2025MNRAS.538.1690V

In the rapidly evolving field of exoplanetary science with missions like JWST and Ariel, there is a pressing need for fast and accurate simulations of disequilibrium chemistry in exoplanet atmospheres. Methods which assume chemical equilibrium simplify calculations but fail to capture the more complex chemical dynamics observed in actual exoplanets. Accurate estimations require complex kinetic codes that are time-intensive due to the need to solve ordinary differential equations. Given an extensive parameter space that needs to be explored to calculate forward models, kinetic codes will become a significant bottleneck, especially when used to match observations or in retrieval processes. We introduce CHEXANET, a novel U-Net-based neural network architecture designed to efficiently simulate disequilibrium chemistry in exoplanetary atmospheres. By utilizing the thermochemical equilibrium state of a hot-Jupiter's atmosphere as an initial condition, which computes in seconds, alongside a set of initial parameters such as the C/O ratio, the network effectively learns to predict the atmosphere in disequilibrium. It significantly enhances computational efficiency, reducing the prediction time for atmospheric disequilibrium states to just one second per atmosphere on a standard personal computer - over a hundred times faster than traditional kinetic models like Full and Reduced Exoplanet Chemical Kinetics distiLLed (FRECKLL).