Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

TO DECAY OR NOT TO DECAY? TEMPORAL BEHAVIOR OF A QUANTUM SYSTEM—ANALYSIS BASED ON A SOLVABLE MODEL—

Hiromichi Nakazato

Published Feb 1, 2003• 0 citations
Publisher

Abstract

The exact expression of the survival amplitude of a model system, which may describe photodetachment of a negative ion, is derived and analyzed in detail with an attention on its behavior as a function of time. The latter is governed by the location of poles of the amplitude in the complex energy plane. Poles located on the real energy axis can prevent the system from decay (known as plasmon poles), while other poles on the second Riemannian sheet give rise to exponentially decaying terms. The solvability of the model enables us to see explicitly the analytic structure of the amplitude, e.g., the movement of the poles in the complex energy plane as functions of parameters of the model. The behavior both at short and long times is also derived exactly and its peculiarity is pointed out and discussed.

DOI
10.1142/9789812704412_0015

Related papers

TEMPORAL BEHAVIOR OF QUANTUM MECHANICAL SYSTEMS

Hiromichi Nakazato, Mikio Namiki, Saverio Pascazio

The temporal behavior of quantum mechanical systems is reviewed. We mainly focus our attention on the time development of the so-called “survival” probability of those systems that are initially prepared in eigenstates of the unperturbed Hamiltonian, by assuming that the latter has a continuous spectrum. The exponential decay of the survival probability, familiar, for example, in radioactive decay phenomena, is representative of a purely probabilistic character of the system under consideration and is naturally expected to lead to a master equation. This behavior, however, can be found only at intermediate times, for deviations from it exist both at short and long times and can have significant consequences. After a short introduction to the long history of the research on the temporal behavior of such quantum mechanical systems, the short-time behavior and its controversial consequences when it is combined with von Neumann’s projection postulate in quantum measurement theory are critically overviewed from a dynamical point of view. We also discuss the so-called quantum Zeno effect from this standpoint. The behavior of the survival amplitude is then scrutinized by investigating the analytic properties of its Fourier and Laplace transforms. The analytic property that there is no singularity except a branch cut running along the real energy axis in the first Riemannian sheet is an important reflection of the time-reversal invariance of the dynamics governing the whole process. It is shown that the exponential behavior is due to the presence of a simple pole in the second Riemannian sheet, while the contribution of the branch point yields a power behavior for the amplitude. The exponential decay form is cancelled at short times and dominated at very long times by the branch-point contributions, which give a Gaussian behavior for the former and a power behavior for the latter. In order to realize the exponential law in quantum theory, it is essential to take into account a certain kind of macroscopic nature of the total system, since the exponential behavior is regarded as a manifestation of a complete loss of coherence of the quantum subsystem under consideration. In this respect, a few attempts at extracting the exponential decay form on the basis of quantum theory, aiming at the master equation, are briefly reviewed, including van Hove’s pioneering work and his well-known “λ2T” limit. In the attempt to further clarify the mechanism of the appearance of a purely probabilistic behavior without resort to any approximation, a solvable dynamical model is presented and extensively studied. The model describes an ultrarelativistic particle interacting with N two-level systems (called “spins”) and is shown to exhibit an exponential behavior at all times in the weak-coupling, macroscopic limit. Furthermore, it is shown that the model can even reproduce the short-time Gaussian behavior followed by the exponential law when an appropriate initial state is chosen. The analysis is exact and no approximation is involved. An interpretation for the change of the temporal behavior in quantum systems is drawn from the results obtained. Some implications for the quantum measurement problem are also discussed, in particular in connection with dissipation.

arXiv:quant-ph/9509016•International Journal of Modern Physics B•Jan 30, 1996•203 citations

Decaying states as complex energy eigenvectors in generalized quantum mechanics

E. C. G. Sudarshan, Charles B. Chiu, Vittorio Gorini

We reexamine the problem of particle decay within the Hamiltonian formalism. By deforming contours of integration, the survival amplitude is expressed as a sum of purely exponential contributions arising from the simple poles of the resolvent on the second sheet plus a background integral along a complex contour $\ensuremath{\Gamma}$ running below the location of the poles. We observe that the time dependence of the survival amplitude in the small-time region is strongly correlated to the asymptotic behavior of the energy spectrum of the system; we compute the small-time behavior of the survival amplitude for a wide variety of asymptotic behaviors. In the special case of the Lee model, using a formal procedure of analytic continuation, we show that a complete set of complex energy eigenvectors of the Hamiltonian can be associated with the poles of the resolvent and the background contour $\ensuremath{\Gamma}$. These poles and points along $\ensuremath{\Gamma}$ correspond to the discrete and the continuum states, respectively. In this context, each unstable particle is associated with a well-defined object, which is a discrete generalized eigenstate of the Hamiltonian having a complex eigenvalue, with its real and negative imaginary parts being the mass and half-width of the particle, respectively. Finally, we briefly discuss the analytic continuation of the scattering amplitude within this generalized scheme, and note the appearance of "redundant poles" which do not correspond to discrete solutions of the modified eigenvalue problem.

Physical review. D. Particles, fields, gravitation, and cosmology/Physical review. D. Particles and fields•Oct 15, 1978•126 citations

Survival Amplitude, Instantaneous Energy and Decay Rate of an Unstable System: Analytical Results

K. Raczyńska, K. Urbanowski

We consider a model of a unstable state defined by the truncated Breit-Wigner energy density distribution function. An analytical form of the survival amplitude $a(t)$ of the state considered is found. Our attention is focused on the late time properties of $a(t)$ and on effects generated by the non--exponential behavior of this amplitude in the late time region: In 1957 Khalfin proved that this amplitude tends to zero as $t$ goes to the infinity more slowly than any exponential function of $t$. This effect can be described using a time-dependent decay rate $γ(t)$ and then the Khalfin result means that this $γ(t)$ is not a constant but at late times it tends to zero as $t$ goes to the infinity. It appears that the energy $E(t)$ of the unstable state behaves similarly: It tends to the minimal energy $E_{min}$ of the system as $t \to \infty$. Within the model considered we find two first leading time dependent elements of late time asymptotic expansions of $E(t)$ and $γ(t)$. We discuss also possible implications of such a late time asymptotic properties of $E(t)$ and $γ(t)$ and cases where these properties may manifest themselves.

arXiv:1802.01441•Acta Physica Polonica B•Jan 1, 2018•12 citations

Analytical study of quadratic and nonquadratic short-time behavior of quantum decay

Sergio Cordero, Gastón García‐Calderón

The short-time behavior of quantum decay of an unstable state initially located within an interaction region of finite range is investigated using a resonant expansion of the survival amplitude. It is shown that in general the short-time behavior of the survival probability $S(t)$ has a dependence on the initial state and may behave as either $S(t)=1\ensuremath{-}O({t}^{3/2})$ or $1\ensuremath{-}O({t}^{2})$. These cases are illustrated by solvable models. The experiment reported by Wilkinson et al. [Nature (London) 387, 575 (1997)] does not distinguish between the above short-time behaviors.

arXiv:1302.3511•Physical Review A•Dec 26, 2012•16 citations

Analytical approach to the wave function of a decaying quantum system

W. van Dijk, Yukihisa Nogami

Recently we outlined an analytical method of solving the time-dependent Schr\"odinger equation for a model which simulates a decaying quantum system such as an $\ensuremath{\alpha}$-decaying nucleus. A particle in the model is initially confined around the origin and leaks out, tunneling through a potential barrier. The solution can be expressed as a linear combination of the Moshinsky functions, each of which is associated with a pole of the scattering $\mathbf{S}$ matrix of the model. In this paper we give a full account of the method with a few explicit examples. We examine deviations from the exponential decay law at very large times. We comment on a recent controversy regarding the t dependence of the survival and nonescape probabilities when t is very large.

Physical Review C•Jan 16, 2002•42 citations

Nonexponential decay propagator and its differential equation for real and complex energy distributions of unstable states

Theodosios Geo Douvropoulos, Cleanthes A. Nicolaides

The survival amplitude $G(t)$ of a nonstationary state decaying into a purely continuous spectrum is treated in terms of an integral transform of an energy distribution with $\ensuremath{\infty}>E\ensuremath{\geqslant}0$. We examine three such distributions. Two are real functions, the Lorentzian ${g}^{L}(E)$ and a modified Lorentzian $G(E)={g}^{L}(E){E}^{1∕2}$, and one is the complex version of ${g}^{L}(E),{g}_{c}^{L}(E)$. Real distributions are associated with Hermitian treatments while complex ones result from non-Hermitian treatments. The difference between the two has repercussions on the $G(t)$ for nonexponential decay (NED) and on the understanding of irreversible decay at the quantum level. For all three distributions, we derive analytically amplitudes (propagators) for NED and then show that these satisfy differential equations, from which additional insight into the decay process for very long and very short times can be obtained. By making analogy with the classical Langevin equation, the terms of the differential equation that are derived when the simpler ${g}^{L}(E)$ and ${g}_{c}^{L}(E)$ are employed, are interpreted using concepts such as friction and fluctuation. On the other hand, when ${g}^{L}(E)$ is multiplied by an energy-dependent factor, as in $G(E)$, the results are, as expected, more complicated and the interpretability of the differential equation satisfied by the NED propagator loses clarity.

Physical Review A•Mar 16, 2004•12 citations
The exact expression of the survival amplitude of a model system, which may describe photodetachment of a negative ion, is derived and analyzed in detail with an attention on its behavior as a function of time. The latter is governed by the location of poles of the amplitude in the complex energy plane. Poles located on the real energy axis can prevent the system from decay (known as plasmon poles), while other poles on the second Riemannian sheet give rise to exponentially decaying terms. The solvability of the model enables us to see explicitly the analytic structure of the amplitude, e.g., the movement of the poles in the complex energy plane as functions of parameters of the model. The behavior both at short and long times is also derived exactly and its peculiarity is pointed out and discussed.
The temporal behavior of quantum mechanical systems is reviewed. We mainly focus our attention on the time development of the so-called “survival” probability of those systems that are initially prepared in eigenstates of the unperturbed Hamiltonian, by assuming that the latter has a continuous spectrum. The exponential decay of the survival probability, familiar, for example, in radioactive decay phenomena, is representative of a purely probabilistic character of the system under consideration and is naturally expected to lead to a master equation. This behavior, however, can be found only at intermediate times, for deviations from it exist both at short and long times and can have significant consequences. After a short introduction to the long history of the research on the temporal behavior of such quantum mechanical systems, the short-time behavior and its controversial consequences when it is combined with von Neumann’s projection postulate in quantum measurement theory are critically overviewed from a dynamical point of view. We also discuss the so-called quantum Zeno effect from this standpoint. The behavior of the survival amplitude is then scrutinized by investigating the analytic properties of its Fourier and Laplace transforms. The analytic property that there is no singularity except a branch cut running along the real energy axis in the first Riemannian sheet is an important reflection of the time-reversal invariance of the dynamics governing the whole process. It is shown that the exponential behavior is due to the presence of a simple pole in the second Riemannian sheet, while the contribution of the branch point yields a power behavior for the amplitude. The exponential decay form is cancelled at short times and dominated at very long times by the branch-point contributions, which give a Gaussian behavior for the former and a power behavior for the latter. In order to realize the exponential law in quantum theory, it is essential to take into account a certain kind of macroscopic nature of the total system, since the exponential behavior is regarded as a manifestation of a complete loss of coherence of the quantum subsystem under consideration. In this respect, a few attempts at extracting the exponential decay form on the basis of quantum theory, aiming at the master equation, are briefly reviewed, including van Hove’s pioneering work and his well-known “λ2T” limit. In the attempt to further clarify the mechanism of the appearance of a purely probabilistic behavior without resort to any approximation, a solvable dynamical model is presented and extensively studied. The model describes an ultrarelativistic particle interacting with N two-level systems (called “spins”) and is shown to exhibit an exponential behavior at all times in the weak-coupling, macroscopic limit. Furthermore, it is shown that the model can even reproduce the short-time Gaussian behavior followed by the exponential law when an appropriate initial state is chosen. The analysis is exact and no approximation is involved. An interpretation for the change of the temporal behavior in quantum systems is drawn from the results obtained. Some implications for the quantum measurement problem are also discussed, in particular in connection with dissipation.
We reexamine the problem of particle decay within the Hamiltonian formalism. By deforming contours of integration, the survival amplitude is expressed as a sum of purely exponential contributions arising from the simple poles of the resolvent on the second sheet plus a background integral along a complex contour $\ensuremath{\Gamma}$ running below the location of the poles. We observe that the time dependence of the survival amplitude in the small-time region is strongly correlated to the asymptotic behavior of the energy spectrum of the system; we compute the small-time behavior of the survival amplitude for a wide variety of asymptotic behaviors. In the special case of the Lee model, using a formal procedure of analytic continuation, we show that a complete set of complex energy eigenvectors of the Hamiltonian can be associated with the poles of the resolvent and the background contour $\ensuremath{\Gamma}$. These poles and points along $\ensuremath{\Gamma}$ correspond to the discrete and the continuum states, respectively. In this context, each unstable particle is associated with a well-defined object, which is a discrete generalized eigenstate of the Hamiltonian having a complex eigenvalue, with its real and negative imaginary parts being the mass and half-width of the particle, respectively. Finally, we briefly discuss the analytic continuation of the scattering amplitude within this generalized scheme, and note the appearance of "redundant poles" which do not correspond to discrete solutions of the modified eigenvalue problem.
We consider a model of a unstable state defined by the truncated Breit-Wigner energy density distribution function. An analytical form of the survival amplitude a(t)a(t)a(t) of the state considered is found. Our attention is focused on the late time properties of a(t)a(t)a(t) and on effects generated by the non--exponential behavior of this amplitude in the late time region: In 1957 Khalfin proved that this amplitude tends to zero as ttt goes to the infinity more slowly than any exponential function of ttt. This effect can be described using a time-dependent decay rate γ(t)γ(t)γ(t) and then the Khalfin result means that this γ(t)γ(t)γ(t) is not a constant but at late times it tends to zero as ttt goes to the infinity. It appears that the energy E(t)E(t)E(t) of the unstable state behaves similarly: It tends to the minimal energy EminE_{min}Emin​ of the system as t→∞t \to \inftyt→∞. Within the model considered we find two first leading time dependent elements of late time asymptotic expansions of E(t)E(t)E(t) and γ(t)γ(t)γ(t). We discuss also possible implications of such a late time asymptotic properties of E(t)E(t)E(t) and γ(t)γ(t)γ(t) and cases where these properties may manifest themselves.
The short-time behavior of quantum decay of an unstable state initially located within an interaction region of finite range is investigated using a resonant expansion of the survival amplitude. It is shown that in general the short-time behavior of the survival probability S(t)S(t)S(t) has a dependence on the initial state and may behave as either $S(t)=1\ensuremath{-}O({t}^{3/2})$ or $1\ensuremath{-}O({t}^{2})$. These cases are illustrated by solvable models. The experiment reported by Wilkinson et al. [Nature (London) 387, 575 (1997)] does not distinguish between the above short-time behaviors.
Recently we outlined an analytical method of solving the time-dependent Schr\"odinger equation for a model which simulates a decaying quantum system such as an $\ensuremath{\alpha}$-decaying nucleus. A particle in the model is initially confined around the origin and leaks out, tunneling through a potential barrier. The solution can be expressed as a linear combination of the Moshinsky functions, each of which is associated with a pole of the scattering S\mathbf{S}S matrix of the model. In this paper we give a full account of the method with a few explicit examples. We examine deviations from the exponential decay law at very large times. We comment on a recent controversy regarding the t dependence of the survival and nonescape probabilities when t is very large.
The survival amplitude G(t)G(t)G(t) of a nonstationary state decaying into a purely continuous spectrum is treated in terms of an integral transform of an energy distribution with $\ensuremath{\infty}>E\ensuremath{\geqslant}0$. We examine three such distributions. Two are real functions, the Lorentzian gL(E){g}^{L}(E)gL(E) and a modified Lorentzian G(E)=gL(E)E1∕2G(E)={g}^{L}(E){E}^{1∕2}G(E)=gL(E)E1∕2, and one is the complex version of gL(E),gcL(E){g}^{L}(E),{g}_{c}^{L}(E)gL(E),gcL​(E). Real distributions are associated with Hermitian treatments while complex ones result from non-Hermitian treatments. The difference between the two has repercussions on the G(t)G(t)G(t) for nonexponential decay (NED) and on the understanding of irreversible decay at the quantum level. For all three distributions, we derive analytically amplitudes (propagators) for NED and then show that these satisfy differential equations, from which additional insight into the decay process for very long and very short times can be obtained. By making analogy with the classical Langevin equation, the terms of the differential equation that are derived when the simpler gL(E){g}^{L}(E)gL(E) and gcL(E){g}_{c}^{L}(E)gcL​(E) are employed, are interpreted using concepts such as friction and fluctuation. On the other hand, when gL(E){g}^{L}(E)gL(E) is multiplied by an energy-dependent factor, as in G(E)G(E)G(E), the results are, as expected, more complicated and the interpretability of the differential equation satisfied by the NED propagator loses clarity.