Joel Mark Bowman is an American physical chemist and educator. He is the Emeritus Samuel Candler Dobbs Professor of Theoretical Chemistry at Emory University.[1]
Bowman's research interests are in basic theories of chemical reactivity.[1] He is well known for his contributions in simulating potential energy surfaces for polyatomic molecules and clusters. Approximately fifty potential energy surfaces for molecules and clusters have been simulated employing his permutationally invariant polynomial method.[3]
Permutationally invariant polynomial (PIP) method
Simulating potential energy surfaces (PESs) for reactive and non-reactive systems is of broad utility in theoretical and computational chemistry. Development of global PESs, or surfaces spanning a broad range of nuclear coordinates, is particularly necessary for certain applications, including molecular dynamics and Monte Carlo simulations and quantum reactive scattering calculations.
Rather than utilizing all of the internuclear distances, theoretical chemists often analytical equations for PESs by using a set of internal coordinates. For systems containing more than four atoms, the count of internuclear distances deviates from the equation 3N−6 (which represents the degrees of freedom in a three-dimensional space for a nonlinear molecule with N atoms).[4][5] As an example, Collins and his team developed a method employing different sets of 3N−6 internal coordinates, which they applied to analyze the H+ CH4 reaction. They addressed permutational symmetry by replicating data for permutations of the H atoms.[6] In contrast to this approach, the PIP method uses the linear least-square method to accurately match tens of thousands of electronic energies for both reactive and non-reactive systems mathematically.
Methodology
Generally, the functions used in fitting potential energy surfaces to experimental and/or electronic structure theory data are based on the choice of coordinates. Most of the chosen coordinates are bond stretches, valence and dihedral angles, or other curvilinear coordinates such as the Jacobi coordinates or polyspherical coordinates.[citation needed] There are advantages to each of these choices.[3] In the PIP approach, the N(N − 1)/2 internuclear distances are utilized. This number of variables is equal to 3N −6 (or 3N − 5 = 1 for diatomic molecules) for N = 3, 4 and differs for N ≥ 5. Thus, N = 5 is an important boundary that affects the choice of coordinates. An advantage of employing this variable set is its inherent closure under all permutations of atoms. This implies that regardless of the order in which atoms are permuted, the resulting set of variables remains unchanged. However, the main focus pertains to permutations involving identical atoms, as the PES must be invariant under such transformations.[3]
PIP utilizing Morse variables of the form , where is the distance between atoms and and is a range parameter) offers a method for mathematically characterizing high-dimensional PESs. By fixing the range parameter in the Morse variable, the PES can be determined through linear least-squares fitting of computed electronic energies for the system at various structural arrangements. The adoption of a permutationally invariant fitting basis, whether in the form of all internuclear distances or transformed variables like Morse variables, facilitates the attainment of accurate fits for molecules and clusters.[7]
Townsend, D.; Lahankar, S. A.; Lee, S. K.; Chambreau, S. D.; Suits, A. G.; Zhang, X.; Rheinecker, J.; Harding, L. B.; Bowman, J. M. (2004), "The roaming atom: Straying from the reaction path in formaldehyde decomposition", Science, 306 (5699): 1158–61, Bibcode:2004Sci...306.1158T, doi:10.1126/science.1104386, PMID15498970, S2CID31464376.
This article needs additional or more specific categories. Please help out by adding categories to it so that it can be listed with similar articles.(July 2023)