Elongation Control in an Algorithmic Chemistry

Algorithmic chemistries intended as computation models seldom model energy. This could partly explain some undesirable phenomena such as unlimited elongation of strings in these chemistries, in contrast to nature where polymerization tends to be unfavored. In this paper, we show that a simple yet sufficiently accurate energy model can efficiently steer resource usage, in particular for the case of elongation control. A string chemistry is constructed on purpose to make strings grow arbitrarily large. Simulation results show that the addition of energy control alone is able to keep the molecules within reasonable length bounds, even without mass conservation, and without explicit length thresholds. A narrow energy range is detected where the system neither stays inert nor grows unbounded. At this operating point, interesting phenomena often emerge, such as clusters of autocatalytic molecules, which seem to cooperate.