Hostname: page-component-7b7567c8c-2n9c9 Total loading time: 0 Render date: 2026-11-16T18:00:25.415Z Has data issue: false hasContentIssue false

Liquid plugs in narrow tubes: application to airway occlusion

Published online by Cambridge University Press:  04 November 2024

Georg F. Dietze*
Affiliation:
Université Paris-Saclay, CNRS, FAST, 91405 Orsay, France
*
†Email address for correspondence: georg.dietze@cnrs.fr

Abstract

We study liquid plugs in the pulmonary airways based on the two-phase axisymmetric weighted residual integral boundary-layer model of Dietze et al. (J. Fluid Mech., vol. 894, 2020, A17), which was originally developed to study liquid films coating the inner surface of a cylindrical tube in interaction with a core gas flow. The augmented form of this model, which was never applied beyond a proof of concept, allows for the representation of liquid pseudo-plugs. Here, we demonstrate its predictive power vs experiments and direct numerical simulations, in terms of the dynamics of plug formation and the characteristics of developed liquid plugs, such as their shape, flow field, speed and length, as well as the associated wall stresses and their spatial derivatives. In particular, we show that the augmented model allows us to establish a direct continuation path from travelling-wave solutions (TWS) to travelling-plug solutions (TPS). We then apply the model to predict mucus plugs in the conducting zone of the tracheobronchial tree, based on the lung architecture model of Weibel. We proceed by numerical continuation of travelling-state solutions in terms of the airway generation, whereby we impose the wavelength of the linearly most-amplified convective instability (CI) mode or that of the absolute instability (AI) mode. We identify the critical airway generation for liquid-plug formation (TWS/TPS transition), maximum potential for wall-stress-induced epithelial cell damage and CI/AI transition, and investigate how these phenomena are affected by the main control parameters, i.e. airway orientation vs gravity, air flow rate, mucus properties and airway size.

Information

Type
JFM Papers
Copyright
© The Author(s), 2024. Published by Cambridge University Press
Figure 0

Figure 1.Liquid plugs (subscript $\mathrm {1}$) enclosing a gas bubble (subscript $\mathrm {2}$) formed by the occlusion of a narrow cylindrical tube by a falling liquid film lining its inner surface. (a) Problem sketch and notations. The tube radius $R^\star$ is used as the length scale, the star superscript designating dimensional quantities; (b) travelling-state solution obtained from our model (2.13) for a vertical configuration: ${\textit {Ka}}=121.4$ (silicone oil II in table 1), $R^\star =1.5\,{\rm mm}$, $L=\varLambda =5.4$, $V_1/{\rm \pi} /R^3=2.85$, $M=1$, $d_{plug}=0.01$, ${\textit {Re}}_1=14.6$, ${\textit {Re}}_2=30.8$, where L, $\varLambda, V_1, M, d_{plug}$ and $\textit{Re}$ denote the domain length, wavelength, liquid volume, normalized gas pressure drop (2.18), pseudo-plug core radius (2.13c) and Reynolds number, respectively. Streamlines in the moving reference frame within the liquid (blue lines) and gas (red lines). Dot-dashed green lines correspond to spherical-cap reconstruction (2.14) between patching points marked by asterisks.

Figure 1

Table 1.Properties of the fluids used in our computations. The Kapitza number, ${\textit {Ka}}$, is defined as ${\textit {Ka}}=\sigma /(\rho _1 g^{1/3} \nu _1^{4/3})$, where $\sigma$, $\rho _1$ and $\nu _1$ denote the surface tension and the density and kinematic viscosity of the liquid, and $g=9.81\,{\rm m}\,{\rm s}^{-2}$ designates the gravitational acceleration.

Figure 2

Figure 2.Spatial linear stability of an annular liquid film in contact with air. Symbols: WRIBL; solid lines: Orr–Sommerfeld. (a) Falling liquid film. Circles: $R^\star =1.5\,{\rm mm}$, ${\textit {Ka}}=121.4$ (silicone oil II/air I in table 1), ${\textit {Re}}_1=15.4$, $M=1$; asterisks: run 13 in Dao & Balakotaiah (2000), $R^\star =3.175\,{\rm mm}$, ${\textit {Ka}}=3.5$ (glycerol(89 %)-water/air I), ${\textit {Re}}_1=0.258$, $M=1$, diamonds: experiment from figure 3(c) of Camassa et al. (2014), ${\textit {Ka}}=3.3\times 10^{-3}$ (silicone oil I, air I in table 1), $R^\star =5\,{\rm mm}$, ${\textit {Re}}_1=4.5\times 10^{-4}$, $M=1$; (b) stability of a pseudo-plug obtained with our augmented WRIBL model (2.13c). Silicone oil film from panel (a): $d_0=d_{plug}=0.01$, $\lambda =1$. Dashed: $\varPi _{CRL}=1$, $\varPi _\mu =\varPi _\rho =0$; dot-dashed: $\varPi _{CRL}=1.01$, $\varPi _\mu =\varPi _\rho =0$; solid: $\varPi _{CRL}=1$ ($-k_i$ has been multiplied by $10^3$).

Figure 3

Figure 3.Transition from TWS to TPS. The TSS are based on our augmented WRIBL model (2.13a). Annular liquid film in contact with air within a vertical cylindrical tube: ${\textit {Ka}}=121.4$ (silicone oil II and air I in table 1), $R^\star =1.5\,{\rm mm}$, $M=1$, $\varLambda =5.4$. (a) Minimal core radius, $d_{min}$, in terms of the normalized liquid volume. Solid blue: $\varPi _{CRL}=\lambda =1$, $d_{plug}=0.01$; dashed red: $\varPi _{CRL}=0$; (b) profiles of TSS corresponding to crosses in panel (a). Toward the tube axis: $V_1/{\rm \pi} /R^3=1, 2, 2.5, 2.5, 2.85$. Dot-dashed blue lines correspond to spherical-cap reconstruction (2.14) between the patching points (blue asterisks), where $|\partial _xd|=\epsilon ^{max}=0.75$.

Figure 4

Figure 4.Transient periodic computation of a liquid plug forming from an initially quasi-uniform film. Parameters according to figure 3: $V_1/{\rm \pi} /R^{3}=2.85$, $\epsilon _{I}=0.015$. (a) Time traces of the instantaneous Reynolds numbers, $\bar {q}^\star _k/\nu _k$, with $\bar {q}_k=\varLambda ^{-1}\int _0^\varLambda q_k\,{{\rm d}\kern0.7pt x}$, in the liquid (solid) and gas (dashed). Dot-dashed green lines: final values for the DNS in figure 6(b); (b) profiles corresponding to crosses in panel (a) (solid) and to the TSS in figure 3(b) (dashed).

Figure 5

Figure 5.The TSS from figure 3 (curves) compared with our own DNS (symbols) using Gerris (Popinet 2009): ${\textit {Ka}}=121.4$, $R^\star =1.5\,{\rm mm}$, $M=1$, $\varLambda =5.4$. Solid blue curve: full model (2.13), dot-dashed black curve: inertialess limit ($S_k=F_{kj}=G_{kj}=0$). (a) Plug celerity $c$; (b) gas Reynolds number, ${\textit {Re}}_2$, quantifying the gas flow rate transported by the gas bubbles in between two liquid plugs.

Figure 6

Figure 6.The TPS forming in a vertical cylindrical tube. Streamlines in the reference frame moving with the plug speed, $c$. Parameters according to figure 3(a): $\varLambda =5.4$, $V_1/{\rm \pi} /R^3=2.85$, $M=1$. (a) Transient periodic computation with our augmented WRIBL model (2.13): ${\textit {Re}}_1=30.8$, ${\textit {Re}}_2=14.6$, $c^\star =0.31\,{\rm m}\,{\rm s}^{-1}$. Dashed green lines correspond to spherical-cap approximation (2.14) between patching points (marked by green asterisks), using $\varepsilon ^{max}=0.75$; (b) periodic DNS with Gerris: ${\textit {Re}}_1=30.4$, ${\textit {Re}}_2=14.4$, $c^\star =0.30\,{\rm m}\,{\rm s}^{-1}$.

Figure 7

Figure 7.Liquid stresses at the tube wall, $r=1$, within a TPS. Parameters according to figure 6. Open green circles: DNS; solid black lines: WRIBL model. (a) Liquid–gas interface, showing gas bubble in between two liquid plugs; (b) pressure difference across the tube, $\Delta p_{w}$ (4.1); (c) axial wall shear stress derivative, $\partial _x\tau _{w}$, according to (4.2b); (d) axial wall pressure derivative, $\partial _x p_{w}$, according to (4.2a). Crosses mark patching points for spherical-cap reconstruction (2.14), using $\varepsilon ^{max}=0.75$, plus signs mark apex of spherical cap and diamonds mark limits of pseudo-plug.

Figure 8

Figure 8.Highly viscous falling liquid film in contact with air: ${\textit {Ka}}=3.3\times 10^{-3}$ (silicone oil I, air in table 1), $R^\star =5\,{\rm mm}$, ${\textit {Re}}_1=4.5\times 10^{-4}$, $M=1$. (a) Experiment from figure 3(c) of Camassa et al. (2014), reproduced here with permission from Cambridge University Press. (b) Open-domain WRIBL computation (true to scale) using inlet noise (3.21): $\epsilon _1=0$, $\epsilon _2=10^{-4}$. Leading and trailing portions of the liquid plugs were reconstructed with the spherical-cap approximation (2.14), using $\epsilon ^{max}=0.68$.

Figure 9

Figure 9.The TSS at $f=f_{max}$ and $M=1$ obtained with our WRIBL model (2.13) for the configuration in figure 8. Solid blue curves with circles: full model; dot-dashed black: $J_k=K_k=L_k=M_k=0$ in (2.13). Filled black diamonds mark AI limit. Lower branch beyond LP2 in panel (a) corresponds to TPS, which are compared with experiments (filled red squares) from figures 3(c) (${\textit {Re}}_1=4.5\times 10^{-4}$) and 3(d) (${\textit {Re}}_1=6.22\times 10^{-4}$) of Camassa et al. (2014). (a) Minimal core radius; (b) wavelength; (c) wave/plug speed; (d) plug profiles corresponding to cross and asterisk in panels (a–c). Solid: ${\textit {Re}}_1=4.5\times 10^{-4}$, $f^\star _{max}=0.145\,{\rm Hz}$; dashed: ${\textit {Re}}_1=6.22\times 10^{-4}$, $f^\star _{max}=0.155\,{\rm Hz}$.

Figure 10

Figure 10.Airway occlusion under imposed gas flow rate for different $h_{VE}$. Linearly most-dangerous TSS vs airway generation $n$ based on lung architecture (5.1) of Weibel & Gomez (1962): ${\textit {Ka}}=30.6$ (mucus I and air I in table 1), ${\textit {Re}}_{20}=\pm 590$, $k=k_{max}$, $R^\star =R^{\star }_{\textit{Weib}}$, ${\textit {Re}}_2={\textit {Re}}_2^{\textit{Weib}}$. Crosses mark transition between CI, where $\omega _i=\partial _\omega k_i=0$ is imposed (3.9a), and AI, where ${\rm d}\omega /{\rm d}k=0$ (with $k$, $\omega \in \mathbb {C}$) is imposed (3.9b). Filled circles to open diamonds: $h_{VE}=0.4$, 0.3, 0.267, 0.24, 0.2 and 0.15. (a,c) Co-current configuration: ${\textit {Re}}_{20}=590$, $M_{dry}<0$; (b,d) counter-current configuration: ${\textit {Re}}_{20}=-590$, $M_{dry}>0$. (a,b) Minimum core radius; (c,d) normalized pressure drop, where $M_{dry}$ (5.3) corresponds to dry airways ($h_{VE}=0$).

Figure 11

Figure 11.Airway occlusion under imposed pressure drop for different values of the liquid holdup, $h_{VE}$. Linearly most-dangerous TSS vs airway generation, $n$, based on lung architecture (5.1) of Weibel & Gomez (1962): ${\textit {Ka}}=30.6$ (mucus I and air I in table 1), $M/M_{dry}=20$, ${\textit {Re}}_{20}^{dry}=590$ ($M_{dry}<0$), $k=k_{max}$, $R^\star =R^{\star }_{\textit{Weib}}$. Filled circles to open diamonds: $h_{VE}=0.4$, 0.3, 0.267, 0.24, 0.2 and 0.15. (a) Minimum core radius; (b) normalized gas Reynolds number. $M_{dry}$ (5.3) and ${\textit {Re}}_2^{dry}$ correspond to dry airways ($h_{VE}=0$) in the absence of gravity (${\textit {Fr}}^{-1}=0$). The LPs mark onset of TPS. Crosses mark transition between CI, where $\omega _i=\partial _\omega k_i=0$ is imposed (3.9a), and AI, where ${\rm d}\omega /{\rm d}k=0$ (with $k$, $\omega \in \mathbb {C}$) is imposed (3.9b).

Figure 12

Figure 12.Length scales of TSS from figure 10, as compared with the airway length, $L_{\textit{Weib}}$ (5.1). (a,b) Relative distance of propagation, $L_{T}/L_{\textit{Weib}}$, during half of one breathing period, 1/$f_{T}$, where $L_{T}=c$/2/$f_{T}$; (c,d) relative wavelength, $\varLambda /L_{\textit{Weib}}$. (a,c) Co-current: ${\textit {Re}}_{20}=590$; (b,d) counter-current: ${\textit {Re}}_{20}=-590$.

Figure 13

Figure 13.Liquid volume, $V_1$, of TSS in figure 10(a) compared with different static dewetting thresholds, $V_{U}$ (5.4) and $V_{P}$ (5.5). From filled circles to open diamonds: $h_{VE}=0.4$, 0.3, 0.267, 0.24, 0.2 and 0.15. (a) Compared with critical volume for the existence of unduloids: $V_{U}=1.73{\rm \pi} R^3$ (5.4); (b) compared with critical volume for the existence of fully wetting spherical liquid plugs: $V_{P}={\rm \pi} R^3(\varLambda -4/3)$ (5.5); (c,d) Bond numbers for liquid unduloids, ${\textit {Bo}}_{U}=\rho _1gh^{\star 2}_{VE}/\sigma$, and liquid plugs, ${\textit {Bo}}_{P}={\textit {Bo}}=\rho _1g R^{\star 2}/\sigma$.

Figure 14

Figure 14.Effect of air flow and mucus properties on airway occlusion. The TSS for reference parameters from figure 10: $k=k_{max}$, $R^\star =R^{\star }_{\textit{Weib}}$, ${\textit {Re}}_2={\textit {Re}}_2^{\textit{Weib}}$, $h_{VE}/R=0.24$, air I in table 1. Circles: $n=10$, pentagons: $n=11$, squares: $n=12$, diamonds: $n=13$ asterisks: $n=14$. Reference state (thin vertical lines): ${\textit {Ka}}=30.6$, ${\textit {Re}}_{2 0}=\pm 590$. (a) Versus ${\textit {Re}}_{20}$, co-current; (b) vs ${\textit {Re}}_{20}$, counter-current; (c) vs ${\textit {Ka}}$ at fixed ${\textit {Ga}}=R^{\star 3}g/\nu _1^2$: ${\textit {Re}}_{2 0}=590$; (d) vs ${\textit {Ka}}$ at fixed ${\textit {Ga}}$: ${\textit {Re}}_{2 0}=-590$. Pluses in panels (a,b) mark turbulence onset.

Figure 15

Figure 15.Effect of mucus surface tension and viscosity on airway occlusion. The TSS for reference parameters of co-current configuration in figure 10: air I in table 1, $k=k_{max}$, $R^\star =R^{\star }_{\textit{Weib}}$, ${\textit {Re}}_2={\textit {Re}}_2^{\textit{Weib}}$, $R_0^\star =9\,{\rm mm}$, $Q_{20}^\star =2.5\times 10^{-4}\,{\rm m}^3\,{\rm s}^{-1}$, $h_{VE}/R=0.24$. Circles: $n=10$, pentagons: $n=11$, squares: $n=12$, diamonds: $n=13$, asterisks: $n=14$. (a) Change in ${\textit {La}}=R^\star \sigma \rho _1/\mu _1^2$ at fixed ${\textit {Ga}}=R^{\star 3}g/\nu _1^2$; (b) change in ${\textit {La}}$ at fixed ${\textit {Bo}}=\rho _1 g R^{\star 2}/\sigma$; (c) data from panel (a) vs ${\textit {Ca}}=\mathcal {U}_2\mu _1/\sigma$; (d) data from panel (b) vs ${\textit {Ca}}$. Absolute instability everywhere except on branch to the right of blue cross. In all panels, open circles correspond to ${\textit {Ka}}=30.6$, i.e. mucus I in table 1.

Figure 16

Figure 16.Self-similar airway contraction/expansion based on lung architecture model (5.1) of Weibel & Gomez (1962). Effect of $R_0^\star$ on TSS: $k=k_{max}$, $R^\star =R^{\star }_{\textit{Weib}}$, ${\textit {Re}}_2={\textit {Re}}_2^{\textit{Weib}}$, ${\textit {Ka}}=30.6$ (mucus I and air I in table 1), $h_{VE}=0.24$. Circles: $R_0^\star =11\,{\rm mm}$, pentagons: $R_0^\star =10\,{\rm mm}$, squares: $R_0^\star =9\,{\rm mm}$, diamonds: $R_0^\star =8\,{\rm mm}$, asterisks: $R_0^\star =7\,{\rm mm}$. (a) Co-current configuration: $M_{dry}<0$, $Q_{20}^\star =2.5\times 10^{-4}\,{\rm m}^3\,{\rm s}^{-1}$; (b) counter-current configuration: $M_{dry}>0$, $Q_{20}^\star =-2.5\times 10^{-4}\,{\rm m}^3\,{\rm s}^{-1}$.

Figure 17

Figure 17.Wall stresses for TSS in figure 10(a). Co-current configuration: ${\textit {Re}}_{20}=590$, ${\textit {Ka}}=30.6$, $k=k_{max}$, $R^\star =R^{\star }_{\textit{Weib}}$, ${\textit {Re}}_2={\textit {Re}}_2^{\textit{Weib}}$. Crosses mark CI/AI transition and other symbols correspond to LPs in figure 10(a), marking the onset of TPS (except for open diamonds). Filled circles to asterisks: $h_{VE}=0.4$, 0.3, 0.267, 0.24, 0.2, and 0.15. (a) Maximum magnitude of wall shear stress, $\tau _{w}$ (4.2b); (b) maximum magnitude of excess pressure, $\Delta p_{w}$ (4.1); (c,d) maximum and minimum values of $\tau _{w}$.

Figure 18

Figure 18.Maximum magnitudes of wall stress derivatives for TSS in figure 10(a). Same parameters and attributions as in figure 17. (a,c) Wall shear stress $\tau _{w}$ according to (4.2b); (b,d) wall pressure $p_{w}$ according to (4.2a); (a,b) spatial derivative $\partial _x$; (c,d) temporal derivative $\partial _t=-c\partial _x$.

Figure 19

Figure 19.Maximum magnitudes of spatial wall stress derivatives for TSS in figure 10(b). Counter-current configuration: ${\textit {Re}}_{20}=-590$, ${\textit {Ka}}=30.6$, $k=k_{max}$, $R^\star =R^{\star }_{\textit{Weib}}$, ${\textit {Re}}_2={\textit {Re}}_2^{\textit{Weib}}$. Crosses mark CI/AI transition and other symbols correspond to LPs in figure 10(a), marking the onset of TPS (except for open diamonds). Filled circles to asterisks: $h_{VE}=0.4$, 0.3, 0.267, 0.24, 0.2 and 0.15. (a) Derivative of wall shear stress, $\tau _{w}$, according to (4.2b); (b) derivative of wall pressure, $p_{w}$, according to (4.2a).

Figure 20

Figure 20.Effect of different control parameters on the maximum magnitude of the spatial wall shear stress derivative, $\partial _x\tau _{w}$ (4.2b). Parameters and symbols as in figures representing corresponding TSS. (a) Effect of air flow: co-current configuration. The TSS according to figure 14(a); (b) effect of air flow: counter-current configuration. The TSS according to figure 14(b); (c) change in ${\textit {Ka}}$ at fixed ${\textit {Ga}}$: co-current. The TSS according to figure 14(c); (d) effect of trachea radius $R_0^\star$: co-current. The TSS according to figure 16(a).