I use Blue Waters to simulate an ultracold inferno. Micheline Soley - PowerPoint PPT Presentation

I use Blue Waters to simulate an ultracold inferno. Micheline Soley Micheline Soley, Harvard University Micheline Soley, Harvard University I use Blue Waters to simulate an ultracold inferno. 1 / 15

K K K K Rb Rb Rb Rb 14 K 4000 K Goal: To introduce the concept of an ultracold inferno in order to make investigation of ultracold reactions possible. Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 2 / 15

Standard Approach Typically, quantum mechanics is used for precise simulation of chemical reactions. Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 3 / 15

Standard Approach However, time-dependent quantum mechanics is computationally intensive and has only been used for a few high-temperature reactions H 2 + OH H + H 2 O Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 4 / 15

Key Challenges Ultracold reactions cannot be studied with existing quantum techniques because: Vectors scale as T − 1 / 2 with temperature T Arrays scale as N D with system size D Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 5 / 15

In contrast, classical mechanics is free of exponential scaling ... but still limited to high temperatures. Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 6 / 15

Solution To make simulation of the ultracold KRb dimer reaction possible, we recognize that it acts like an ultracold inferno. Image reproduced from MBS, E. J. Heller, Phys. Rev. A, vol. 98, no. 5, pp. 052702, Nov. 2018. Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 7 / 15

The ultracold inferno enables simulation of the the reaction through semiclassical treatment of the hot intermediate and quantum treatment of the cold products. Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 8 / 15

Monte Carlo integration is used to determine the number of ways that atoms can be arranged in the hot intermediate semiclassically. Blue Waters enables the use of quintillions of sampling points via data paralellism. Images reproduced with permission from nicoguaro and Se Roger via Wikimedia (Creative Commons). Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 9 / 15

Random matrix theory and R -matrix theory are used to analyze the cold products quantum mechanically. -1 -1 - - ( ) ( ) ( ) 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 - - ( ) ( ) * ( ) x 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Parallelization on Blue Waters with ScaLAPACK enables manipulation of large, non-sparse matrices. Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 10 / 15

Accomplishments Preliminary results suggest semiclassical mechanics is not only computationally e ffi cient, but also nearly exact for the hot intermediate. Type Molecules Semiclassical Quantum Analytic 1 . 224 · 10 5 1 . 225 · 10 5 1 . 224 · 10 5 K 2 Rigid Rotors 3 . 06 · 10 5 3 . 058 · 10 5 3 . 063 · 10 5 Rb 2 1 . 875 · 10 10 1 . 874 · 10 10 1 . 874 · 10 10 K 2 , Rb 2 Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 11 / 15

Why It Matters The combination of classical and quantum mechanics makes possible computational simulation of systems previously considered o ff -limits. Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 12 / 15

Broader Impact Simulation of the ultracold KRb dimer reaction on Blue Waters has the power to impact a wide range of fields. Theoretical ultracold chemistry Experimental ultracold chemistry Computational chemistry Physics, chemistry, and computer science communities Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 13 / 15

Blue Waters Contributions Education on supercomputing techniques Computational support and resources not otherwise available Discussion with researchers from other fields Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 14 / 15

Acknowledgments Blue Waters Graduate Professor Eric J. Heller and Research Fellowship Professor Kang-Kuen Ni National Science Foundation Blue Waters Project Sta ff Graduate Research Fellowship under Grant No. DGE1144152 And you, for your attention! Micheline Soley, Harvard University Theoretical Chemistry at a New Frontier 15 / 15