The paper proposes that the dark sector of the Universe can be a single substance that acts like ordinary dark matter early on and then changes into a kind of relativistic solid at late times to drive cosmic acceleration. The authors give a concrete realization of this idea with a generalized Chaplygin-type solid. In this scenario the same component supplies the gravitational pull needed to form structure at high redshift and the negative-pressure behavior that speeds up expansion at low redshift.

To build the model the researchers use an effective field theory description of a continuous medium. They track the motion of volume elements with three scalar fields that label each element. The dark Lagrangian is written in terms of one quantity that measures compression (called b) and two quantities that measure shape at fixed volume (called Y and Z). When the Lagrangian depends only on b the medium behaves like a perfect fluid (the dark‑matter‑like phase). Allowing dependence on Y and Z turns on shear or rigidity and produces the solid phase.

A key advantage of the solid description is how it changes the behavior of small perturbations. Earlier unified models based on perfect fluids—like the standard Chaplygin gas—run into trouble because their effective sound speed grows when the component begins to accelerate the expansion. That growth produces strong acoustic oscillations or excessive suppression of structure, which observations disfavor. In the solid model the shear response adds new, independent propagation speeds for vector and longitudinal disturbances. This extra freedom prevents the large, problematic oscillations that appear in the perfect‑fluid case. The authors derive stability conditions for tensor, vector and scalar perturbations to identify viable parameter ranges.

The solid phase also gives several observational signatures that differ from perfect‑fluid unification. Structure growth can be suppressed at late times. The model predicts a nontrivial gravitational slip (a difference in the two potentials that describe how space and time bend) and an effective, time‑dependent mass for gravitational waves. Tensor modes still travel at the speed of light, but their mass evolves because of the medium’s shear rigidity. Because the shear-related effects only become important after the transition, the model leaves high‑redshift cosmology essentially unchanged while producing testable features at low redshift. The paper discusses the longitudinal transfer function, the gravitational slip, and changes to the inferred gravitational‑wave distance as possible observable targets.