The strongest glasses have an Achilles' heel that causes them to fail catastrophically when pushed past their limit. They do not bend or stretch, as all damage concentrates into a single plane and the material fails in an instant. This brittleness has long capped the usefulness of high-stability amorphous solids, from bulk metallic glasses to engineered metamaterials.
Rashmi Priya and Smarajit Karmakar from the Tata Institute of Fundamental Research (TIFR), Hyderabad , in collaboration with Jürgen Horbach from Heinrich Heine University (HHU), Düsseldorf , report a potential way around this problem: 'lacing' the glass with self-propelled particles while it is being sheared markedly reduces its brittleness. This also allows the material to bear higher stress, making it stronger than before. This theoretical framework strengthens our understanding of brittleness in glasses and links shear to self-propulsion.
Where does the brittleness come from?
Unlike a crystal, a glass has no repeating atomic pattern. Its particles are disordered, much as they were in the liquid from which the glass formed, but are trapped in place. How firmly they are trapped depends on how the glass is prepared. Let us picture a rugged landscape of hills and valleys, where each valley represents one possible arrangement of the particles. A slowly cooled or well-aged glass settles into a deep valley. It is stable and difficult to perturb, but brittle when pushed past its limit. A rapidly cooled glass is caught in a shallower valley. Being less stable, it is weaker but more ductile.
The difference shows up under shear: imagine fixing the bottom of a block and pushing its top sideways. In a brittle glass, deformation remains suppressed until the stress reaches a large yield value. The particles suddenly organise into a shear band: a thin plane of intense rearrangement slicing across the whole sample, with a sudden drop in stress. Less stable glasses usually deform more gradually; rearrangements occur across the material, but these glasses cannot bear much load. The strength and ductility of a system trade against each other, and the researchers asked whether this trade-off could be changed after the glass had already been prepared.
From one crack to a network
The researchers took a theoretical approach and simulated how a glass deforms when doped with a small fraction of self-propelled particles (SPPs). SPPs are particles that carry their own fuel and push themselves along, like bacteria crowded in a suspension or synthetic colloids that swim when lit. Under the right conditions, the active particles do not soften the glass; in fact, the stress-strain curve, which is the material's mechanical fingerprint, turns from a cliff into a rounded hill. The glass yields later, and at higher stress. The single plane where damage once accumulated is now replaced by a network that spreads it across many smaller bands, which gradually connect. Activity changes not just when the glass fails, but how it fails.
This behaviour arises from a competition among the time over which shear deforms the glass, the time for which active particles keep moving in one direction, and how fast a shear band propagates. External shear and activity together compete with shear-band propagation, determining whether deformation becomes concentrated in a single band or spreads across multiple bands. Behind this lies an interesting equivalence: how quickly or slowly a glass is sheared can be traded against the local active force. As a result, a rapidly sheared glass with weak activity can yield like a slowly sheared glass with strong activity. This equivalence also persists under creep, where a fixed stress is applied and increasing activity delays flow and lowers the rate at which the glass deforms.
The key lies in the persistence time: how long each active particle keeps pushing in one direction before reorienting. When the persistence time is short, a swimmer rattles in place among its neighbours, never pushing long enough to escape the cage they form around it. That rattling is what strengthens the glass, and what produces the shear-band network. Given long enough to travel in one direction, the same swimmer can break out of its cage, making the glass easier to deform and more prone to flow.
Why it matters, and where next
This study establishes a preparation-independent route to control the failure mechanism by introducing and tuning an additional internal timescale through active doping. It also deepens a growing conceptual bridge between active matter and the mechanics of disordered solids.
For now, the work is computational, so the next step is to test these predictions at the laboratory bench. Dense colloidal systems doped with photoswitchable active particles are a promising testbed, though the experimental routes available today remain largely confined to gels and colloidal suspensions rather than to dense solid glasses. Beyond the laboratory lies the biological horizon, where living tissues constantly experience external forces and can tune their internal activity to withstand them. Whether the compensatory picture established here extends to such systems remains an open question.
