As we reach the end of Moore’s law, the search is on for alternatives to classical semiconductor computers. These new devices are being designed to calculate hard problems faster, store information more densely, or require less energy and cost to operate. With a large space of molecules and reactions to choose from, computing with chemistry offers the possibility for massive parallelization and high information density, while remaining accessible and cost-effective.
The Aspuru-Guzik group is collaborating with experimentalists to design and implement molecular computers to solve combinatorial optimization problems, such as the traveling salesperson, the wedding table seating arrangement, and problems in logistics optimization.
In our approach, a computation is performed by a network of chemical reactions within an array of compartments (e.g. microdroplets), each representing a bit of information. For a given problem, we first program the objective and constraints into the array. After the system evolves to a steady state, the solution is obtained from the final states of each droplet. In effect, we are annealing the effective spin system to its ground state.
Our research aims to answer chiefly: How can we harness chemical reactions to perform computation? What computations can it perform? Which types of reactions are amenable to our approach? Can the device be scaled up to useful sizes, such as a million bits?
The field of excitonics is related to the understanding, control, and harnessing of electronic excitations in nanoscale environments. One of the goals of our group is the development of novel theoretical approaches for studying excitonic energy transfer in systems such as photosynthetic complexes, organic photovoltaic materials, J-aggregates and nano-materials such as quantum dot assemblies. We are carrying out fundamental work in the development of top-down and bottom-up approaches to the theoretical description of excitonic transfer processes in natural and artificial systems. From these studies, we would like to define basic physical principles underlying the efficient energy transfer which could, in turn, be employed to design new devices.