Biomimetic Tri-Phase Hierarchical Interface Enables Ultra-Efficient Enrichment and Sensing of Respiratory Viruses in Exhaled Breath.

The analysis of exhaled breath condensate (EBC) for respiratory pathogen detection is fundamentally limited by the extreme dilution of target analytes, leading to insufficient sensitivity in point-of-care platforms. Here, we present a paradigm-shifting solution inspired by the efficient multi-stage capture mechanism of the human respiratory system. We engineered a tri-phasic hierarchical interface that orchestrates a synergistic cascade for viral enrichment and detection: (1) microscale pillar arrays that mimic bronchial bifurcations to direct airflow and enhance particle-surface collisions via inertial impaction; (2) a TiO2 nanowire forest that emulates the mucin network, providing a high-surface-area scaffold for analyte retention; and (3) a functionalized polymer coating for specific antibody-antigen recognition. This bio-inspired architecture, optimized through computational fluid dynamics and fabricated via wafer-level MEMS processes, achieved a 3.6-fold increase in condensate collection and a 500% enhancement in capture-site density. When integrated into a portable fluorescence detection system, the platform demonstrated high specificity (>95%) and detected SARS-CoV-2 pseudovirus down to 3 TU/mL under the tested simulated-breath workflow. Critically, the binding kinetics, analyzed using the Hill model, further supports a saturation-like, heterogeneous capture behavior of the hierarchical interface. This work not only delivers a powerful tool for non-invasive diagnostics but also establishes a generalizable biomimetic design paradigm for next-generation biosensors targeting ultra-dilute biomarkers.