Artificially Intelligent Formaldehyde Sensing using Printed Graphene-Based Aerogels

The identification and analysis of formaldehyde as a hazardous air pollutant in indoor environments present an immediate and critical concern. The endeavour to achieve real-time recognition of formaldehyde in the presence of interfering gases represents a formidable challenge, particularly for room temperature sensors that are susceptible to noise and baseline drift. Exploring the unique properties of dimensionally hybrid systems based on graphene and leveraging artificial intelligence (AI) offer promising pathways.<br/>In this study, we tackle these challenges by utilising the 3D-printed porous structure of graphene-based aerogels as interconnected frameworks for rapid gas diffusion and sensitising the aerogels with tin oxide quantum dots (QDs) for high-performance chemical sensing at room temperature. Through reaction-coupled diffusion modelling of both filament-structured and film-like aerogels, we explain and experimentally demonstrate the superior gas-sensing capabilities of thinner filaments with higher surface porosity. With optimised morphology and doping of the printed structures, we achieve an outstanding response of 15.23% towards 1 ppm formaldehyde at room temperature.<br/>Furthermore, we explore the versatility of different structures and chemical doping to construct gas sensor arrays capable of distinguishing formaldehyde from ammonia and nitrogen dioxide. To enable real-time gas species classification, we develop gas recognition algorithms based on dynamic response features that are independent of steady-state response or testing sequences. These algorithms demonstrate remarkable accuracies, even in the presence of significant noise and baseline drift. Our findings present a novel artificially intelligent solution for the detection and real-time identification of formaldehyde at room temperature, offering promising prospects of graphene-based aerogels for addressing this pressing environmental issue.