Nonlinear dynamics of air invasion in one-dimensional compliant fluid networks

Vascular networks exhibit a remarkable diversity of architectures and transport mechanisms across biological systems. Inspired by embolism propagation in plant xylem, where air invades water-filled conduits under negative pressure, we study air penetration in compliant one-dimensional hydrodynamic networks experiencing mass loss by pervaporation. Using a theoretical framework grounded in biomimetic models, we show that embolism dynamics are shaped by the interplay between network compliance and viscous dissipation. In particular, the competition between two timescales (the pressure diffusion time, $τ_\mathrm{diff}$, and the pervaporation time, $τ_\mathrm{pv}$) governs the emergence of complex, history-dependent behaviors. When $τ_\mathrm{diff} \sim τ_\mathrm{pv}$, we uncover a nonlinear feedback between the internal pressure field and the embolism front, leading to transient depressurization and delayed interface motion. These results offer a minimal framework for understanding embolism dynamics in slow-relaxing vascular systems and provide design principles for soft microfluidic circuits with tunable, nonlinear response.

💡 Research Summary

The authors present a theoretical study of air invasion (embolism) in a one‑dimensional compliant fluid network that loses liquid volume through pervaporation. The network consists of a series of micro‑channels fabricated in PDMS, each channel being connected to its neighbor by a narrow constriction. Liquid loss through the thin PDMS wall is described by a pervaporation flux j that depends on channel geometry, material diffusivity, saturation concentration, and ambient relative humidity. As water evaporates, the pressure inside each channel drops; because the PDMS walls are elastic, the pressure drop is coupled to a volume change through a compliance C, derived from thin‑plate theory and expressed in terms of channel length, width, wall thickness, Young’s modulus and Poisson’s ratio.

Fluid flow between adjacent channels occurs only through the constrictions, whose hydraulic resistance R is calculated from Stokes flow in a rectangular slit. The pressure difference across a constriction drives water flow, while the air‑water interface can cross the constriction only when the downstream pressure falls below a Laplace pressure threshold p_c, which is set by the surface tension of water and the curvature of the constriction.

The dynamics of each channel’s pressure p_i is governed by a discrete diffusion‑like equation:

 dp_i/dt = (p_{i+1} – p_i)/RC – (p_i – p_{i‑1})/RC – j l/C,

where the first two terms represent pressure diffusion through neighboring resistances and the last term accounts for the uniform pervaporation loss. When the embolism front is halted at a constriction, the pressure of the upstream channel follows a similar equation but with only one neighboring term. Once p_i drops below p_c, the front jumps forward, the channel volume instantly adjusts to V = l + C p_c (elastic relaxation is assumed instantaneous), and the water volume in that channel then evolves according to a balance of flow through the constriction and ongoing pervaporation.

Two characteristic timescales emerge from the model:

1. Pressure diffusion time τ_diff = N₀² R C, which measures how fast pressure equilibrates along the whole series of N₀ channels.
2. Pervaporation time τ_pv = V/(j l), which measures how quickly the total liquid volume is depleted by evaporation.

When τ_diff ≪ τ_pv (fast pressure diffusion), the pressure field remains nearly uniform; the front advances as soon as the local Laplace condition is met, producing an almost exponential (truncated) progression that matches earlier work on low‑resistance xylem analogues. In this regime, viscous dissipation in the constrictions is negligible compared with the global pressure drop, and the embolism front exhibits a smooth, monotonic advance.

When τ_diff ≈ τ_pv or τ_diff > τ_pv (slow pressure diffusion), a qualitatively different behavior appears. Pressure gradients become localized near the front, and the pervaporation‑induced global depressurization competes with the viscous pressure drop across the resistive network. The system then experiences a transient pressure minimum (p_min) as the front stalls, followed by a rapid “catch‑up” when the overall pressure falls sufficiently low to overcome the Laplace barrier. This creates an inflection point in the front‑versus‑time curve, reminiscent of the intermittent embolism propagation observed in real leaves (sub‑minute bursts separated by hour‑long pauses). Increasing the hydraulic resistance R (by narrowing w_c or lengthening l_c) amplifies τ_diff, making the delay more pronounced and, for sufficiently large R, can almost arrest the front. Conversely, reducing R restores the fast‑diffusion regime.

The authors also develop a continuous limit of the discrete equations, showing that the pressure field obeys a diffusion equation with a source term proportional to the pervaporation flux. Analytical asymptotic solutions are obtained for the two limiting regimes, confirming the numerical findings.

Importantly, all model parameters (channel width w, height h, wall thickness δ, Young’s modulus E, Poisson’s ratio ν, constriction dimensions w_c, l_c, ambient humidity RH, etc.) are experimentally accessible. This makes the framework a practical design tool for soft microfluidic circuits: by tuning compliance C and resistance R, one can engineer devices that either quickly propagate a pressure front, exhibit delayed, history‑dependent motion, or even store mechanical memory through hysteresis.

In summary, the paper demonstrates that the interplay between pressure diffusion and pervaporation introduces a nonlinear feedback loop that governs embolism dynamics in compliant networks. This minimal model captures key features of plant xylem embolism—intermittent propagation, pressure depressurization, and dependence on anatomical resistance—while also offering quantitative guidelines for designing biomimetic fluidic systems with tunable, nonlinear responses.