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

Nuovo cimento della Società italiana di fisica. A, Nuclei, particles and fields

Time operator in quantum mechanics: I: Nonrelativistic case

V. S. Olkhovsky, Erasmo Recami, A. J. Gerasimchuk

Published Jul 1, 1974• 79 citations
Publisher
DOI
10.1007/bf02813438

Related papers

On the Time Operator in Quantum Mechanics. II: -- General Prescription in One Dimensional Case --

Toru Goto, S. Naka, K. Yamaguchi

Progress of Theoretical Physics•Dec 1, 1981•10 citations

Non-self-adjoint operators as observables in quantum theory and nuclear physics

V. S. Olkhovsky, Sergei P. Maydanyuk, Erasmo Recami

The aim of this paper is to show the importance and applicability of non-self-adjoint operators for description of observables both in relativistic and nonrelativistic quantum mechanics, as well as in quantum electrodynamics. The paper considers (i) a maximally Hermitian (but non-self-adjoint) time operator in nonrelativistic quantum mechanics and in quantum electrodynamics and (ii) the problem of defining the four-dimensional coordinate and momentum operators describing relativistic particles with zero spin (each of these operators contains both Hermitian and anti-Hermitian parts). Some other applications of non-selfadjoint and non-Hermitian operators in physics are analyzed as well. The paper ends with an analysis of a few cases of T -non-invariant interactions, including quantum dissipation and nuclear optical potentials.

Physics of Particles and Nuclei•Jul 1, 2010•6 citations

Time and energy operators in quantum mechanics

Chun-Fang Li, Qi Wang

Journal of Shanghai University (English Edition)•Mar 1, 1998•1 citation

The Dirac Equation and Time Flux

Francesco R. Ruggeri

It was argued in a number of previous notes that for the nonrelativistic case, W(x,t) transforms under x->x+a, t-> t+b (with a and b infinitesimal) as exp(ipx - iEt), except with p replaced with = - i d/dx W(x,t) and E replaced with i d/dt W(x,t). It was suggested this was a driving condition of nonrelativistic quantum mechanics, and if this is the case, one might expect these ideas to carry over into relativistic quantum mechanics. For relativistic quantum mechanics, however, one has the Einstein energy-momentum relationship E*E= p*p + mo*mo (c=1) which leads to -d/dt d/dt W(x,t) and -d/dx d/dx W(x,t) being used. The momentum term is fine because it is related to < p*p conditional >, but the second derivative of time seems to be a little confusing as it represents < E*E conditional > and not <E conditional >. If <E conditional> has physical significance, as it seems to be the case in nonrelativistic quantum mechanics, one might expect an equation with this time flux i.e. i d/dt W(x,t) / W(x,t) = < E conditional > and not < E*E conditional >. Dirac wrote such an equation, but obtaining a single time derivative comes at a price. The Dirac equation, in one dimension, contains matrices, d/dx and d/dt in addition to a two vector wavefunction (U(x,t), V(x,t)) instead of W(x,t). Thus, the Dirac equation in one dimension leads to coupled equations with U(x,t) and V(x,t). The single derivative d/dx is deceptive because U(x,t) can be written in terms of d/dx V(x,t) and vice versa, so ultimately the Dirac equations contain d/dx d/dx which is needed to obtain < p*p conditional >. The question is: How can one obtain < E conditional >? This may be done by using a (d/dt + constant ) -1 operator i,e. almost an inverse time operator, in other words either (E+m)-1 or (E-m) -1 where E can be thought of as id/dt. Thus, the RHS of the Dirac equations is i d/dt U(x,t) or i d/dt V(x,t) which suggests an infinitesimal step forward in time, while the LHS side contains (id/dt +m)-1 or (-id/dt-m) -1 which seems to be representative of undoing such a step. It should be noted that one may take Einstein’s energy-momentum equation E*E = p*p + mo*mo, without any knowledge of quantum mechanics, and cast it as 2x2 matrix equation, linear in E, namely H (U,V) = E (U,V). The matrix H serves as an evolution operator and another matrix serves as a velocity operator. One may derive the Schrodinger zitterbewegung equation for the velocity matrix without any use of wavefunctions and derivatives which are key to quantum mechanics. Nevertheless, the idea of noncommuting matrices, vectors and zitterbewegung is present. There are several choices for the matrix H, but one leads to the one dimensional Dirac coupled equations Thus, the “spin” features may be obtained directly from E*E= p*p+mo*mo without any quantum mechanics. (1), (2). The purpose of this note is to examine some of these ideas.

•Jul 24, 2019•0 citations

A time operator in quantum mechanics

Annals of Physics•Sep 1, 1983•4 citations

One-Particle Quantum Mechanics-General Formalism

Stephen L. Adler

Abstract We begin the discussion of nonrelativistic quaternionic quantum mechanics by analyzing the allowed form of H for a single-particle system. We will assume, as is done in complex nonrelativistic quantum mechanics, that H is invariant under Galilean transformations and that the kinetic or noninteracting part of H is rotation and translation invariant. In analyzing the implications of these assumptions, we follow the method used in Sec. 13-4 of Jauch (1968a) in the complex quantum mechanics case [see also Mackey (1968), and Feynman as described in Dyson (1990)], and so work in the coordinate representation (cf. Sec. 2.4), and use the Heisenberg picture (cf. Sec. 3.3) to describe time evolution. For brevity of notation, we omit the subscript H for Heisenberg picture operators and denote the coordinate representation Hamiltonian H(x) as simply H2.

Oxford University Press eBooks•Apr 27, 1995•0 citations
The aim of this paper is to show the importance and applicability of non-self-adjoint operators for description of observables both in relativistic and nonrelativistic quantum mechanics, as well as in quantum electrodynamics. The paper considers (i) a maximally Hermitian (but non-self-adjoint) time operator in nonrelativistic quantum mechanics and in quantum electrodynamics and (ii) the problem of defining the four-dimensional coordinate and momentum operators describing relativistic particles with zero spin (each of these operators contains both Hermitian and anti-Hermitian parts). Some other applications of non-selfadjoint and non-Hermitian operators in physics are analyzed as well. The paper ends with an analysis of a few cases of T -non-invariant interactions, including quantum dissipation and nuclear optical potentials.
It was argued in a number of previous notes that for the nonrelativistic case, W(x,t) transforms under x->x+a, t-> t+b (with a and b infinitesimal) as exp(ipx - iEt), except with p replaced with = - i d/dx W(x,t) and E replaced with i d/dt W(x,t). It was suggested this was a driving condition of nonrelativistic quantum mechanics, and if this is the case, one might expect these ideas to carry over into relativistic quantum mechanics. For relativistic quantum mechanics, however, one has the Einstein energy-momentum relationship E*E= p*p + mo*mo (c=1) which leads to -d/dt d/dt W(x,t) and -d/dx d/dx W(x,t) being used. The momentum term is fine because it is related to < p*p conditional >, but the second derivative of time seems to be a little confusing as it represents < E*E conditional > and not <E conditional >. If <E conditional> has physical significance, as it seems to be the case in nonrelativistic quantum mechanics, one might expect an equation with this time flux i.e. i d/dt W(x,t) / W(x,t) = < E conditional > and not < E*E conditional >. Dirac wrote such an equation, but obtaining a single time derivative comes at a price. The Dirac equation, in one dimension, contains matrices, d/dx and d/dt in addition to a two vector wavefunction (U(x,t), V(x,t)) instead of W(x,t). Thus, the Dirac equation in one dimension leads to coupled equations with U(x,t) and V(x,t). The single derivative d/dx is deceptive because U(x,t) can be written in terms of d/dx V(x,t) and vice versa, so ultimately the Dirac equations contain d/dx d/dx which is needed to obtain < p*p conditional >. The question is: How can one obtain < E conditional >? This may be done by using a (d/dt + constant ) -1 operator i,e. almost an inverse time operator, in other words either (E+m)-1 or (E-m) -1 where E can be thought of as id/dt. Thus, the RHS of the Dirac equations is i d/dt U(x,t) or i d/dt V(x,t) which suggests an infinitesimal step forward in time, while the LHS side contains (id/dt +m)-1 or (-id/dt-m) -1 which seems to be representative of undoing such a step. It should be noted that one may take Einstein’s energy-momentum equation E*E = p*p + mo*mo, without any knowledge of quantum mechanics, and cast it as 2x2 matrix equation, linear in E, namely H (U,V) = E (U,V). The matrix H serves as an evolution operator and another matrix serves as a velocity operator. One may derive the Schrodinger zitterbewegung equation for the velocity matrix without any use of wavefunctions and derivatives which are key to quantum mechanics. Nevertheless, the idea of noncommuting matrices, vectors and zitterbewegung is present. There are several choices for the matrix H, but one leads to the one dimensional Dirac coupled equations Thus, the “spin” features may be obtained directly from E*E= p*p+mo*mo without any quantum mechanics. (1), (2). The purpose of this note is to examine some of these ideas.
Abstract We begin the discussion of nonrelativistic quaternionic quantum mechanics by analyzing the allowed form of H for a single-particle system. We will assume, as is done in complex nonrelativistic quantum mechanics, that H is invariant under Galilean transformations and that the kinetic or noninteracting part of H is rotation and translation invariant. In analyzing the implications of these assumptions, we follow the method used in Sec. 13-4 of Jauch (1968a) in the complex quantum mechanics case [see also Mackey (1968), and Feynman as described in Dyson (1990)], and so work in the coordinate representation (cf. Sec. 2.4), and use the Heisenberg picture (cf. Sec. 3.3) to describe time evolution. For brevity of notation, we omit the subscript H for Heisenberg picture operators and denote the coordinate representation Hamiltonian H(x) as simply H2.