephant Ear impellers. The Stokes number of the LSPs is approx. 0.004 on a macro-scale. The vertical probability of presence, axial velocity profiles, circulation time distributions, and residence time distributions are quantified to analyze single-phase mixing heterogeneities, detect hydrodynamic compartments and conduct a Lagrangian regime analysis. Results reveal a similarly distributed probability of presence in the vertical reactor center but emphasize the LSP's sensitivity to fluctuating densities. Axial velocity distributions illustrate characteristic impeller-induced flow patterns, and circulation time distributions identify three compartments with comparatively shorter times in the axial center. Residence time distributions exhibit a similar compartmentalized profile. Moreover, the study estimates a potential oxygen deprivation zone for CHO cells in the upper compartment and demonstrates the LSP's efficacy in characterizing impeller systems. Contrary to literature, the ratio of examined global mixing times to circulation times is 1.0, highlighting macro-scale mixing. The research underscores that LSPs offer crucial insights into industrial-scale STRs, specifically for determining hydrodynamic compartments without having optical access.Publicationtype: Journal Article27 128