a very interesting paper.
in jacs asap.
Solving maze problems is not only relevant to the everyday issues
of urban transportation1 and to experimental psychology2 but is
also one of the model problems of network and graph theory3 as
well as robotics.4 With the advent of computers, algorithms for
maze solving have become automated, but the solution times still
scale unfavorably with maze size/complexity.5 Several groups have
thus explored the possibility of maze solving by physical, chemical,
or even biological systems: microfluidic networks,6 chemical waves7
or plasmas,8 or microorganisms growing in response to food
gradients within the maze.9 Inspired by the latter example, we
wished to create a system in which an inanimate/chemical construct
would be self-propelled and solve mazes in response to chemical
stimuli. Here we describe one such system comprising small
droplets powered by the combination of acid/base chemistry and
surface tension effects. When subject to a pH gradient within a
maze, these droplets move toward regions of low pH and find the
shortest of multiple possible paths. Taxis in our system is over
distances of several centimeters and derives from the convection
flows developed outside of the droplets.1
