Multiscale Mechanics of Active Transient Fiber Networks: From Turnover-Driven Fluctuations to Stochastic Constitutive Theory

The structural backbone of living cells is a filamentous network, dynamically crosslinked and driven by molecular motors to generate internal forces, that continuously replaces its own constituents over minutes to hours. Such active transient fiber networks offer resilience and reconfigurability, a combination rare in engineered materials but desirable where a structure must bear load while it adapts, repairs, or morphs. Yet how this remarkable balance on the macroscale emerges from microscale processes, including active force generation and continuous scaffold rebuilding, remains elusive. This award supports fundamental theoretical and computational research to uncover this multiscale connection and thereby advance the science and engineering of active transient materials. Because the same networks govern how cells divide and migrate, the resulting understanding informs the mechanics of development, wound healing, and diseases such as cancer metastasis and fibrosis, besides raising the potential of engineered material applications in soft robotics, adaptive structures, and sustainable materials. The project also supports training of undergraduate and graduate students in the mechanics of active materials, hands-on educational outreach activities, such as workshops for students and exhibits for public, and dissemination of open-source network mechanics simulation tools for the engineering community. Continuum descriptions of active transient fiber networks rest on a classical assumption, that microstructural disorder self-averages to a deterministic response beyond a representative volume. This generally holds for passive networks, but its validity remains unclear when a network is driven out of equilibrium and continuously remodeled. This research develops a multiscale framework linking the microscopic dynamics of these networks to their scale-dependent response, combining discrete, agent-based simulations with data-driven coarse-graining. These simulations first resolve how crosslinker kinetics and turnover govern the spatial organization and lifetime of the force-bearing backbone, and how remodeling modulates the mean active stress. The statistics and scale-dependence of active stress fluctuations are then quantified across motor density, turnover rate, and crosslinker type, revealing the scale below which fluctuations dominate and above which a deterministic constitutive law suffices. Data-driven coarse-graining synthesizes these results into a stochastic constitutive theory based on the conformation tensor, in which active driving terms and state-dependent noise are learned from simulation data subject to thermodynamic and symmetry constraints, predicting both mean rheology and fluctuation statistics from microscopic parameters. Together, these advances deliver a bottom-up, thermodynamically consistent alternative to phenomenological constitutive models, where the mean and fluctuating mechanics follow systematically from molecular parameters. The resulting framework offers a predictive foundation for the rational design of engineered active materials and for an improved understanding of the mechanics of living systems that inspire them. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.