Showing papers in "Annual Review of Physical Chemistry in 1994"

Propagation Methods for Quantum Molecular Dynamics

Abstract: INTROIDUCTION Our current understanding of molecular dynamics uses quantum mechanics as the basic underlying theory to elucidate thc processes involved. Establishing numerical schemes to solve the quantum equations of motion is crucial for understanding realistic molecular encounters. The introduction of pseudo-spectral methods has been an important step in this direction. These methods allow an extremely accurate representation of the action of an operator, usually the Hamiltonian, on a wavefunction: q~ = fI~p. A solution for the quantum molecular dynamics can be obtained by recursively applying the elementary mapping step. This recursive application of the elementary step, termed the propagator, is the subject of this review. The na~:ural application of a propagator is in a time-dependent description of quantum molecular dynamics, where the propagator U(z) maps the waw~function at time t, ~p(t) to the wavefunction at time t + ~: 0(t+ ~) = ~(z)0(t). The decomposition into a recursive application of the elementary step is performed by a polynomial expansion of the propagator. The introduction of the Chebychev polynomial expansion (l) first created a propagation scheme that could match the accuracy of the

Intramolecular dynamics from eigenstate-resolved infrared spectra

Abstract: Understanding intramolecular energy flow in molecules is one of the central goals of chemical physics. Statistical theories [such as Rice-Ram­ sperger-Ka:;sel-Marcus (RRKM) theory] of unimolecular reaction rates, which assume that the vibrational degrees of freedom rapidly exchange energy, have proved extremely successful ( 1). Recent work on small mol­ ecules has provided rigorous tests of these theories (2). With the success of unimolecular reaction theory, it has become clear that intramolecular vibrational energy redistribution (IVR) is a nearly universal phenomenon when molecules have enough energy to break bonds. However, despite this success, several impottant issues about the nature 'Of the assumed statistical motion have been left unresolved. Definifive measurements of the lime scale of the IVR process have only recently become possible. Unimolecular reaction theories only require IVR to be faster than the dissociation rate. For most chemical reactions, these

Orientation and Alignment of Reaction Products

Abstract: are intrinsically anisotropic because certain angles of approach of A to BC and of separation of AB from C are preferred, as are certain planes of rotation of the reagent and product molecules. Most of the experimental and theoretical efforts aimed at understanding elementary reactions have been applied to the study of the scalar properties of Reaction 1, such as how the rate of reaction varies with the energies of the reagents or how the energy available after reaction is partitioned among the internal and translational degrees of freedom of the products. To concentrate exclusively on the scalar properties of the reaction is to neglect, however, the vector properties that are key indicators of the anisotropic forces present in the reaction. Vector properties, such as velocities and angular momenta, possess not only magnitudes that can be directly related to translational and rotational energies, but also well-defined directions. Only by understanding the scalar and vector properties together, as well as possible

Kinetics of Surface Growth: Phenomenology, Scaling, and Mechanisms of Smoothening and Roughening

Abstract: The growth of solid surfaces via vapor phase processes can be viewed as proceeding in either of two directions: outward by physical or chemical deposition or inward by physical or chemical etching. Figure I is a sche­ matic illustration of the evolution of a surface, S(r, t), where S is the z­ coordinate of the growing surface at the position r = (x,y) and time t, and the initial condition S(r) = 0 at t = 0 corresponds to initiating the growth on a perfectly fiat surface. The average value of the new surface height at any particular time

Generic Long-Range Correlations in Molecular Fluids

Abstract: In this ·review wc are concerned with the nature of the local microscopic fluctuations in fluid systems that are in a well-defined macroscopic state; this may be either a thermodynamic equilibrium state or a nonequilibrium steady-state, i.e. a nonequilibrium state in which the parameters that describe the state do not depend on time. To specify the spatial and tem­ poral character of these fluctuations, one introduces various correlation functions (1 -3). For this purpose, we consider dynamical variables, A(r; r) and B(r; r), that depend explicitly on the position r and im­ plicitly on any or all of the particle coordinates. We denote by f the set of coordinates and momenta of all the particles at some initial time, and by ret) the values of these same variables as a result of the dynamical motion of the collection of particles at a later time t. The correlation function of A[r); ret)] and B(r2; f) is then defined as

Dynamics by Semiclassical Methods

Abstract: quantum approa ch. The evaluation of the quantum time evolution of a system requires the solution of the time-dependent Schr6dinger e:quation. A full numerical solution of the quantum problem is generally not feasible for cases involving more than a few degrees of freedom. The: situation is similar for the time-independent calculation of quantum transition probabilities in collision problems. Semiclassical methods build up approximate quantum solutions, which are numerically relatively easy to obtain, even for moderately long times, from information obtained along classical trajectories. Another reason for the attractiveness of semiclassical methods is their intuitive appeal. A semiclassical analysis of a problem allows the results to be interpreted in terms of classical trajectories, and this can provide a clearer picturc of the bchavior of the system than might be possible from quantum calculatio ns. Classical mechanics provides an accurate approximation of the dynamics of macroscopic systems, while quantum effects are very important on the microscopic lev el. Electronic states of atoms and molecules are known to be highly quantized, while the motion of molecules in a liquid are often well described by classical mechani cs. The scattering of molecules at rela

Ab initio studies of hydrogen bonds: the water dimer paradigm.

Abstract: The textbook definition of a hydrogen bond consists of the interaction between the covalent X-H bond of one molecule and the lone electron pair of the Y atom of another, presuming both X and Y are electronegative atoms like 0, N, or F. This ostensibly simple interaction has turned out to be one of the most intriguing in all of chemistry, as well as one of the most important. The first realistic notions about the H-bond date back to the work of Huggins, Latimer, and Rodebush in the early part of this century (1 -4). Interest in these bonds increased when studies reported the bonds' intimate involvement in the structure and function of biological macromolecules (5, 6). In the early 1 950s, these bonds were accepted as primary elements in protein structure (7) and in transmission of the genetic code (8). Most of the early research on these bonds centered around the mani­ festations of hydrogen bonding in condensed phases. Infrared and Raman spectroscopy provided windows for viewing changes in the bonding pat­ terns of the molecules involved in the H-bond (3, 4). Nuclear magnetic resonance spectra produced dynamic data (9-1 3) dealing with the for­ mation and dissociation ofH-bonds and with proton cxchanges on various time scales that provided information about equilibria and rates. Radial distribution patterns revealed the manner in which the molecules organized themselves to accommodate the formation of H-bonds, and diffraction

High-Resolution, Direct Infrared Laser Absorption Spectroscopy in Slit Supersonic Jets: Intermolecular Forces and Unimolecular Vibrational Dynamics in Clusters

Abstract: A long-standing goal in the field of cluster research has been to elucidate the transition between the properties of isolated monomer species in the gas phase and the corresponding properties in the condensed phase. Some of the most exacting methods of studying small clusters in this transition region are offered by high resolution spectroscopies (1-6), particularly in the nearand far-IR, which have succeeded in providing the first detailed glimpses of the quantum level structure that corresponds to both low frequency van der Waals motion and high frequency intramolecular motion in the molecular aggregates. In a very tangible sense, we are beginning to realize one of the long-standing dreams of the physical chem­ ist-that of watching molecules in the process of a collision by measuring

Surface Reconstruction and Catalysis

Abstract: The location and bonding of atoms and molecules on surfaces is of great interest to surface chemists and to those interested in the application of surfaces through exploitation of their unique properties. These properties include chemical properties that give rise to selective adsorption and heterogeneous catalysis; mechanical properties that control adhesion, fric­ tion, slide, or fracture; electrical properties utilized in microelectronic circuitry and xerography; magnetic properties used in information storage on tape or disk drives; and optical properties that give rise to nonlinear effects such as second harmonic and sum frequency generation. During the past 25 years, over 50 new techniques have been developed that permit the investigation of surfaces on the molecular level (l). The ability to study surfaces with increased time and spatial resolution (and energy resolution, when applicable) controls the development of many surface technologies. The contributions of science push and surface technology pull have resulted in an exponential growth in the field of surface science and pro­ pelled it among the frontier fields of physical chemistry. Several new surface science techniques permit quantitative deter-

Excess Electrons in Liquids: Geometrical Perspectives

Abstract: The behavior of excess electrons in liquids is remarkably varied. To under­ stand it, one must draw on a range of topics, all of which are pertinent to the nature of electronic states in disordered materials in general. This review describes the current theoretical understanding. Our focus is on single-electron properties in bulk fluids. Related research on electrons in clusters (e.g. 1 , l a) is not discussed in this review, though we do describe some implications and some recent work on correlated many electron systems. Recent n:views of experiments on electron mobility in nonpolar liquids were presented in 1989 by Holroyd & Schmidt (2), and in 1991 by Schmidt (3) and Munoz (4). Work on the behavior of electrons in polar fluids is routinely reviewed in the Colloque Weyl series; the seventh volume appeared in 1 991 (5). Early and influential theoretical work was reviewed several years ago by Davis & Brown (6). The general perspectives held at that time are also captured in the collection of theoretical and experimental articles published in a 1977 issue of the Canadian Journal of Chemistry (7). Hernandez reviewed these ideas and some new results in 1991 (8). A rekindling of interest in this topic occurred in the mid 1980s, spurred in part by new-found computational abilities. These improved calculations and a new theoretical perspective caused revision and clarification of the earlier concepts. Simulation work has been reviewed by Sprik & Klein (9), by Rossky & Schnitker ( 1 0) and by Coker & Berne ( 11). The underlying