Project Title: Comprehensive study of 3D liquid flow fields in additively manufactured structures for SMART reactors using large-scale vertical magnetic resonance imaging and computational fluid dynamics - Project: SFB1615 - Subproject: B03- Magnetic resonance imaging of large-scale multiphase and reactive flow systems - Project: SFB1615 - Subproject: C01-Integration of components into adaptive geometries Related Publication: https://doi.org/10.48550/arXiv.2602.09750 https://doi.org/10.1016/j.cej.2026.176536 https://doi.org/10.15480/882.16859 Authors: Timo Merbach (https://orcid.org/0000-0002-7723-5444) (Institute of Multiphase Flows, Hamburg University of Technology, Eißendorfer Straße 38, 21073 Hamburg, Germany) Muhammad Adrian (https://orcid.org/0009-0008-3634-9518) (Institute of Process Imaging, Hamburg University of Technology, Denickestraße 17, 21073 Hamburg, Germany) Christoph Wigger (https://orcid.org/0009-0006-2648-9184) (Institute of Multiphase Flows, Hamburg University of Technology, Eißendorfer Straße 38, 21073 Hamburg, Germany) Selma Iraqi Houssaini (https://orcid.org/0009-0008-9856-1776) (Institute of Multiphase Flows, Hamburg University of Technology, Eißendorfer Straße 38, 21073 Hamburg, Germany) Benedict Bayer (https://orcid.org/0009-0008-7896-7322) (Institute of Multiphase Flows, Hamburg University of Technology, Eißendorfer Straße 38, 21073 Hamburg, Germany) Artyom Tsanda (https://orcid.org/0009-0009-7765-4604) (Institute for Biomedical Imaging, Hamburg University of Technology, Am Schwarzenberg-Campus 3, 21073 Hamburg, Germany and Section for Biomedical Imaging, University Medical Center Hamburg-Eppendorf, Lottestraße 55, 22529 Hamburg, Germany) Serhan Acikgöz (https://orcid.org/0009-0002-8643-192X) (Institute for Industrialization of Smart Materials, Hamburg University of Technology, Harburger Schloßstraße 28, 21079 Hamburg, Germany) Funding Acknowledgement: This dataset was generated as part of the DFG-funded project CRC 1615: SMART Reactors for Future Process Engineering (DFG Project Number: 503850735). ________________________________________ 1. General Information Dataset Title: Data supplement for publication: Comprehensive study of 3D liquid flow fields in additively manufactured structures for SMART reactors using large-scale vertical magnetic resonance imaging and computational fluid dynamics Short Description: The TUHH Open Research (TORE) repository contains raw and processed data associated with the publication "Comprehensive study of 3D liquid flow fields in additively manufactured structures for SMART reactors using large-scale vertical magnetic resonance imaging and computational fluid dynamics". The dataset comprises experimental magnetic resonance imaging velocimetry data, numerical results from OpenFOAM simulations, and corresponding design files. The publication investigates the flow behaviour in additively manufactured TPMS structures using a large-bore vertical 3 T MRI system. Cross-validation between experimental measurements and numerical simulations is performed. Three different TPMS geometries are analysed within the Darcy-Forchheimer regime. Date of Data Collection: [2025 – 2026] Geographical Coverage: Germany Keywords: Porous media, Magnetic resonance imaging, Computational fluid dynamics, Triply periodic minimal surfaces ________________________________________ 2. Methodological Information Data Collection and Processing: Design: The design of Gyroid TPnS and Schwarz-Diamond TPSf structures is conducted using nTopology 5.31.3, a specialized software platform for advanced geometric modelling and simulation. Rectangular blocks are generated and cut into cylindrical geometries with an inner diameter of 38 mm and a length of 100 mm. These cylindrical structures are then integrated with a surrounding wall and clamping connectors. The .stl files required for manufacturing are generated directly within the nTopology software and are included in the dataset in the Design folder. -> Date of creation of the data and author: 05/2025; Timo Merbach, Serhan Acikgöz Manufacturing: The unrotated Gyroid TPnS modules are manufactured using an EOS Eosint P396 (EOS GmbH, Germany) PBF-LB/P system. The rotated Gyroid TPnS and Schwarz-Diamond TPSf structures are fabricated using a Formlabs Form 3+ system (Formlabs, USA), which employs VPP-UVL. Manufacturing has been executed at Fraunhofer Research Institution for Additive Manufacturing Technologies IAPT, Am Schleusengraben 14, 21029 Hamburg, Germany and Institute of Multiphase Flows, Hamburg University of Technology, Eißendorfer Straße 38, 21073 Hamburg, Germany. -> Date of creation of the data and author: 07/2025; Timo Merbach, Serhan Acikgöz o Experimental: Experimental measurements were conducted using magnetic resonance imaging to analyse the flow field of unrotated and rotated Gyroid TPnS and Schwarz-Diamond TPSf structures under varying operating conditions. Experimental data collected for this purpose are stored in HDF5 format. Data collection has been performed at Institute of Process Imaging, Hamburg University of Technology, Denickestraße 17, 21073 Hamburg, Germany. -> Structural integrity: The free cross-sectional area from design geometry is compared with the free cross-sectional area obtained from experimental data. Design files are stored as STL file. -> Flow field: The flow field is analysed in terms of mass flow rate and divergence using MATLAB R2024a. Visualization of the flow field, hydraulic tortuosity and vorticity are computed using MATLAB R2024a. Magnetic resonance imaging velocimetry data are provided in HDF5 data (Experimental folder). -> Date of creation of the data and author: 05/2025 – 07/2025; Timo Merbach, Muhammad Adrian, Artyom Tsanda, Selma Iraqi Houssaini, Benedict Bayer o Numerical: All numerical simulations are performed using the open-source software OpenFOAM (Version 2406) with the simpleFoam solver, steady state flow conditions as indicated by the experimental data. CFD simulations were conducted to analyse the flow field of unrotated Gyroid TPnS structures. The Workflow and the data structure are provided in Section 5. Simulation data are provided within the Numerical folder. Data collection has been performed at Institute of Multiphase Flows, Hamburg University of Technology, Eißendorfer Straße 38, 21073 Hamburg, Germany. -> Date of creation of the data and author: 03/2026 – 04/2026; Christoph Wigger Experimental Design: The experimental setup comprises a gear pump (Gather Industries, Germany), a Coriolis mass flow meter (Endress & Hauser, Germany), and a degassing unit (Binder GmbH, Germany) connected to a vacuum pump (Edwards Vacuum, UK). The flow passes through an acrylic pipe (D = 38 mm, L = 760 mm), followed by two TPMS modules (each 100 mm in length) connected via BioClamps (BioPure Technology Ltd, UK). The working conditions in terms of mass flow rate are (0.755 ± 0.03) kg/min (operating point 1), (1.15 ± 0.01) kg/min (operating point 2), (1.5 ± 0.01) kg/min (operating point 3) for the flow through the TPMS modules. MRI velocimetry measurements are performed on a large-bore 3 T vertical MRI system. MRI-derived velocity fields are quantified by using bipolar flow-encoding gradients. The pulse sequence used for data acquisition is the Quantitative Flow (QFlow) sequence provided by Philips (Philips N.V., Netherlands), based on a spoiled gradient-echo sequence. Numerical Design: The numerical flow simulations are conducted using the same design models that served as the basis for additive manufacturing. The computational domain comprises two cylindrical elements, supplemented by an additional pipe section to ensure the development of an established flow profile, together with an outlet section. Moreover, the simulations are performed under the same operating conditions as the experiments. Data Validation and Quality Assurance: Data accuracy, completeness, and consistency were ensured through a combination of experimental validation, physical plausibility checks, and numerical verification. In particular, fluid dynamic consistency was evaluated by comparing calculated mass flow rates from the velocity fields with independently measured mass flow rates, ensuring agreement within experimental uncertainty. Additionally, the divergence criterion was assessed to confirm mass conservation within the reconstructed flow fields. Structural integrity and geometric consistency were verified by comparing the processed data with the geometry files, including checks of free cross-sectional area. This ensured that the reconstructed and analysed datasets accurately represent the underlying physical structures. Furthermore, numerical simulations were conducted using grid-independent meshes, and all simulations followed established best-practice CFD guidelines. ________________________________________ 3. Data and File Overview The dataset is organised into three main folders: Design, Experimental, and Numerical. In addition, it includes this ReadMe file and a separate file describing the overall folder structure (Folder_structure.pdf). List of Files and Structure: Folder Description Format Size Design/ Structure files .stl 140 MB Experimental/ Experimental data .hdf5 255 MB Numerical/ Numerical data Various 245 MB File Naming Convention: The file naming conventions for the structures are as follows: Gyroid-TPnS_0: unrotated Gyroid TPnS Gyroid-TPnS_45: rotated Gyroid TPnS Schwarz-Diamond-TPSf: Schwarz-Diamond TPSf o File names follow a structured convention, for the design files: 20260430_CRC1615_C01_Gyroid-TPnS_0.stl -> 20260430 = date of creation -> CRC1615 = research project -> C01 = project number -> Gyroid-TPnS_0 = description of the structure -> .stl = file format The folders including the numerical data have an additional identifier that specifies the operating condition: Example: 20260430_CRC1615_C01_Gyroid-TPnS_0_op_point_1 -> 20260430 = date of creation -> CRC1615 = research project -> C01 = project number -> Gyroid-TPnS_0 = description of the structure -> op_point_1 = Operating point, flow conditions All operating conditions: -> Operating point 1: (0.755 ± 0.03) kg/min -> Operating point 2: (1.15 ± 0.01) kg/min -> Operating point 3: (1.5 ± 0.01) kg/min The experimental dataset is stored in HDF5 format and contains measurement data organized in hierarchical groups. The HDF5 file is organized into the following top-level groups: -> /op_point_1 Operating point 1: (0.755 ± 0.03) kg/min -> /op_point_2 Operating point 2: (1.15 ± 0.01) kg/min -> /op_point_3 Operating point 3: (1.5 ± 0.01) kg/min Each group corresponds to a specific configuration of the system. Each group contains two datasets representing numerical data arrays, namely magnitude and velocity data. Datasets are stored as multidimensional arrays and can be accessed using standard HDF5-compatible tools. The data within the HDF5 files are stored as four-dimensional matrices with dimensions (3 × 20 × 176 × 176), representing the velocity components (v, u, w) along the axial direction (z) and the two lateral directions (x, y). Number of Records: Three different structures have been analysed experimentally including unrotated and rotated Gyroid TPnS and Schwarz-Diamond TPSf. Three different operating points 1, 2, and 3 have been considered. A numerical study has been conducted for unrotated Gyroid TPnS for all operating points. ________________________________________ 4. Access and Licensing Information Repository and Persistent Identifier: Published via TORE, DOI: https://doi.org/10.15480/882.16072 License for Use: CC BY 4.0 Access Restrictions: Open ________________________________________ 5. Reproducibility, Software Dependencies, and Workflow of Numerical Simulations Software Required: nTopology 5.31.3, MATLAB R2024a, OpenFOAM-v2406 (Linux operating system) Scripts and Workflow for numerical simulations In this section, the structure of the considered OpenFOAM cases is outlined including boundary conditions, solver settings and material and mesh properties. o Initial conditions Initial and boundary conditions are included in 0/. In this case, boundary conditions have been defined for the velocity field U and pressure field p, as simpleFOAM is selected as a solver. o Material and Mesh Properties Information associated with material and mesh properties is included in the constant/ directory. Fluid properties are defined in transportProperties containing information about the dynamic viscosity of the liquid. Additionally, the mesh is saved in triSurface/ after performing the required mesh procedure. o Numerical Settings In system/ all numerical settings, solver details, and runtime controls are included. The general settings about the simulation are included in controlDict. Furthermore, solver settings are listed in fvSolution and discretisation schemes are defined in fvSchemes. During this simulation two different sets of discretisation schemes are employed. First a set of robust schemes is applied to ensure a stable start-up procedure. After a total number of 200 iterations, another set of discretisation schemes is employed to guarantee second order accuracy, while avoiding oscillatory or diverging behaviour. o Running a Case For every case, a .bash-file is packaged containing the necessary commands to perform the simulations. To run a simulation, OpenFOAM must be compiled before running the simulations. All simulations have been performed using OpenFOAM-v2406. It is therefore recommended to perform the simulations with this version to ensure reproducibility. Reproducibility Notes: All data required to reproduce the results are included in the dataset, comprising flow measurements, numerical simulation outputs, and geometry files. All analyses follow standard fluid dynamics principles, enabling direct comparison between experimental and numerical data. ________________________________________ 7. Ethical and Legal Aspects Data Protection: Not applicable, as the dataset does not contain any personal or sensitive data. Consent Statement: Not applicable, as no human participants or personal data are involved. ________________________________________ 8. Versioning and Updates Version Number: v1.0 Date of Release: [2026-04-30] ________________________________________ 9. Contact Information Corresponding Author: Name: Timo Merbach Institution: Institute of Multiphase Flows, Hamburg University of Technology Email: timo.merbach@tuhh.de ORCID: 0000-0002-7723-5444 Project Website: https://www.tuhh.de/sfb1615/welcome