Colloidal particles endowed with specific time-dependent interactions are a promising route for realizing artificial materials that have the properties of living ones. Previous work has demonstrated how this system can give rise to self-replication. Here, we introduce the process of colloidal catalysis, in which clusters of particles catalyze the creation of other clusters through templating reactions. Surprisingly, we find that simple templating rules generically lead to the production of huge numbers of clusters. The templating reactions among this sea of clusters give rise to an exponentially growing catalytic cycle, a specific realization of Dyson's notion of an exponentially growing metabolism. We demonstrate this behavior with a fixed set of interactions between particles chosen to allow a catalysis of a specific sixparticle cluster from a specific seven-particle cluster, yet giving rise to the catalytic production of a sea of clusters of sizes between 2 and 11 particles. The fact that an exponentially growing cycle emerges naturally from such a simple scheme demonstrates that the emergence of exponentially growing metabolisms could be simpler than previously imagined.catalytic cycles | DNA-coated colloids | metabolism | templating T he origin of life is usually associated with self-replication, the ability of an entity to create copies of itself (1). Thus inspired, there have been many efforts in recent years aimed at creating artificial systems with self-replicative capacity. These are typically inspired by the linear chain mechanism of DNA. However, living systems are more than individual entities that can replicate themselves. Dyson (2) and Oparin (3) argued that a more critical aspect of living systems is the creation of a metabolism, a complex cascade of chemical reactions that together are able to accomplish more than any single chemical reaction can do on its own. Using a simple mathematical model, the so-called "garbage bag model," Dyson presented a scenario through which such a complex metabolism could arise spontaneously. His idea is that a random set of catalysts will catalyze arbitrary chemical reactions in a nonsynergistic fashion. However, if each catalyst is more likely to function when there are others that are synergistic with it, then there is a critical amount of cooperativity above which metabolic cycles spontaneously emerge. This idea has been explored through abstract simulations (4). In a similar spirit, Kauffman and coworkers have shown that if the probability of one species catalyzing the formation of another is above a threshold, then catalytic cycles naturally emerge (5-7).Whether this scenario can naturally occur in practice remains obscure. The fundamental question is to determine how likely it is for a soup of interacting catalysts to self-organize into an exponentially growing catalytic cycle. How special do the interactions between the different components need to be for spontaneous exponentially growing catalytic cycles to emerge? In this work, we present an explicit demon...