arxiv: 2604.26695 · v1 · submitted 2026-04-29 · ⚛️ physics.flu-dyn

Recognition: unknown

Compartment Modelling of Multiphase Reactors using Unsupervised Clustering

Michael Mitterlindner , Maximilian Graber , Regina Kratzer , Markus Reichhartinger , Stefan Radl

Authors on Pith no claims yet

Pith reviewed 2026-05-07 12:33 UTC · model grok-4.3

classification ⚛️ physics.flu-dyn

keywords compartment modelingmultiphase reactorsunsupervised clusteringCFD reductionmass conservationinterphase mass transferreactor optimizationmodel reduction

0 comments

The pith

Unsupervised clustering of CFD data produces accurate compartment models for multiphase reactors.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces a toolbox called CLARA that turns detailed but expensive CFD simulations of multiphase reactors into simpler compartment models. It does so by grouping the flow data with unsupervised clustering, then using graph adjustments and optimization to keep compartments connected and mass conserved while including interphase mass transfer and reactions inside each group. Verification on analytical benchmarks and full reactive multiphase simulations shows these reduced models reproduce both overall reactor performance and local species distributions. Because the compartment models run far faster than the original CFD, they support real-time control and repeated design changes that would otherwise be impractical. The central advance is automating this reduction for multiphase systems where prior methods handled only single-phase cases.

Core claim

CLARA automates the generation of compartment models from CFD data for multiphase reactors by applying unsupervised clustering algorithms, followed by graph reassignment and optimization routines that enforce mass conservation and spatial connectivity, while modeling multiphase phenomena and interphase mass transfer within compartments. Verification studies with analytical benchmarks and reactive multiphase CFD simulations confirm that the resulting compartment models accurately reproduce reactor performance metrics and spatial species distributions.

What carries the argument

The CLARA toolbox, which partitions CFD data via unsupervised clustering combined with graph reassignment and optimization to create connected, mass-conserving compartments that incorporate multiphase effects and interphase transfer.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same clustering-plus-reassignment pipeline could be applied to other transport-dominated systems such as porous-media flows or atmospheric dispersion to create control-oriented reduced models.
Integration with existing process simulators might allow hybrid CFD-compartment workflows where only critical regions retain full resolution.
If mass-conservation enforcement proves robust across scales, the approach could lower the computational barrier for closed-loop optimization in industrial chemical reactors.

Load-bearing premise

The unsupervised clustering step plus later adjustments will always identify groups that capture the essential multiphase flow patterns and transfer rates without missing critical features in new geometries.

What would settle it

Generate a CLARA compartment model from a new reactive multiphase CFD dataset on an unseen reactor geometry, then compare its predicted outlet concentrations and internal species profiles over time against the original full CFD run; systematic deviations beyond numerical tolerance would falsify the accuracy claim.

Figures

Figures reproduced from arXiv: 2604.26695 by Markus Reichhartinger, Maximilian Graber, Michael Mitterlindner, Regina Kratzer, Stefan Radl.

**Figure 1.** Figure 1: Simulation geometry with a quarter-width pure gas inlet on one side and initial filling height of 0 view at source ↗

**Figure 2.** Figure 2: Absolute error of the CM solution compared to the analytical solution for (i) reactive single phase (constant, view at source ↗

**Figure 3.** Figure 3: Time-averaged flow field of the quarter-gassed reactor: (a) liquid volume fraction view at source ↗

**Figure 4.** Figure 4: Steady-state normalised liquid-phase oxygen concentration view at source ↗

**Figure 5.** Figure 5: Local concentration error ϵc as a function of the number of clusters nc used during clustering. For the three reaction orders, obtained with agglomerative clustering using obvious features. Solid lines: volume fraction α liq as clustering feature; dashed lines: dissolved oxygen concentration c liq O2 as feature. 4.2.4. Details for the 0.5th-order reaction As established in the previous subsection, the 0.5t… view at source ↗

**Figure 6.** Figure 6: Representative CLARA solution for the 0.5 view at source ↗

**Figure 7.** Figure 7: Local concentration error ϵc (17) for the 0.5th-order case as a function of the number of clusters nc for six feature combinations, comparing agglomerative and k-means clustering. The dotted gray line ( ) marks the error of the conventional single-compartment model. Several important trends emerge from view at source ↗

**Figure 8.** Figure 8: Effect of over-clustering on the predicted concentration field for the 0.5th-order reaction, with c liq O2 as the used feature. Increasing nc from 8 to 16 enlarges the predicted depletion zone (purple color) well beyond the CFD reference, a direct consequence of compartments bridging the bubble-rich and the return-flow region. The second mechanism is the suppression of (turbulent) mixing at compartment bou… view at source ↗

**Figure 9.** Figure 9: Concentration field for the 0.5th-order reaction with Sct = 1 × 107 , obtained with agglomerative clustering using c liq O2 and nc = 10 clusters. (a) Cluster index: A single large cluster captures the headspace region, suppressing turbulent diffusivity creates a large depleted bulk region surrounded by a thin saturation rim, within the liquid region. (b) CLARA concentration prediction and (c) CFD reference… view at source ↗

**Figure 10.** Figure 10: Local concentration error ϵc (17) as a function of nc for the case with suppressed turbulent diffusivity (Sct = 1 × 107 ). The dotted gray line ( ) marks the single-compartment baseline (22 [%]). A second observation further supports this interpretation: in the baseline case, including the turbulent kinetic energy k turb liq as a clustering feature stabilised the error by guiding compartment boundaries aw… view at source ↗

**Figure 11.** Figure 11: Cluster assignment at nc = 20 using c liq O2 as the clustering feature. Agglomerative clustering produces spatially coherent, concentric shells around the depletion zone. K-means with graph reassignment yields fragmented, geometrically irregular compartments that lack physical interpretability. The result for the agglomerative clustering algorithm shown in panel (11a) clearly exhibits the thin structures… view at source ↗

read the original abstract

Detailed Computational Fluid Dynamics (CFD) simulations are too computationally expensive for the real-time control and design optimization of multiphase flow reactors. To address these limitations, we introduce CLARA, a software toolbox that automates the generation of Compartment Models (CM) via the unsupervised clustering of CFD data. Unlike previous studies, our toolbox enables the modelling of multiphase phenomena and interphase mass transfer within each compartment. CLARA employs unsupervised clustering algorithms, graph reassignment, and optimization routines to ensure mass conservation and spatial connectivity across all compartments. Verification studies utilizing analytical benchmarks and reactive multiphase CFD simulations demonstrate that the CMs produced by CLARA accurately reproduce reactor performance and spatial species distributions. The significantly reduced computational demand of CMs compared to full CFD models enables the optimal control of multiphase reactors and facilitates their rational design and optimization.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

read the letter

CLARA automates compartment model creation from multiphase CFD data with interphase transfer included, but the accuracy claims rest on thin verification details and unclear feature choices for clustering. The paper shows a clear pipeline that starts with unsupervised clustering of CFD results, adds graph reassignment for connectivity, and finishes with optimization to enforce mass conservation. That combination lets the models handle multiphase flow and transfer inside each compartment, which goes beyond many earlier compartment approaches that treated phases more simply or ignored transfer altogether. The verification against analytical benchmarks and full reactive CFD runs is a positive step, and the reduced compute cost is a real practical gain for control or repeated design calculations. The authors lay out the steps plainly enough that someone could try to reproduce the workflow. The main gaps are in the supporting numbers and setup choices. The abstract states that the models accurately reproduce performance and species fields, yet no error metrics, sensitivity to compartment count, or direct side-by-side plots appear in the summary. More importantly, the clustering features are not specified. If they are limited to velocity and volume fraction without species concentrations or reaction rates, compartments can average across zones that have very different driving forces for transfer; the later optimization only corrects global balances and cannot restore local fidelity. The paper needs to state the exact feature vector and show cases where the lumped rates still match the original CFD within stated tolerances. This work is aimed at chemical engineers and reactor modelers who already generate CFD data but need faster surrogates for optimization loops or real-time use. A reader focused on multiphase reactor design will get a usable method description and a starting toolbox. It deserves a serious referee because the core idea is testable and the pipeline is described in enough detail for reviewers to check the implementation and quantitative results. I would recommend sending it to peer review.

Referee Report

2 major / 1 minor

Summary. The paper introduces CLARA, a toolbox that automates generation of Compartment Models (CMs) for multiphase reactors by applying unsupervised clustering to CFD data, followed by graph reassignment and optimization to enforce mass conservation and spatial connectivity. It incorporates modeling of multiphase phenomena and interphase mass transfer inside compartments and claims verification via analytical benchmarks and reactive multiphase CFD simulations, showing that the resulting CMs accurately reproduce reactor performance and spatial species distributions while offering substantially lower computational cost than full CFD for control and optimization.

Significance. If the accuracy claims hold with quantitative support, the work would be significant for fluid dynamics and reactor engineering: it offers a practical, automated bridge between expensive CFD and reduced-order models that retain interphase transfer fidelity, enabling real-time control and rational design optimization of multiphase systems that current CFD cannot support at scale.

major comments (2)

[Abstract] Abstract: The central claim that 'the CMs produced by CLARA accurately reproduce reactor performance and spatial species distributions' rests on the unsupervised clustering step, yet the abstract (and pipeline description) does not specify whether species concentrations or reaction rates are included in the feature vector alongside hydrodynamic quantities (velocity, volume fraction). If species data participate only in post-processing, compartments can average across zones with dissimilar transfer driving forces; the subsequent mass-conservation optimization then enforces only global balances, not local fidelity. This is load-bearing for the accuracy claim in reactive multiphase cases.
[Abstract] Verification studies (referenced in Abstract): The manuscript states that analytical benchmarks and reactive multiphase CFD simulations demonstrate accurate reproduction, but provides no explicit error metrics (e.g., relative L2 errors on species fields, reactor performance indices, or mass-balance residuals) or implementation details (clustering algorithm, hyperparameters, feature scaling). Without these, the support for the 'accurate reproduction' claim cannot be assessed and the weakest assumption—that clustering plus reassignment preserves essential multiphase phenomena—remains untested in the supplied text.

minor comments (1)

[Abstract] Abstract: The acronym CLARA is introduced without expansion; define it on first use for clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful review and insightful comments, which have helped us improve the clarity and rigor of our presentation. We respond to each major comment below.

read point-by-point responses

Referee: The central claim that 'the CMs produced by CLARA accurately reproduce reactor performance and spatial species distributions' rests on the unsupervised clustering step, yet the abstract (and pipeline description) does not specify whether species concentrations or reaction rates are included in the feature vector alongside hydrodynamic quantities (velocity, volume fraction). If species data participate only in post-processing, compartments can average across zones with dissimilar transfer driving forces; the subsequent mass-conservation optimization then enforces only global balances, not local fidelity. This is load-bearing for the accuracy claim in reactive multiphase cases.

Authors: We appreciate the referee's concern regarding the feature selection in the clustering step. Upon review, the clustering in CLARA is indeed performed solely on hydrodynamic quantities such as velocity and volume fraction to identify compartments based on flow similarity. Species concentrations are used in the post-processing and modeling phase to compute interphase transfers within the defined compartments. This design choice is intentional to focus on flow-based zoning, which is standard in compartment modeling. However, to ensure local fidelity, the optimization enforces mass conservation and connectivity. We have revised the abstract to clarify this and added a discussion in the methodology section explaining why hydrodynamic features are used and how it still captures the essential multiphase phenomena for the reactive cases considered. revision: yes
Referee: The manuscript states that analytical benchmarks and reactive multiphase CFD simulations demonstrate accurate reproduction, but provides no explicit error metrics (e.g., relative L2 errors on species fields, reactor performance indices, or mass-balance residuals) or implementation details (clustering algorithm, hyperparameters, feature scaling). Without these, the support for the 'accurate reproduction' claim cannot be assessed and the weakest assumption that clustering plus reassignment preserves essential multiphase phenomena remains untested in the supplied text.

Authors: We acknowledge that the abstract and initial description do not include specific numerical error metrics or detailed implementation parameters. The verification is supported by visual and qualitative comparisons in the manuscript, but to provide quantitative evidence, we will include in the revised manuscript a table summarizing the error metrics, such as relative L2 errors on species concentration fields and reactor performance indices, along with mass balance residuals. We have also added the specific clustering algorithm used, the chosen hyperparameters, and the feature scaling method in the methods section. This strengthens the support for the accuracy claims. revision: yes

Circularity Check

0 steps flagged

No circularity: CLARA applies clustering/optimization to independent CFD data with external verification

full rationale

The paper describes a toolbox that ingests external CFD simulation results, applies unsupervised clustering plus graph reassignment and mass-conservation optimization, and verifies the resulting compartment models against both analytical benchmarks and the original CFD data. No equation or claim reduces a derived quantity to a parameter fitted from the same quantity; the central performance claims rest on comparison to independent reference solutions rather than on internal redefinition or self-citation chains. The method is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

2 free parameters · 1 axioms · 0 invented entities

The approach relies on standard clustering algorithms and optimization for conservation, with free parameters likely including the number of compartments and clustering hyperparameters.

free parameters (2)

number of compartments
Determines model granularity and is likely selected or optimized during the process.
clustering hyperparameters
Parameters for unsupervised clustering algorithms that affect compartment formation.

axioms (1)

domain assumption Unsupervised clustering can group CFD data into spatially connected compartments that preserve mass conservation and capture multiphase phenomena.
Core assumption enabling the automated generation of accurate CMs.

pith-pipeline@v0.9.0 · 5452 in / 1127 out tokens · 35431 ms · 2026-05-07T12:33:18.486138+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

31 extracted references · 23 canonical work pages

[1]

V . P. I. Laborda, L. Puiman, T. Groves, C. Haringa, L. K. Nielsen, Unsupervised learning bioreactor regimes, Computers & Chemical Engineering 194 (2025) 108891.doi:10. 1016/j.compchemeng.2024.108891. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0098135424003090

work page arXiv 2025
[2]

Jamshidzadeh, A

M. Jamshidzadeh, A. U. Griesz, J. W. Jensen, U. Krühne, J. M. Woodley, K. V . Gernaey, P. I. Nikel, H. Junicke, CFD-guided scaling of Pseudomonas putida fermentation, Biochemical Engineering Journal 213 (2025) 109549.doi:10.1016/j.bej.2024.109549. URLhttps://linkinghub.elsevier.com/retrieve/pii/S1369703X2400336X

work page doi:10.1016/j.bej.2024.109549 2025
[3]

Nadal-Rey, D

G. Nadal-Rey, D. D. McClure, J. M. Kavanagh, B. Cassells, S. Cornelissen, D. F. Fletcher, K. V . Gernaey, Development of dynamic compartment models for industrial aerobic fed- batch fermentation processes, Chemical Engineering Journal 420 (2021) 130402.doi: 10.1016/j.cej.2021.130402. URLhttps://linkinghub.elsevier.com/retrieve/pii/S1385894721019884

work page doi:10.1016/j.cej.2021.130402 2021
[4]

Schwarz, T

S. Schwarz, T. Frey, M. Hoffmann, M. Grünewald, M. Schlüter, CFD-Based Compartment Modeling of Continuous Polymer Reactors in Milli-Structured Apparatuses by Use of the Mean Age Theory, Industrial & Engineering Chemistry Research 62 (31) (2023) 12109– 12119.doi:10.1021/acs.iecr.3c00947. URLhttps://pubs.acs.org/doi/10.1021/acs.iecr.3c00947

work page doi:10.1021/acs.iecr.3c00947 2023
[5]

Danckwerts, Absorption by simultaneous diffusion and chemical reaction, Transactions of the Faraday Society 46 (1950) 300–304, publisher: Royal Society of Chemistry

P. Danckwerts, Absorption by simultaneous diffusion and chemical reaction, Transactions of the Faraday Society 46 (1950) 300–304, publisher: Royal Society of Chemistry

1950
[6]

Le Nepvou De Carfort, T

J. Le Nepvou De Carfort, T. Pinto, U. Krühne, An Automatic Method for Generation of CFD-Based 3D Compartment Models: Towards Real-Time Mixing Simulations, Bioengi- neering 11 (2) (2024) 169.doi:10.3390/bioengineering11020169. URLhttps://www.mdpi.com/2306-5354/11/2/169

work page doi:10.3390/bioengineering11020169 2024
[7]

Maldonado De León, A

H. Maldonado De León, A. Straathof, C. Haringa, Dynamic compartment models: Towards a rapid modeling approach for fed-batch fermentations, Chemical Engineering Science 308 (2025) 121396.doi:10.1016/j.ces.2025.121396. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0009250925002192

work page doi:10.1016/j.ces.2025.121396 2025
[8]

Le Nepvou De Carfort, M

J. Le Nepvou De Carfort, M. Raptotasios, V . Puig I Laborda, K. V . Gernaey, U. Krühne, Modeling pH gradients in industrial-scale lactic acid bacteria fermentation using flow- informed compartment models, Computers & Chemical Engineering (2026) 109568doi: 10.1016/j.compchemeng.2026.109568. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0098135426000207

work page doi:10.1016/j.compchemeng.2026.109568 2026
[9]

Lambauer, A

V . Lambauer, A. Permann, Z. Petrášek, V . Suboti ´c, C. Hochenauer, R. Kratzer, M. Re- ichhartinger, Automatic Control of Chemolithotrophic Cultivation of Cupriavidus necator: Optimization of Oxygen Supply for Enhanced Bioplastic Production, Fermentation 9 (7) (2023) 619.doi:10.3390/fermentation9070619. URLhttps://www.mdpi.com/2311-5637/9/7/619 28

work page doi:10.3390/fermentation9070619 2023
[10]

Savarese, A

M. Savarese, A. Cuoci, W. De Paepe, A. Parente, Machine learning clustering algorithms for the automatic generation of chemical reactor networks from CFD simulations, Fuel 343 (2023) 127945.doi:10.1016/j.fuel.2023.127945. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0016236123005586

work page doi:10.1016/j.fuel.2023.127945 2023
[11]

Rupprecht, Q

S. Rupprecht, Q. Gao, T. Karia, A. M. Schweidtmann, Multi-agent systems for chemical engineering: a review and perspective, Current Opinion in Chemical Engineering 51 (2026) 101209.doi:10.1016/j.coche.2025.101209. URLhttps://linkinghub.elsevier.com/retrieve/pii/S2211339825001212

work page doi:10.1016/j.coche.2025.101209 2026
[12]

Backpropagation through time and the brain.Current Opinion in Neurobiology, 55:82–89, 2019

S. Zhang, Y . R. Tan, C. M. Nguyen, D. Karan, S. Ganguly, N. Jose, M. Q. Lim, S. Guo, L. Jiang, M. Kraft, A. A. Lapkin, A semantic framework for chemical process digitalisation using ontologies, Chemical Engineering Journal 533 (2026) 174361.doi:10.1016/j. cej.2026.174361. URLhttps://linkinghub.elsevier.com/retrieve/pii/S1385894726018206

work page doi:10.1016/j 2026
[13]

Le Nepvou De Carfort, V

J. Le Nepvou De Carfort, V . Puig I Laborda, L. K. Nielsen, K. V . Gernaey, U. Krühne, Flow-informed clustering of bioreactor volumes to build CFD-based compartment models, Chemical Engineering Science 320 (2026) 122539.doi:10.1016/j.ces.2025.122539. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0009250925013600

work page doi:10.1016/j.ces.2025.122539 2026
[14]

Delafosse, M.-L

A. Delafosse, M.-L. Collignon, S. Calvo, F. Delvigne, M. Crine, P. Thonart, D. Toye, CFD- based compartment model for description of mixing in bioreactors, Chemical Engineering Science 106 (2014) 76–85.doi:10.1016/j.ces.2013.11.033. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0009250913007690

work page doi:10.1016/j.ces.2013.11.033 2014
[15]

Promma, M

I. Promma, M. G. Aucoin, N. M. Abukhdeir, H. Budman, A coupled metabolic flux/compartmental hydrodynamic model for large-scale aerated bioreactors, Computers & Chemical Engineering 189 (2024) 108806.doi:10.1016/j.compchemeng.2024. 108806. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0098135424002242

work page doi:10.1016/j.compchemeng.2024 2024
[16]

M. Kim, S. Park, D. Lee, S. Lim, M. Park, J. M. Lee, Modeling long-time behaviors of industrial multiphase reactors for CO2 capture using CFD-based compartmental model, Chemical Engineering Journal 395 (2020) 125034.doi:10.1016/j.cej.2020.125034. URLhttps://linkinghub.elsevier.com/retrieve/pii/S1385894720310263

work page doi:10.1016/j.cej.2020.125034 2020
[17]

Puiman, C

L. Puiman, C. Bokelmann, S. D. Simpson, A. M. Spormann, R. Takors, Dos and don’ts for scaling up gas fermentations, Current Opinion in Biotechnology 93 (2025) 103294. doi:10.1016/j.copbio.2025.103294. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0958166925000382

work page doi:10.1016/j.copbio.2025.103294 2025
[18]

Weber, M

B. Weber, M. V on Campenhausen, T. Maßmann, A. Bednarz, A. Jupke, CFD based compartment-model for a multiphase loop-reactor, Chemical Engineering Science: X 2 (2019) 100010.doi:10.1016/j.cesx.2019.100010. URLhttps://linkinghub.elsevier.com/retrieve/pii/S2590140019300176

work page doi:10.1016/j.cesx.2019.100010 2019
[19]

WIREs Data Mining and Knowledge Discovery2(1), 86–97 (2012) https://doi

F. Murtagh, P. Contreras, Algorithms for hierarchical clustering: an overview, WIREs Data Mining and Knowledge Discovery 2 (1) (2012) 86–97.doi:10.1002/widm.53. URLhttps://wires.onlinelibrary.wiley.com/doi/10.1002/widm.53 29

work page doi:10.1002/widm.53 2012
[20]

Tajsoleiman, R

T. Tajsoleiman, R. Spann, C. Bach, K. V . Gernaey, J. K. Huusom, U. Krühne, A CFD based automatic method for compartment model development, Computers & Chemical Engineer- ing 123 (2019) 236–245.doi:10.1016/j.compchemeng.2018.12.015. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0098135418308950

work page doi:10.1016/j.compchemeng.2018.12.015 2019
[21]

Omidi, T

M. Omidi, T. Baumann, Integrated numerical simulation of reactive bubble columns: A coupled approach with openFOAM and PhreeqC, Chemical Engineering Journal 518 (2025) 163301.doi:10.1016/j.cej.2025.163301. URLhttps://linkinghub.elsevier.com/retrieve/pii/S138589472504135X

work page doi:10.1016/j.cej.2025.163301 2025
[22]

Pedregosa, F

F. Pedregosa, F. Pedregosa, G. Varoquaux, G. Varoquaux, N. Org, A. Gramfort, A. Gram- fort, V . Michel, V . Michel, L. Fr, B. Thirion, B. Thirion, O. Grisel, O. Grisel, M. Blon- del, P. Prettenhofer, P. Prettenhofer, R. Weiss, V . Dubourg, V . Dubourg, J. Vanderplas, A. Passos, A. Tp, D. Cournapeau, Scikit-learn: Machine Learning in Python, MACHINE LEARNI...

2011
[23]

Least Squares Quantization in PCM,

S. Lloyd, Least squares quantization in PCM, IEEE Transactions on Information Theory 28 (2) (1982) 129–137.doi:10.1109/TIT.1982.1056489. URLhttps://ieeexplore.ieee.org/document/1056489/

work page doi:10.1109/tit.1982.1056489 1982
[24]

Murtagh, P

F. Murtagh, P. Legendre, Ward’s Hierarchical Agglomerative Clustering Method: Which Algorithms Implement Ward’s Criterion?, Journal of Classification 31 (3) (2014) 274–295. doi:10.1007/s00357-014-9161-z. URLhttp://link.springer.com/10.1007/s00357-014-9161-z

work page doi:10.1007/s00357-014-9161-z 2014
[25]

R. O. Fox, Computational models for turbulent reacting flows, Cambridge series in chemi- cal engineering, Cambridge university press, Cambridge New York, 2003

2003
[26]

Sander, Compilation of Henry’s law constants (version 5.0.0) for water as solvent, Atmospheric Chemistry and Physics 23 (19) (2023) 10901–12440.doi:10.5194/ acp-23-10901-2023

R. Sander, Compilation of Henry’s law constants (version 5.0.0) for water as solvent, Atmospheric Chemistry and Physics 23 (19) (2023) 10901–12440.doi:10.5194/ acp-23-10901-2023. URLhttps://acp.copernicus.org/articles/23/10901/2023/

2023
[27]

Schiller, A

L. Schiller, A. Naumann, Über die grundlegenden Berechnungen bei der Schwerkraftauf- bereitung, Zeitschrift des Vereines Deutscher Ingenieure 77 (12) (1933) 318–320

1933
[28]

Frössling, Über die Verdunstung fallender Tropfen, Gerlands Beiträge zur Geophysik 52 (1938) 170–216

N. Frössling, Über die Verdunstung fallender Tropfen, Gerlands Beiträge zur Geophysik 52 (1938) 170–216

1938
[29]

Dhaouadi, S

H. Dhaouadi, S. Poncin, J. Hornut, N. Midoux, Gas–liquid mass transfer in bubble column reactor: Analytical solution and experimental confirmation, Chemical Engineering and Pro- cessing: Process Intensification 47 (4) (2008) 548–556.doi:10.1016/j.cep.2006.11. 009. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0255270106002984 Appendix A. Simulation Par...

work page doi:10.1016/j.cep.2006.11 2008
[30]

experimental

∂c∗ ∂ξ∗ 1 =−Dac ∗ (B.3) where Da= k1st L ⟨u1⟩ξ2 is the Damköhler number, representing the ratio of the residence time to the reaction time scale. Integrating Equation B.3 with respect toξ∗ 1 subject to the boundary condition c∗(0, ξ∗ 2)=1 yields the general solution: c∗(ξ∗ 1, ξ∗ 2)=exp −Da·ξ ∗ 1 u∗ 1(ξ∗ 2) ! (B.4) Appendix B.1.1. Analytical Solutions for ...
[31]

presents an analytical solution for gas–liquid mass transfer in a bubble column reactor, focusing on the transient response during gassing-in experiments. Assuming plug flow for the liquid phase and a steady gas phase, the model incorporates interface mass transfer governed by Henry’s law, as well as static pressure effects on gas solubility and the dynam...