Xinlu Deng

Controllable adhesive interfaces offer unique opportunities for dynamic object capture, especially for aerospace engineering. However, existing approaches often lack the capability to efficiently dissipate impact energy and rapidly protect the adhesive interface during rebound and typically rely on complex mechanical and control systems to fulfill multiple functions. Here, we present a sequential phase-switching adhesion strategy using shape memory polymers (SMPs) to facilitate full-cycle capture operations. The engineered SMP-frame adhesive structure captures targets with normalized kinetic energies exceeding 3800 joules per square meter, empowered by the intrinsic viscoelasticity of the material to provide both strong energy dissipation and rapid adhesive interface strengthening. The dynamic strengthening mechanism delays crack propagation and greatly enhances detachment toughness, enabling unprecedented rebound suppression within the short timescales of collision. The adhesive-integrated system successfully performs dynamic capture, manipulation, and release under free-floating conditions, demonstrating the effectiveness of the sequential phase-switching adhesion strategy. The proposed concept and design offer a paradigm for robust, multifunctional capture in high-dynamics environments.

“Living” organisms in nature exhibit robust and biologically intelligent surface anti-wrinkling. Nonetheless, the complexities of self-regulating stress or structural characteristics through growth or gene expression render the anti-wrinkling of “nonliving” artificial surfaces using bionic principles a pressing yet formidable challenge. Here, inspired by nonliving dehydrated leaves, we propose an on-demand customizable, material invariant, parametric surface anti-wrinkling strategy using leaf vein–imitated boundary curvature–coupled constraints. This strategy allows for an exact surface customization with enhanced anti-wrinkling capability, tailored to specific anti-wrinkling demands while maintaining the original cross-section materials. The defined parameters, anti-wrinkling width and concave radius, are customized by the anti-wrinkling design criteria via the dimensionless dual-correction stiffness model, which are simple linear or quadratic functions of anti-wrinkling demands and cross-section properties. Experiments at different scales and materials validate the correctness of the design criteria. The strategy in this study is effective on diverse wrinkle-prone surfaces at multiple scales and can inform real engineering design of the nonliving artificial surfaces.

Extrinsic fiber acoustic sensors have gained marked attention for their potential in remarkable sensitivity and minimum detectable pressure (MDP), together with advantages of immunity to polarization fading. Their extreme sensitivity, however, leads to excessive drifts that bring substantial 1/ f noise and complicate the demodulation process. Here, we present a stabilized triple-phase demodulation architecture that effectively suppresses 1/ f noise and ensures a nondrifting phase output with a low, flat noise floor of 173 μrad/ Hz without any prior calibration. We also propose a labyrinth-inspired multiresonant membrane for wide-bandwidth sound capture. As a result, an outstanding linear dynamic range of 115.72 dB and wideband acoustic spectrum coverage from 20 Hz to 15 kHz are achieved. Our sensor also displays a high sensitivity of 1.02 rad/Pa (−119.83 dB ref: 1 rad/μPa) and an MDP of 81.84 μPa/ Hz at its first resonance. We demonstrate exceptional sound detection capabilities across a large dynamic range of pressures, such as ecosystem perception, speech recognition, and shock detection. These findings highlight our sensor’s potential in diverse voice interaction and hyper-alert acoustic sensing.