Design of intelligent autocatalytic reaction networks

This thesis investigates the design of autocatalytic chemical reaction networks (CRNs) for out-of-equilibrium conditions in order to create intelligent chemical systems capable of decision-making, memory, and adaptation. Autocatalysis provides nonlinear feedback dynamics that mimic regulatory processes found in biological systems, allowing chemical systems to exhibit functions such as logic processing, signal response, and environmental adaptation. By carefully tuning flow conditions, catalytic composition, and the reactor properties, it becomes possible to program CRNs to store memory, perform logic operations, and propagate spatial information. Such systems lay the foundation for a new class of programmable, self-regulating materials with potential applications in neuromorphic computation, sensing, soft robotics, and synthetic biology. As this field progresses, intelligent chemical systems may increasingly perform intricate tasks like real-time stimulus processing, history-dependent decision-making, and spatial navigation. These abilities suggest broad applicability: in environmental monitoring, chemical systems could autonomously detect and respond to pollutants; in healthcare, they may underpin wearable devices that react to physiological changes; in soft robotics, they offer embedded sensing and actuation in complex terrains. The integration of chemical intelligence with digital platforms could further expand their real-world utility, enabling hybrid systems that combine chemical adaptability with computational precision. As design principles mature, it may become possible to build chemical circuits that replicate essential electronic components — like memory switches or oscillators — but operate through fluidic or molecular interactions instead of electrons. Ultimately, the promise of autocatalytic CRNs lies not only in emulating biology but in surpassing current limitations of semiconductor technology by creating biodegradable, low-energy, decentralized systems. This thesis brings us closer to realizing a new technological paradigm, where chemistry, computation, and biology merge into sustainable, intelligent systems for the future.