Search

We propose a methodology to unify electronic and nuclear quantum wavepacket dynamics in molecular processes including nonadiabatic chemical reactions. The canonical and traditional approach in the full quantum treatment both for electrons and nuclei rests on the Born–Oppenheimer fixed nuclei strategy, the total wavefunction of which is described in terms of the Born–Huang expansion. This approach is already realized numerically but only for small molecules with several number of coupled electronic states for extremely hard technical reasons. Besides, the stationary-state view of the relevant electronic states based on the Born–Oppenheimer approximation is not always realistic in tracking real-time electron dynamics in attosecond scale. We therefore incorporate nuclear wavepacket dynamics into the scheme of nonadiabatic electron wavepacket theory, which we have been studying for a long time. In this scheme thus far, electron wavepackets are quantum mechanically propagated in time along nuclear paths that can naturally bifurcate due to nonadiabatic interactions. The nuclear paths are in turn generated simultaneously by the so-called matrix force given by the electronic states involved, the off-diagonal elements of which represent the force arising from nonadiabatic interactions. Here we advance so that the nuclear wavepackets are directly taken into account in place of path (trajectory) approximation. The nuclear wavefunctions are represented in terms of the Cartesian Gaussians multiplied by plane waves, which allows for feasible calculations of atomic and molecular integrals together with the electronic counterparts in a unified manner. The Schrödinger dynamics of the simultaneous electronic and nuclear wavepackets are to be integrated by means of the dual least action principle of quantum mechanics [K. Takatsuka, J. Phys. Commun. 4, 035007 (2020)], which is a time-dependent variational principle. Great contributions of Vincent McKoy in the electron dynamics in the fixed nuclei approximation and development in time-resolved photoelectron spectroscopy are briefly outlined as a guide to the present work.

The path-branching theory as a nonadiabatic electron wavepacket theory (Yonehara et al., 2012), in which nonadiabatic electron wavepackets are propagated in time along branching nuclear paths, is extended so that Gaussian nuclear wavepackets are to be evolved in time along the variational quantum paths, which are determined consistently with the electron dynamics.

We propose a general theoretical scheme of relativistic electron-nucleus coupled dynamics of molecules in radiation fields, which is derived from quantum electrodynamical formalism. Aiming at applications to field-induced dynamics in ultrastrong laser pulses to the magnitude of 1016 W/cm2 or even larger, we derive a nonperturbative formulation of relativistic dynamics using the Tamm-Dancoff expansion scheme, which results in, within the lowest order expansion, a time-dependent Schrodinger equation with the Coulombic and retarded transversal photon-exchange interactions. We also discuss a wavepacket type nuclear dynamics adapted for such dynamics.

Time-dependent variational principle (TDVP) provides powerful methods in solving the time-dependent Schröinger equation. As such Kan developed a TDVP (Kan 1981 Phys. Rev. A 24, 2831) and found that there is no Legendre transformation in quantum variational principle, suggesting that there is no place for the Maupertuis reduced action to appear in quantum dynamics. This claim is puzzling for the study of quantum–classical correspondence, since the Maupertuis least action principle practically sets the very basic foundation of classical mechanics. Zambrini showed within the theory of stochastic calculus of variations that the Maupertuis least action principle can lead to the Nelson stochastic quantization theory (Zambrini 1984 J. Math. Phys. 25, 1314). We here revisit the basic aspect of TDVP and reveal the hidden roles of Maupertuis-Hamilton least action in the Schrödinger wavepacket dynamics. On this basis we propose a dual least (stationary) action principle, which is composed of two variational functionals; one responsible for ‘energy related dynamics’ and the other for ‘dynamics of wave-flow’. The former is mainly a manifestation of particle nature in wave-particle duality, while the latter represents that of matter wave. It is also shown that by representing the TDVP in terms of these inseparably linked variational functionals the problem of singularity, which is inherent to the standard TDVPs, is resolved. The structure and properties of this TDVP are also discussed.

The Longuet-Higgins (Berry) phase arising from nonadiabatic dynamics and the Aharonov-Bohm phase associated with the dynamics of a charged particle in the electromagnetic vector potential are well known to be individually a manifestation of a class of the so-called geometrical phase. We herein discuss another similarity between the force working on a charged particle moving in a magnetic field, the Lorentz force, and a force working on nuclei while passing across a region where they have a strong quantum mechanical kinematic (nonadiabatic) coupling with electrons in a molecule. This kinematic force is indeed akin to the Lorentz force in that its magnitude is proportional to the velocity of the relevant nuclei and works in the direction perpendicular to its translational motion. Therefore this Lorentz-like nonadiabatic force is realized only in space of more or equal to three dimensions, thereby highlighting a truly multi-dimensional effect of nonadiabatic interaction. We investigate its physical significance qualitatively in the context of breaking of molecular spatial symmetry, which is not seen otherwise without this force. This particular symmetry breaking is demonstrated in application to a coplanar collision between a planar molecule and an atom sharing the same plane. We show that the atom is guided by this force to the direction out from the plane, resulting in a configuration that distinguishes one side of the mirror plane from the other. This can serve as a trigger for the dynamics towards molecular chirality.

The nonadiabatic nuclear wavepacket dynamics on the coupled two lowest (1)Σ(+) states of the LiF molecule under the action of a control pulse is investigated. The control is achieved by a modulation of the characteristics of the potential energy curves using an infrared field with a cycle duration comparable to the time scale of nuclear dynamics. The transition of population between the states is interpreted on the basis of the coupled nuclear wavepacket dynamics on the effective potential curves, which are transformed from the adiabatic potential curves with use of a diabatic representation that diagonalizes the dipole-moment matrix of the relevant electronic states. The basic feature of the transition dynamics is characterized in terms of the notion of the collision between the dynamical crossing point and nuclear wavepackets running on such modulated potential curves, and the transition amplitude is mainly dominated by the off-diagonal matrix element of the time-independent electronic Hamiltonian in the present diabatic representation. The importance of the geometry dependence of the intrinsic dipole moments as well as of the diabatic coupling potential is illustrated both theoretically and numerically.

A practical quantum theory for unifying electronic and nuclear dynamics, which were separated by the Born–Oppenheimer approximation, is proposed. The theory consists of two processes. Nonadiabatic (quantum) electron wavepacket dynamics on branching (non-Born–Oppenheimer) nuclear paths are first constructed. Since these paths are not the classical trajectories, most of the existing semiclassical theories to generate quantum wavepacket do not work. Therefore, we apply our own developed semiclassical wavepacket theory to these generated non-Born–Oppenheimer paths. This wavepacket is generated based on what we call the action decomposed function, which does not require the information of the so-called stability matrix. Thus, the motion of nuclei is also quantized, and consequently the total wavefunction is represented as a series of entanglement between the electronic and nuclear wavepackets. © 2009 Wiley Periodicals, Inc. Int J Quantum Chem, 2009

Classical trajectory study of nuclear motion on the Born-Oppenheimer potential energy surfaces is now one of the standard methods of chemical dynamics. In particular, this approach is inevitable in the studies of large molecular systems. However, as soon as more than a single potential energy surface is involved due to nonadiabatic coupling, such a naive application of classical mechanics loses its theoretical foundation. This is a classic and fundamental issue in the foundation of chemistry. To cope with this problem, we propose a generalization of classical mechanics that provides a path even in cases where multiple potential energy surfaces are involved in a single event and the Born-Oppenheimer approximation breaks down. This generalization is made by diagonalization of the matrix representation of nuclear forces in nonadiabatic dynamics, which is derived from a mixed quantum-classical representation of the electron-nucleus entangled Hamiltonian [Takatsuka, K. J. Chem. Phys. 2006, 124, 064111]. A manifestation of quantum fluctuation on a classical subsystem that directly contacts with a quantum subsystem is discussed. We also show that the Hamiltonian thus represented gives a theoretical foundation to examine the validity of the so-called semiclassical Ehrenfest theory (or mean-field theory) for electron quantum wavepacket dynamics, and indeed, it is pointed out that the electronic Hamiltonian to be used in this theory should be slightly modified.

We investigate how the nuclear dynamics at an avoided crossing is affected and can be controlled by the introduction of a laser field whose cycle is comparable to the time-scale of the nuclear dynamics. By introducing the concepts of light-induced effective potential energy curves and dynamical avoided crossing, we describe the laser controlled nuclear dynamics and present basic control scenarios, giving a detailed explanation of the underlying dynamical mechanisms. The scenarios presented allow for examples to understand from a different perspective the results of dynamic Stark control experiments. The proposed interpretation is applied to the laser-controlled nonadiabatic dynamics between the two lowest (1)Σ(+) states of LiF, where the usefulness of the concepts developed is elucidated.

In light of the fundamental importance and renewed interest of the tunnel phenomena, we review the recent development of semiclassical tunneling theory, particularly from the view point of “tunneling path”, beginning from a simple one-dimensional formula to semiclassical theories making use of the analytic continuation, in time, coordinates, or momentum, which are the stationary solutions of semiclassical approximations to the Feynman path integrals. We also pay special attention to the instanton path and introduce various conventional and/or intuitive ideas to generate tunneling paths, to which one-dimensional tunneling theory is applied. Then, we review the recent progress in generalized classical mechanics based on the Hamilton–Jacobi equation, in which both the ordinary Newtonian solutions and the instanton paths are regarded as just special cases. Those new complex-valued solutions are generated along real-valued paths in configuration space. Such non-Newtonian mechanics is introduced in terms of a quantity called “parity of motion”. As many-body effects in tunneling, illustrative numerical examples are presented mainly in the context of the Hamilton chaos and chemical reaction dynamics, showing how the multidimensional tunneling is affected by the system parameters such as mass combination and anisotropy of potential functions.

For Hamilton dynamics on a potential that has multiple local basins as in structural isomerization reaction of clusters, a notion of interbasin mixing is introduced that is responsible for Markov-type stochastic appearance of molecular structures. An extension of the Lyapunov exponent to quantify the time scale to reach interbasin mixing is proposed. The present dynamics also serves as a prototype of multichannel chemical reactions.

The dependence of lifetime distribution in isomerization dynamics of Ar7-like clusters on the potential topography is reported. Using the scaled Morse potential V=∑i

The effects of multidimensionality in the quantum mechanical tunneling of chemical reactions are investigated. The aim of the present report is twofold. In the first place, we construct a new semiclassical theory to describe the tunneling by incorporating nonclassical solutions of the time-dependent Hamilton–Jacobi equation into the Feynman kernel. A systematic class of complex-valued (nonclassical) solutions for the time-independent Hamilton–Jacobi equation has been found that are generated along non-Newtonian paths in real-valued configuration space [K. Takatsuka and H. Ushiyama, Phys. Rev. A 51, 4353 (1995)]. In the present paper, the straightforward extension is applied to the time-dependent Hamilton–Jacobi equation, the solutions of which describe the tunneling in chemical reactions. It is shown that no damping factor due to the tunneling arises from the preexponential factor in the thus obtained nonclassical kernel, since it is still real valued, aside from the complex phase due to the Maslov index,...

An explicit solution to the Maslov-type semiclassical theory for propagating a wave function, rather than evolving in time the Feynman kernel, is presented. It turns out that the present solution bears distinguished advantages over the semiclassical kernel, one of the most remarkable examples of which is the far less number of classical trajectories required for the propagation, basically proportional to P{sup 2N}{approximately}P{sup 3N} for the kernel while only to P{sup N} in our solution, where {ital N} is the dimension in configuration space and {ital P} is the number of sampling points in each dimension. As an illustrative example to show the validity of the solution, the theory is applied to the calculation of eigenvalues of the Morse oscillators, giving accurate results in a compact way. {copyright} {ital 1997} {ital The American Physical Society}

Through nonadiabatic interaction due to electron transfer as that in alkali halides, vibrational dynamics on the ionic potential energy surface (large dipole moment) is coupled to that on the covalent surface (small dipole moment). Thus, population transfer between the states should cause long-range electron jump between two remote sites, which thereby leads to a sudden change of the large molecular dipole moment. Therefore, by making repeated use of the dynamical Stark effect, one may expect emission of photons from it. We show with coupled quantum wavepacket dynamics calculation that such photoemission can indeed occur and can be controlled by an external field. The present photoemission can offer an alternative scheme to study femtosecond and subfemtosecond vibrational and electronic dynamics and may serve as a unique optical source.

A new method to calculate eigenfunctions and eigenvalues in a given energy range is proposed, which can therefore be applied to highly excited states of electronic and/or vibrational states of a molecule. The spectral components of a wave packet that lie outside the energy range are projected out through the time evolution; that is, the packet is screened onto the energy range. If the range includes only a single root, the corresponding eigenfunction is screened first, and the eigenvalue follows as its expectation value. For a case where there is more than a single root, several methods can be figured out. One typical and effective procedure is to construct local basis functions in terms of the aforementioned energy screened wave packets to represent the Hamiltonian in them and to diagonalize it. The concept to construct a local basis was originally developed by Neuhauser [J. Chem. Phys. 93, 2611 (1990)]. The present method performs it in a more efficient and theoretically satisfactory way.

When a nonadiabatic system that has an ionic state (large dipole moment) and a covalent state (small dipole moment) is located in a strong laser field, the crossing point of the two potential energy curves is forced to oscillate due to the oscillating laser field and to meet wavepackets moving on the potential curves many times. This leads to additional transitions between the two states, and under favorable conditions, the wavepacket may be confined in a spatial region rich in nonadiabatic interaction. In this paper, taking the LiF molecule system in a continuous-wave driving field as a prototypical example, the dynamical origins of the wavepacket confinement are theoretically investigated.

A complicated yet interesting induced photon emission can take place by a nonadiabatic intramolecular electron transfer system like LiF under an intense CW laser [Y. Arasaki, S. Scheit, and K. Takatsuka, J. Chem. Phys. 138, 161103 (2013)]. Behind this phenomena, the crossing point between two potential energy curves of covalent and ionic natures in diabatic representation is forced to oscillate, since only the ionic potential curve is shifted significantly up and down repeatedly (called the Dynamical Stark effect). The wavepacket pumped initially to the excited covalent potential curve frequently encounters such a dynamically moving crossing point and thereby undergoes very complicated dynamics including wavepacket bifurcation and deformation. Intramolecular electron transfer thus driven by the coupling between nonadiabatic state-mixing and laser fields induces irregular photon emission. Here in this report we discuss the complicated spectral features of this kind of photon emission induced by infrared laser. In the low frequency domain, the photon emission is much more involved than those of ultraviolet/visible driving fields, since many field-dressed states are created on the ionic potential, which have their own classical turning points and crossing points with the covalent counterpart. To analyze the physics behind the phenomena, we develop a perturbation theoretic approach to the Riccati equation that is transformed from coupled first-order linear differential equations with periodic coefficients, which are supposed to produce the so-called Floquet states. We give mathematical expressions for the Floquet energies, frequencies, and intensities of the photon emission spectra, and the cutoff energy of their harmonic generation. Agreement between these approximate quantities and those estimated with full quantum calculations is found to be excellent. Furthermore, the present analysis provides with notions to facilitate deeper understanding for the physical and mathematical mechanisms of the present photon emission.

A mechanism of quantizing chaos is discussed and an analytical quantization condition for weak chaos is derived. In the asymptotic evaluation of the density of states, we include the effect of the rapid change of the amplitude factor in the Feynman kernel, whereas the Gutzwiller trace formula considers only the violent oscillation of the usual quantum-phase represented by the action integral. Instead of the intractable application of the steepest descent method, we extrC\ct essential information from the periodic-orbit theory based on the smooth relationship between the steepest descent method and the stationary phase approximation. For weak chaos is set a quantization condition that detects which periodic orbits are supposed to correlate with the quantizing steepest-descent-solutions. It is shown that the true energy to be quantized is shifted from that of such periodic orbits. The energy thus quantized has the same form as the Helmholtz free energy within the framework of our thermodynamic characterization of quantum chaos. As the instability disappears, this quantization condition is correctly reduced to the resonant quantization condition, through which it is connected with the Einstein-Brillouin-Keller conditions.

The coherent path integral is generalized such that the identity operator represented in a complete (actually overcomplete) set of the coherent states with the ''time-variable'' exponents are inserted between two consecutive short-time propagators. Since such a complete set of any given exponent can constitute the identity operator, the exponent may be varied from time to time without loss of generality as long as it is set common to all the Gaussians. However, a finite truncation of the coherent state expansion should result in different values of the propagator depending on the choice of the exponents. Furthermore, approximation methodology to treat with the exact propagator can also depend on this choice, and thereby many different semiclassical propagators may emerge from these combinations. Indeed, we show that the well-known semiclassical propagators such as those of Van Vleck, Herman-Kluk, Heller's thawed Gaussian, and many others can be derived in a systematic manner, which enables one to comprehend these semiclassical propagators from a unified point of view. We are particularly interested in our generalized form of the Herman-Kluk propagator, since the relative accuracy of this propagator has been well established by Kay, and since, nevertheless, its derivation was not necessarily clear. Thus our generalizedmore » Herman-Kluk propagator replaces the classical Hamiltonian with a Gaussian averaged quantum Hamiltonian, generating non-Newtonian trajectories. We perform a numerical test to assess the quality of such a family of generalized Herman-Kluk propagators and find that the original Herman-Kluk gives an accurate result. The reason why this has come about is also discussed.« less

Is proposed quasi-action variable as a means to analyze the onset of classical chaos in molecular vibrational systems. The basic idea rests on a symplectic area generated by a classical trajectory in phase space, from which the geometrical information of a torus and its breakdown in extracted. The Fourier spectrum of the time derivative of this symplectic area centers on the following definition and findings: (1) in an integrable system, the action variables can be simply calculated in terms of the above Fourier amplitudes, (2) the quasi-action variable is also defined in a similar way and is a good approximation to the corresponding action variable, but (3) the construction of the quasi-action variable does not depend on the integrability and hence it it defined as well even for a chaotic system, and (4) the characteristics of chaos can be analyzed in the continuous spectrum of the quasi-action variable. Some numerical examples of the quasi-action variable are presented for a system of what we call phase...

We present a new concept of large-amplitude motion for molecular vibrations called phase-space large-amplitude motion (PSLAM), which is an outcome of a study of Hamiltonian chaos. The time-dependent behavior of PSLAM is shown to be extremely irregular and unpredictable. PSLAM is anticipated to be identified in experimental observations.

We extend our formerly proposed theory for non-Born-Oppenheimer electronic and nuclear wavepacket dynamics within on-the-fly scheme [T. Yonehara, S. Takahashi, and K. Takatsuka, J. Chem. Phys. 130, 214113 (2009)] to a case of nonadiabatic dynamics under an intense laser field: electron wavepacket in a molecule is propagated in attosecond time-scale along non-Born-Oppenheimer nuclear paths that smoothly branch due to nonadiabatic coupling and/or optical interactions. Such branching paths are determined consistently with the motion of the electron wavepackets. Furthermore, these nuclear paths are quantized in terms of Gaussian wavepackets (action decomposed function), which can be applied to nonclassical paths. Both electronic wavepacket dynamics and quantization of non-Born-Oppenheimer paths are generalized so as to include the direct effects of the classical vector potential of electromagnetic fields. In the second half of this paper, we perform numerical studies to explore nonadiabatic dynamics in a laser field by examining two cases: one is a two-state model system having an avoided crossing, and the other is two-state dynamics in HF molecule on the two low lying ab initio potential curves. Both are placed in laser fields. With the former system, we survey some basic properties of the coupling of nonadiabatic dynamics and laser interaction varying the relevant coupling parameters such as the laser timing with respect to the incident of nonadiabatic transition. This investigation will set a foundation for the future studies of control of electronic states in realistic multidimensional molecular systems. Application to the latter system shows that non-Born-Oppenheimer quantum chemistry in laser fields is indeed useful in the study of dynamics in ab initio level. Through the comparison with full quantum data, we verify that the formalism and methodology developed here work accurately. Furthermore, we attain some basic insight about the characteristics of molecules in laser fields.

We report the results of theoretical studies of the time-resolved femtosecond photoelectron spectroscopy of quantum wavepackets through the conical intersection between the first two (2)A' states of NO(2). The Hamiltonian explicitly includes the pump-pulse interaction, the nonadiabatic coupling due to the conical intersection between the neutral states, and the probe interaction between the neutral states and discretized photoelectron continua. Geometry- and energy-dependent photoionization matrix elements are explicitly incorporated in these studies. Photoelectron angular distributions are seen to provide a clearer picture of the ionization channels and underlying wavepacket dynamics around the conical intersection than energy-resolved spectra. Time-resolved photoelectron velocity map images are also presented.

The fast fourier transformation (FFT) is well known to be extremely fast and useful. However, its spectrum is quite often not accurate, because it is a discrete transformation and, further, the effect of finite range of sampling, the so-called Gibbs phenomenon, produces long tails. Here a very simple and efficient method to extract the accurate frequencies and the amplitudes of discrete spectra from FFT data is proposed. No window function is used in the present method. Indeed, our numerical examples show that the resultant frequencies and amplitudes are extremely accurate.

A practical quantum theory for unifying electronic and nuclear dynamics, which were separated by the Born–Oppenheimer approximation, is proposed. The theory consists of two processes. Nonadiabatic (quantum) electron wavepacket dynamics on branching (non-Born–Oppenheimer) nuclear paths are first constructed. Since these paths are not the classical trajectories, most of the existing semiclassical theories to generate quantum wavepacket do not work. Therefore, we apply our own developed semiclassical wavepacket theory to these generated non-Born–Oppenheimer paths. This wavepacket is generated based on what we call the action decomposed function, which does not require the information of the so-called stability matrix. Thus, the motion of nuclei is also quantized, and consequently the total wave function is represented as a series of entanglement between the electronic and nuclear wavepackets. In the last half of the article, we show the practice to demonstrate how these independent theories can be unified to give electron-nuclear wavepackets in a two-state model. The wavepackets up to the phases and resultant transition probabilities are compared to the full quantum-mechanical counterparts. It turns out that the lowest level approximation to the wavepacket approach already shows a good agreement with the full quantum quantities. Thus, the present theoretical framework gives a basic method with which to study non-Born–Oppenheimer electronic and nuclear wavepacket states relevant to ultrafast chemical events.

Since the early stage of the study of Hamilton chaos, semiclassical quantization based on the low‐order Wentzel‐Kramers‐Brillouin (WKB) theory and their variants have been suffering from difficulties such as divergence in the correlation function, non‐convergence in the trace formula, and so on. It is now widely recognized that the essential drawback of these semiclassical theories commonly originates from the erroneous feature of the amplitude factors based on the stability matrix, when applied to classically chaotic dynamics. In this article, we clarify the mechanism of semiclassical quantization of energy spectrum with the Fourier analysis of phase interference in a trajectory‐wise time correlation function. It is accordingly shown that the energy quantization of chaos in semiclassical regime is, in principle, possible in terms of constructive and destructive interference of the phases alone, and the role of the semiclassical amplitude factor is indeed negligibly small.

Methods to search for periodic orbits are usually implemented with the Newton-Raphson type algorithms that extract the orbits as fixed points. When used to find periodic orbits in flows, however, many such approaches have focused on using mappings defined on the Poincare surfaces of section, neglecting components perpendicular to the surface of section. We propose a Newton-Raphson based method for Hamiltonian flows that incorporates these perpendicular components by using the full monodromy matrix. We investigated and found that inclusion of these components is crucial to yield an efficient process for converging upon periodic orbits in high dimensional flows. Numerical examples with as many as nine degrees of freedom are provided to demonstrate the effectiveness of our method.

The mechanism of energy quantization is studied for classical dynamics on a highly anharmonic potential, ranging from integrable, mixed, and chaotic motions. The quantum eigenstates (standing waves) are created by the phase factors (the action integrals and the Maslov index) irrespective of the integrability, when the amplitude factors are relatively slowly varying. Indeed we show numerically that the time Fourier transform of an approximate semiclassical correlation function in which the amplitude factors are totally removed reproduces the spectral positions (energy eigenvalues) accurately in chaotic regime. Quantization with the phase information alone brings about dramatic simplification to molecular science, since the amplitude factors in the lowest order semiclassical approximation diverge exponentially in a chaotic domain.

A serious difficulty in the semiclassical Ehrenfest theory for nonadiabatic transitions is that a path passing across the avoided crossing is forced to run on a potential averaged over comprising adiabatic potential surfaces that commit the avoided crossing. Therefore once a path passes through the crossing region, it immediately becomes incompatible with the standard view of "classical trajectory" running on an adiabatic surface. This casts a fundamental question to the theoretical structure of chemical dynamics. In this paper, we propose a non-Born-Oppenheimer path that is generated by an anti-Hermitian Hamiltonian, whose complex-valued eigenenergies can cross in their real parts and avoid crossing in the imaginary parts in the vicinity of the nonadiabatic transition region. We discuss the properties of this non-Born-Oppenheimer path and thereby show its compatibility with the Born-Oppenheimer classical trajectories. This theory not only allows the geometrical branching of the paths but gives the nonadiabatic transition amplitudes and quantum phases along the generated paths.

A renormalized semiclassical quantization method for rescalable Hamiltonians is proposed. A classical Hamilton system having a potential function that consists of homogeneous polynomials like the Coulombic potential can have a scale invariance in its extended phase space (phase space plus time). Consequently, infinitely many copies of a single trajectory constitute a one-parameter family that is characterized in terms of a scaling factor. This scaling invariance in classical dynamics is lost in quantum mechanics due to the presence of the Planck constant. It is shown that in a system whose classical motions have a self-similarity in the above sense, classical trajectories adopted in the semiclassical scheme interact with infinitely many copies of their own that are reproduced by the relevant scaling procedure, thereby undergoing quantum interference among themselves to produce a quantized spectrum.

We develop the path-branching representation for nonadiabatic electron wavepacket dynamics [T. Yonehara and K. Takatsuka, J. Chem. Phys. 132, 244102 (2010)] so as to treat dynamics in an energy range comparable to the barrier height of adiabatic potential energy curves. With this representation two characteristic chemical reaction dynamics are studied, in which an incident nuclear wavepacket encounters a potential barrier, on top of which lies another nonadiabatically coupled adiabatic potential curve: (1) Dynamics of initial paths coming into the nonadiabatic interaction region with energy lower than the barrier height. They branch into two pieces (and repeat branching subsequently), the upper counterparts of which can penetrate into a classically inaccessible high energy region and eventually branch back to the product region on the ground state curve. This is so to say surmounting the potential barrier via nonadiabatically coupled excited state, and phenomenologically looks like the so-called deep tunneling. (2) Dynamics of classical paths whose initial energies are a little higher than the barrier but may be lower than the bottom of the excited state. They can undergo branching and some of those components are trapped on top of the potential barrier, being followed by the population decay down to the lower state flowing both to product and reactant sites. Such expectations arising from the path-branching representation are numerically confirmed with full quantum mechanical wavepacket dynamics. This phenomenon may be experimentally observed as time-delayed pulses of wavepacket trains.

Chaos in the multimode nonadiabatic system constructed by Heller [J. Chem. Phys. 92, 1718 (1990)], which consists of two diabatic two-dimensional harmonic potentials with the Condon coupling, is studied. A thorough investigation is carried out by scanning the magnitudes of the Condon coupling and the Duschinsky angle. To elucidate mechanisms that can cause chaos in this quantum system, the statistical properties of the energy levels and eigenfunctions of the system are investigated. We find an evidence in terms of the nearest-neighbor spacing distribution of energy levels and other measures that a certain class of chaos is purely induced by the nonadiabatic interaction due to breakdown of the Born-Oppenheimer approximation. Since the nonadiabatic transition can induce repeated bifurcation and merging of a wave packet around the region of quasicrossing between two potential surfaces, and since this interaction does not have a counterpart in the lower adiabatic system, the present chaos deserves being called "nonadiabatic chaos." Another type of chaos in a nonadiabatic system was previously identified [D. M. Leitner et al., J. Chem. Phys. 104, 434 (1996)] that reflects the inherent chaos of a corresponding adiabatic potential. We present a comparative study to establish the similarity and difference between these kinds of chaos.

We derive an extended expression of the density of states for Hamiltonian chaos in the present paper and will discuss the possibility of explicit construction of the quantization condition for chaos in a future article. In view of the controversial validity of Gutzwiller's density of states, we apply our phase-space semiclassical mechanics [K. Takatsuka, Phys. Rev. Lett. 61, 503 (1988); Phys. Rev. A 39, 5961 (1989)] in order to construct independently the expression for both regular and irregular spectra in a unified manner. It is shown that the Liapunov exponent, Greene's residue in classical chaos, and the Maslov index in the Einstein-Brillouin-Keller conditions are closely related to each other through the amplitude factor of our phase-space kernel. In particular, the Maslov index is interpreted as a quantum-mechanical phase due to the spinning motion of a volume element that is to be carried by a phase flow along a periodic orbit. The difficulty in semiclassical mechanics is argued in terms of the exponential decay of a time correlation function. Furthermore, a formal aspect of the thermodynamic characterization of quantum chaos is addressed, defining the dynamical temperature and entropy through the analytic structure of the density of states.

A theory of quantum electron wavepacket dynamics that nonadiabatically couples with classical nuclear motions in intense optical fields is studied. The formalism is intended to track the laser-driven electron wavepackets in terms of the linear combination of configuration-state functions generated with ab initio molecular orbitals. Beginning with the total quantum Hamiltonian for electrons and nuclei in the vector potential of classical electromagnetic field, we reduce the Hamiltonian into a mixed quantum-classical representation by replacing the quantum nuclear momentum operators with the classical counterparts. This framework gives equations of motion for electron wavepackets in an intense laser field through the time dependent variational principle. On the other hand, a generalization of the Newtonian equations provides a matrix form of forces acting on the nuclei for nonadiabatic dynamics. A mean-field approximation to the force matrix reduces this higher order formalism to the semiclassical Ehrenfest theory in intense optical fields. To bring these theories into a practical quantum chemical package for general molecules, we have implemented the relevant ab initio algorithms in it. Some numerical results in the level of the semiclassical Ehrenfest-type theory with explicit use of the nuclear kinematic (derivative) coupling and the velocity form for the optical interaction are presented.

We propose a nonempirical statistical theory to give the reaction rate and the kinetic energy distribution of fragments for molecular evaporation from highly nonrigid atomic and van der Waals clusters. To quantify the theory, an efficient and accurate method to evaluate the absolute value of classical density of states (the Thomas-Fermi density in phase space) and the flux at the so-called dividing surface is critically important, and we have devised such an efficient method. The theory and associated methods are verified by numerical comparison with the corresponding molecular dynamics simulation through the study of Ar(2) evaporation from Ar(8) cluster, in which evaporation is strongly coupled with structural isomerization dynamics. It turns out that the nonempirical statistical theory gives quite an accurate reaction rate. We also study the kinetic energy release (KER) arising from these evaporations and its Boltzmann-like distribution both for atomic and diatomic evaporations. This provides a general relation between the KER and temperature of the fragments.

Since the early stage of the study of Hamilton chaos, semiclassical quantization based on the low-order Wentzel-Kramers-Brillouin theory, the primitive semiclassical approximation to the Feynman path integrals (or the so-called Van Vleck propagator), and their variants have been suffering from difficulties such as divergence in the correlation function, nonconvergence in the trace formula, and so on. These difficulties have been hampering the progress of quantum chaos, and it is widely recognized that the essential drawback of these semiclassical theories commonly originates from the erroneous feature of the amplitude factors in their applications to classically chaotic systems. This forms a clear contrast to the success of the Einstein-Brillouin-Keller quantization condition for regular (integrable) systems. We show here that energy quantization of chaos in semiclassical regime is, in principle, possible in terms of constructive and destructive interference of phases alone, and the role of the semiclassical amplitude factor is indeed negligibly small, as long as it is not highly oscillatory. To do so, we first sketch the mechanism of semiclassical quantization of energy spectrum with the Fourier analysis of phase interference in a time correlation function, from which the amplitude factor is practically factored out due to its slowly varying nature. In this argument there is no distinction between integrability and nonintegrability of classical dynamics. Then we present numerical evidence that chaos can be indeed quantized by means of amplitude-free quasicorrelation functions and Heller's frozen Gaussian method. This is called phase quantization. Finally, we revisit the work of Yamashita and Takatsuka [Prog. Theor. Phys. Suppl. 161, 56 (2007)] who have shown explicitly that the semiclassical spectrum is quite insensitive to smooth modification (rescaling) of the amplitude factor. At the same time, we note that the phase quantization naturally breaks down when the oscillatory nature of the amplitude factor is comparable to that of the phases. Such a case generally appears when the Planck constant of a large magnitude pushes the dynamics out of the semiclassical regime.

The validity range of the Born-Oppenheimer (BO) approximation is studied with respect to the variation of the mass (m) of negatively charged particle by substituting an electron (e) with muon (mu) and antiproton (p) in hydrogen molecule cation. With the use of semiclassical quantization applied to these (ppe), (ppmu), and (ppp) under a constrained geometry, we estimate the energy difference of the non-BO vibronic ground state from the BO counterpart. It is found that the error in the BO approximation scales to the power of 3/2 to the mass of negative particles, that is, m(1.5). The origin of this clear-cut relation is analyzed based on the original perturbation theory due to Born and Oppenheimer, with which we show that the fifth order term proportional to m(5/4) is zero and thereby the first correction to the BO approximation should arise from the sixth order term that is proportional to m(6/4). Therefore, the validity range of the Born-Oppenheimer approximation is wider than that often mistakenly claimed to be proportional to m(1/4).

The role of destructive quantum interference in semiclassical quantization of molecular vibrational states is studied. This aspect is crucial for correct quantization, since failure in the appropriate treatment of destructive interference quite often results in many spurious peaks and broad background to hide the true peaks. We first study the time-Fourier transform of the autocorrelation function without performing summation over the trajectories. The resultant quantity, the prespectrum which is a function of individual classical trajectories, provides a clear view about how destructive interference among the trajectories should function. It turns out that the prespectrum is oscillatory but never a random noise. On the contrary, it bears a systematic and regular structure, which is sometimes characterized in terms of very sharp and high peaks in the energy space of the sampled classical trajectories. We have found an extended quantization condition that is responsible for generating these peaks in the prespectrum, which we call the prior quantization condition. Integration of the prespectrum over the trajectory space is supposed to give "zero" (practically a small value of the order of the Planck constant) at a noneigenvalue energy, which is actually a materialization of the destructive interference. Besides, certain finite peaks in the prespectrum survive after the integration to form the true spikes (eigenvalues) in the final spectrum, if they satisfy an additional resonance condition. For these resonance components, the prior quantization condition is reduced to the Einstein-Brillouin-Keller quantization condition. Based on these analyses, we propose a rather conventional filtering technique to efficiently handle tedious computation for destructive interference, and numerically verify that it works well even for multidimensional chaotic systems. This filtering technique is further utilized to extract a few trajectories that dominate an eigenstate of molecular vibration.

We propose a variational method for scattering in which the functional is of a fractional form as for the Schwinger variational principle. However, our functional does not involve the Green's function, but the Hamiltonian and the potential function. This method shows features of both the Schwinger-type variational principles and the Kohn-type standard variational principles. As a result, our method can derive distinct advantages from both of these approaches. The resultant K matrix is symmetric and anomaly-free. Some other properties, including a minimum principle, which is useful in the selection of an optimum basis for the expansion of the scattering functions are also discussed.

In this contribution we want to discuss several features and applications of the Schwinger variational principle to the study of collisions of low energy electrons with molecules and molecular ions. The Schwinger variational principle has long been known to be a potentially useful formulation of the collision problem but until recently there have been very few applications of this variational principle to electron collision problems.1

In this paper we reply to the preceding Comment by Abdel-Raouf concerning several features of an amplitude-independent variational principle which we have recently proposed. The most substantive discussion concerns the nature of the singularities which can arise in one of these variational principles. The specific variational principle involved contains the Hamiltonian and not the Green's-function operator.

The dynamical characteristic function (DCF) introduced previously as a kind of phase space distribution function is generalized so as to give an overlap integral of two wave packets which are to be propagated on different potential energy hypersurfaces. The development of our new semiclassical theory is motivated by the fact that the scattering (S) matrix is just one of this kind of overlap integrals. In this theory the semiclassical DCF is evolved in time by running a pair of classical trajectories, which are determined by two different Hamiltonians, total scattering Hamiltonian of the system, and unperturbed final channel Hamiltonian. The DCF becomes an overlap integral of two wave packets, if these two trajectories coincide with each other in the exit region at t=∞. The validity of this semiclassical theory is shown to be ensured, if the oscillatory wave packets are employed to construct the DCF. The S matrix in the stationary state scattering theory is given as a superposition of the wave packet DCF’s.

A new orbital theory is proposed, in which general spin orbitals (GSO) are introduced in the spin‐optimized (SO) SCF scheme. In this SO–SCF–GSO theory, the effective Hamiltonian for each orbital takes the form of a 2×2 matrix composed of the eigenfunctions for two‐component spinors. It is found that the GSO’s thus defined should still satisfy a general form of Koopmans’ theorem. The SO–SCF GSO’s are to be obtained by solving two sets of coupled SCF equations for the spin coupling coefficients and the linear combination coefficients for basis functions. Using an STO‐6G basis set of the double ζ quality, sample calculations have been carried out for the doublet state of the linear H3 system for which the bond lengths are fixed at 1.470 and 2.984 bohr. The total energy obtained is ∼3 kcal/mole lower than the values which have resulted from the SO–SCF–DODS and the spin‐extended Hartree–Fock (SEHF) GSO calculations with the same basis set. The resulting orbitals are found to be more delocalized over the entire system than those obtained by the SO–SCF–DODS theory.