Chemical Morphology Controls Reactivity of OH Radicals at the Air-Ice Interface.

At the air-ice interface, some aromatic compounds such as benzene and anthracene are surprisingly unreactive toward OH. This may be a consequence of the poor solvation of these compounds at the interface, resulting in clustering there. We test this hypothesis by comparing the reaction of OH with pyrene, a 4-ring polyaromatic hydrocarbon (PAH), to reactions of OH with the more water-soluble compounds coumarin and 7-hydroxycoumarin (7OHC). We observe that OH reacts readily with coumarin and 7OHC at both liquid and frozen air-water interfaces. Pyrene, a much less soluble compound, reacts with OH at the liquid surface but not at the air-ice interface. We report evidence of pyrene aggregation at the ice surface based on its broadened and red-shifted emission spectrum alongside fluorescence mapping of anthracene, a closely related 3-ring PAH, which shows bunching at the ice surface. By contrast, fluorescence mapping shows that coumarin is fairly homogeneously distributed at the air-ice interface. Together, these results suggest that the limited reactivity of some compounds toward OH at the ice surface may be a consequence of their propensity to self-aggregate, demonstrating that chemical morphology can play an important role in reactions at the ice surface.

Improved Approximation Algorithms for Bounded-Degree Local Hamiltonians.

The low-temperature properties of interacting quantum systems are believed to require exponential resources to compute in the general case. Quantifying the extent to which such properties can be approximated using efficient algorithms remains a significant open challenge. Here, we consider the task of approximating the ground state energy of two-local quantum Hamiltonians with bounded-degree interaction graphs. Most existing algorithms optimize the energy over the set of product states. We propose and analyze a family of shallow quantum circuits that can be used to improve the approximation ratio achieved by a given product state. The algorithm takes as input an n-qubit product state with variance Var and improves its energy by an amount proportional to Var^{2}/n. In a typical case, this results in an extensive improvement in the estimated energy. We extend our results to k-local Hamiltonians and entangled initial states.