Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Time-resolved X-ray radiography of through-thickness liquid transport in partly saturated needle-punched nonwovens

Patrick Wegele, Zisheng Yao, Jonas Tejbo, Julia K. Rogalinski, Tomas Rosén, Alexander Groetsch, Kim Nygård, Eleni Myrto Asimakopoulou, Pablo Villanueva-Perez, L. Daniel Söderberg

Abstract

Nonwoven fibre networks underpin filtration, insulation and geotextiles, where liquid uptake, redistribution and release govern performance. In needle-punched felts, barbed needles mechanically entangle fibres and partially reorient them toward the thickness direction ($z$), creating out-of-plane "pillars" and heterogeneity. While mechanical and structural consequences of needling are well documented, dynamic $z$-direction transport in partly saturated networks remains difficult to access due to opacity and sub-second timescales. Here we combine micro-CT ($μ$CT) of dry structure with time-resolved X-ray radiography during droplet addition to quantify through-thickness transport as a function of saturation and needling intensity, using a compact Washburn-type descriptor for dynamics. Results show an exponential dependence of $z$-directional liquid transport on saturation, consistent with previous models for in-plane relative permeability of nonwoven networks. Additionally, increased needle-punch intensity reorients fibres toward the $z$-direction, forming preferential flow pathways that enhance through-thickness transport, even as single-phase permeability decreases. These findings underscore needle-punch as a key design parameter for tuning liquid transport in nonwoven fibre networks. The approach provides an experimental and modelling framework for dynamic, capillarity-driven transport in opaque fibrous materials.

Time-resolved X-ray radiography of through-thickness liquid transport in partly saturated needle-punched nonwovens

Abstract

Nonwoven fibre networks underpin filtration, insulation and geotextiles, where liquid uptake, redistribution and release govern performance. In needle-punched felts, barbed needles mechanically entangle fibres and partially reorient them toward the thickness direction (), creating out-of-plane "pillars" and heterogeneity. While mechanical and structural consequences of needling are well documented, dynamic -direction transport in partly saturated networks remains difficult to access due to opacity and sub-second timescales. Here we combine micro-CT (CT) of dry structure with time-resolved X-ray radiography during droplet addition to quantify through-thickness transport as a function of saturation and needling intensity, using a compact Washburn-type descriptor for dynamics. Results show an exponential dependence of -directional liquid transport on saturation, consistent with previous models for in-plane relative permeability of nonwoven networks. Additionally, increased needle-punch intensity reorients fibres toward the -direction, forming preferential flow pathways that enhance through-thickness transport, even as single-phase permeability decreases. These findings underscore needle-punch as a key design parameter for tuning liquid transport in nonwoven fibre networks. The approach provides an experimental and modelling framework for dynamic, capillarity-driven transport in opaque fibrous materials.
Paper Structure (19 sections, 21 equations, 12 figures, 3 tables)

This paper contains 19 sections, 21 equations, 12 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Materials
  3. Methods
  4. Synchrotron Setup
  5. Evaluation of network saturation
  6. Derivation of Liquid Transport Function
  7. Introduction of a $z$-directional liquid transport index $T_z(\Theta)$
  8. Results and Discussion
  9. Principal behaviour
  10. Influence of the needle-punch intensity
  11. Conclusion
  12. Experimental uncertainty analysis
  13. Porosity $\phi_{0,i}$
  14. Pore Size Distribution
  15. Fibre Orientation $T_{zz}(z)$
  16. ...and 4 more sections

Figures (12)

  • Figure 1: Tomographic reconstruction and analysis of the samples used in the underlying experiment. See Appendix \ref{['Sec:App0']} for an uncertainty analysis of the dry network evaluations. a) Rendering of the reconstruction of samples Ref, High-NPI and Low-NPI. b) Pore size distribution of samples High-NPI and Low-NPI. Sample Ref is not investigated in terms of pore size and fibre orientation due to the differing composition, which does not allow a comparison to samples High-NPI and Low-NPI c) $zz$-component of the fibre orientation tensor as a function of the through-thickness coordinate $z$.
  • Figure 2: Schematic representation of the setup as used during the experiment. An X-ray beam is penetrating the top surface of the sample while subsequent droplets are released from the Kapton tube. The process is recorded on the X-ray detector, generating a stack of 2D-greyscale images. See Appendix \ref{['Appendix_A4']} for an uncertainty analysis of the experimental setup.
  • Figure 3: Flat-field corrected VanNieuwenhove2015 results as obtained on the X-ray detector for the first droplet penetrating dry sample Ref. The droplet is coming from the right and penetrating the network visible on the left. a) Dry state, reference for dry network. b): The droplet (indicated with a dashed yellow line) is arriving. c): Droplet burst, moment of absolute saturation in top level and reference for $\tau$=0 s for later model fitting. d): Liquid transport within the network decreases the absolute saturation in the FoV. e): Static situation where liquid has redistributed just before the next droplet is released.
  • Figure 4: Static X-ray projection scan of samples High-NPI and Low-NPI in the µCT after the experiment was conducted. Dark areas are the result of dried KI, indicating that the KI-stained liquid was present within these areas.
  • Figure 5: Evaluation algorithm to determine the mean greyscale value $\bar{I_j}$ for each frame of the processed stacks.
  • ...and 7 more figures