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

Physical Review

Quantum-Mechanically Correct Form of Hamiltonian Function for Conservative Systems

Boris Podolsky

Published Nov 1, 1928• 465 citations
Publisher

Abstract

Dirac showed that, if in the Hamiltonian $H$ momenta ${\ensuremath{\eta}}_{r}$ conjugate to the co-ordinates ${\ensuremath{\xi}}_{r}$ are replaced by $(\frac{h}{2\ensuremath{\pi}i})\frac{\ensuremath{\partial}}{\ensuremath{\partial}{\ensuremath{\xi}}_{r}}$, the Schr\"odinger equation appropriate to the coordinate system ${\ensuremath{\xi}}_{r}$ is $(H\ensuremath{-}E){\ensuremath{\psi}}_{\ensuremath{\xi}}=0$. Applied to coordinate systems other than cartesian this usually leads to incorrect results. The difficulty is here traced partially to the way in which ${\ensuremath{\psi}}_{\ensuremath{\xi}}$ is normalized and partly to the choice of $H$. In $H$ expressions such as $\mathrm{qp}{q}^{\ensuremath{-}1}p$ and ${p}^{2}$ are not equivalent, and the simplified form is generally incorrect. A formula satisfying all the requirements of quantum mechanics for a Hamiltonian of a conservative system, in an arbitrary coordinate system, is therefore developed $H=\frac{1}{2\ensuremath{\mu}}\ensuremath{\Sigma}\stackrel{r=n}{r=1}\ensuremath{\Sigma}\stackrel{s=n}{s=1}{g}^{\ensuremath{-}\frac{1}{4}}{p}_{r}{g}^{\frac{1}{2}}{g}^{\mathrm{rs}}{p}_{s}{g}^{\ensuremath{-}\frac{1}{4}}+U$ This formula is applied to a case of plane polar coordinates and leads to correct results.

DOI
10.1103/physrev.32.812

Related papers

Generalized Momentum Operators in Quantum Mechanics

Peter D. Robinson, Joseph O. Hirschfelder

The usual form P0 for the quantum-mechanical operator P conjugate to a generalized coordinate q1 is, in atomic units, P0=−ig−12 ∂/∂q1(g12),where g is the Jacobian of the transformation from Cartesian coordinates to the generalized coordinates. However, in some cases, this plusible form for P is not Hermitian with respect to physically acceptable bound-state wavefunctions, as it must be if it is to represent a real observable quantity. In this paper, the general form P=−i(A·grad+12 div A)is justified. Here A = h1q̂1, where h1 is the metric scale factor corresponding to q1, and q̂1 is the unit vector in the direction of increasing q1. This general form reduces to P0 if the usual formula for the divergence is applied. In the cases where P0 is not Hermitian, it transpires that q̂1 is ill-defined at one or both of the end points α and β of the range of q1, and the divergence formula is thus invalid at such points. It is shown that, in order to obtain a Hermitian form for P, certain terms involving delta functions similar to Dirac's must be added to the usual formula for div A. These terms can be regarded as implicit in div A. If q̂1 is ill-defined at the lower limit q1 = α, then the resulting new Hermitian form for P, which we propose as the correct one, is P=−i[g−12 ∂/∂q1(g12)+12 δ+(q1−α)]. If q̂1 is, in addition, ill-defined at the upper limit q1 = β, then the extra term +½i δ_(β − q1) must be added. Corresponding new forms are obtained for the Laplacian operator. In addition, the new formulas for P are applied to hypervirial relations. In the Appendix, Charles Goebel obtains a similar expression for the momentum operators by replacing the metric scale function by θg where θ is a step function, unity inside and zero outside the range of definition of the generalized coordinates. The differentiation of θ then produces the delta functions.

Journal of Mathematical Physics•Mar 1, 1963•29 citations

Derivation of the coordinate-momentum commutation relations from canonical invariance

Moshe Shapiro

Based on (i) the spectral resolution definition of the momentum operator, (ii) the linearity of correspondence between physical observables and quantum Hermitian operators, (iii) the definition of conjugate coordinate-momentum variables in classical mechanics, and (iv) the invariance of the classical Hamiltonian to canonical transformations (transformations that change from one set of conjugate variables to another), we show that the $[\stackrel{\ifmmode \hat{}\else \^{}\fi{}}{x},\stackrel{\ifmmode \hat{}\else \^{}\fi{}}{p}]$ coordinate-momentum commutator must have the value $ih$ where $h$ is a real nonzero number (which we can identify with the experimentally determined $\ensuremath{\hbar}$). The results are then extended to include all generalized coordinates and their conjugate momenta as well as the Cartesian special-relativistic case.

Physical Review A•Oct 31, 2006•5 citations

THE LAGRANGIAN IN QUANTUM MECHANICS

P. A. M. Dirac

Quantum mechanics was built up on a foundation of analogy with the Hamiltonian theory of classical mechanics. This is because the classical notion of canonical coordinates and momenta was found to be one with a very simple quantum analogue, as a result of which the whole of the classical Hamiltonian theory, which is just a structure built up on this notion, could be taken over in all its details into quantum mechanics. Now there is an alternative formulation for classical dynamics, provided by the Lagrangian. This requires one to work in terms of coordinates and velocities instead of coordinates and momenta. The two formulations are, of course, closely related, but there are reasons for believing that the Lagrangian one is the more fundamental. In the first place the Lagrangian method allows one to collect together all the equations of motion and express them as the stationary property of a certain action function. (This action function is just the time-integral of the Lagrangian.) There is no corresponding action principle in terms of the coordinates and momenta of the Hamiltonian theory. Secondly the Lagrangian method can easily be expressed relativistically, on account of the action function being a relativistic invariant; while the Hamiltonian method is essentially non-relativistic in form, since it marks out a particular time variable as the canonical conjugate of the Hamiltonian function. For these reasons it would seem desirable to take up the question of what corresponds in the quantum theory to the Lagrangian method of the classical theory. A little consideration shows, however, that one cannot expect to be able to take over the classical Lagrangian equations in any very direct way. These equations involve partial derivatives of the Lagrangian with respect to the coordinates and velocities and no meaning can be given to such derivatives in quantum mechanics. The only differentiation process that can be carried out with respect to the dynamical variables of quantum mechanics is that of forming Poisson brackets and this process leads to the Hamiltonian theory. 1 We must therefore seek our quantum Lagrangian theory in an indirect way. We must try to take over the ideas of the classical Lagrangian theory, not the equations of the classical Lagrangian theory.

WORLD SCIENTIFIC eBooks•Aug 1, 2005•288 citations

Hamiltonian Path-Integral Methods

Claude Garrod

A path-integral formulation of quantum mechanics is investigated which is closely related to that of Feynman. It differs from Feynman's formulation in that it involves the Hamiltonian function of the canonically conjugate coordinates and momenta. The classical limit yields the variational principle: $\ensuremath{\delta}ff(p\ifmmode\ddagger\else\textdaggerdbl\fi{}\ensuremath{-}N)dt=0$. A path-integral formula is also obtained for the energy eigenstate projection operator associated with the time-independent Schr\"odinger equation. The classical limit of the projection operator formula yields a modified form of the well-known variational principle for the phase-space orbit of given energy. Relativistically covariant Hamiltonian variational principles are analyzed and lead naturally to a relativistic scalar wave equation which involves a proper time variable which is canonically conjugate to the mass in the same manner as the ordinary time variable is conjugate to the energy in nonrelativistic quantum theory.

Reviews of Modern Physics•Jul 1, 1966•161 citations

Momentum, quasi-momentum and hamiltonian operators in terms of arbitrary curvilinear coordinates, with special emphasis on molecular hamiltonians

André Nauts, Xavier Chapuisat

Since the conformation of physical systems is often advantageously described with the help of generalized (i.e. curvilinear) coordinates, the following questions are raised and general answers given to them: (i) What are the components of the momentum operators, those of the adjoint momentum operators and those of the hermitian momentum operators? (ii) What is the general form of the hamiltonian operator? (iii) What do the general forms of the momentum and hamiltonian operators become when expressed in terms of quasi-momentum operators (e.g. angular momentum component operators)? (iv) How are the answers to the above questions affected by an arbitrary choice of normalization convention for the total wave-function? A complete set of formulae (whatever the independent choices of coordinates, quasi-momentum operators and normalization) is given and various ‘historical’ formulae are shown to be particular instances of the general formulae proposed.

Molecular Physics•Aug 20, 1985•130 citations

Hamiltonian path integral quantization in polar coordinates

Ashok Kumar Kapoor, Pankaj Sharan

Using a scheme proposed earlier we set up Hamiltonian path integral quantization for a particle in two dimensions in plane polar coordinates.This scheme uses the classical Hamiltonian, without any $O(\hbar^2)$ terms, in the polar varivables. We show that the propagator satisfies the correct Schrödinger equation.

arXiv:quant-ph/9804037•Apr 15, 1998•3 citations
Dirac showed that, if in the Hamiltonian HHH momenta \ensuremathηr{\ensuremath{\eta}}_{r}\ensuremathηr​ conjugate to the co-ordinates \ensuremathξr{\ensuremath{\xi}}_{r}\ensuremathξr​ are replaced by (h2\ensuremathπi)\ensuremath∂\ensuremath∂\ensuremathξr(\frac{h}{2\ensuremath{\pi}i})\frac{\ensuremath{\partial}}{\ensuremath{\partial}{\ensuremath{\xi}}_{r}}(2\ensuremathπih​)\ensuremath∂\ensuremathξr​\ensuremath∂​, the Schr"odinger equation appropriate to the coordinate system \ensuremathξr{\ensuremath{\xi}}_{r}\ensuremathξr​ is (H\ensuremath−E)\ensuremathψ\ensuremathξ=0(H\ensuremath{-}E){\ensuremath{\psi}}_{\ensuremath{\xi}}=0(H\ensuremath−E)\ensuremathψ\ensuremathξ​=0. Applied to coordinate systems other than cartesian this usually leads to incorrect results. The difficulty is here traced partially to the way in which \ensuremathψ\ensuremathξ{\ensuremath{\psi}}_{\ensuremath{\xi}}\ensuremathψ\ensuremathξ​ is normalized and partly to the choice of HHH. In HHH expressions such as qpq\ensuremath−1p\mathrm{qp}{q}^{\ensuremath{-}1}pqpq\ensuremath−1p and p2{p}^{2}p2 are not equivalent, and the simplified form is generally incorrect. A formula satisfying all the requirements of quantum mechanics for a Hamiltonian of a conservative system, in an arbitrary coordinate system, is therefore developed H=12\ensuremathμ\ensuremathΣr=1r=n\ensuremathΣs=1s=ng\ensuremath−14prg12grspsg\ensuremath−14+UH=\frac{1}{2\ensuremath{\mu}}\ensuremath{\Sigma}\stackrel{r=n}{r=1}\ensuremath{\Sigma}\stackrel{s=n}{s=1}{g}^{\ensuremath{-}\frac{1}{4}}{p}_{r}{g}^{\frac{1}{2}}{g}^{\mathrm{rs}}{p}_{s}{g}^{\ensuremath{-}\frac{1}{4}}+UH=2\ensuremathμ1​\ensuremathΣr=1r=n\ensuremathΣs=1s=ng\ensuremath−41​pr​g21​grsps​g\ensuremath−41​+U This formula is applied to a case of plane polar coordinates and leads to correct results.
The usual form P0 for the quantum-mechanical operator P conjugate to a generalized coordinate q1 is, in atomic units, P0=−ig−12 ∂/∂q1(g12),where g is the Jacobian of the transformation from Cartesian coordinates to the generalized coordinates. However, in some cases, this plusible form for P is not Hermitian with respect to physically acceptable bound-state wavefunctions, as it must be if it is to represent a real observable quantity. In this paper, the general form P=−i(A·grad+12 div A)is justified. Here A = h1q̂1, where h1 is the metric scale factor corresponding to q1, and q̂1 is the unit vector in the direction of increasing q1. This general form reduces to P0 if the usual formula for the divergence is applied. In the cases where P0 is not Hermitian, it transpires that q̂1 is ill-defined at one or both of the end points α and β of the range of q1, and the divergence formula is thus invalid at such points. It is shown that, in order to obtain a Hermitian form for P, certain terms involving delta functions similar to Dirac's must be added to the usual formula for div A. These terms can be regarded as implicit in div A. If q̂1 is ill-defined at the lower limit q1 = α, then the resulting new Hermitian form for P, which we propose as the correct one, is P=−i[g−12 ∂/∂q1(g12)+12 δ+(q1−α)]. If q̂1 is, in addition, ill-defined at the upper limit q1 = β, then the extra term +½i δ_(β − q1) must be added. Corresponding new forms are obtained for the Laplacian operator. In addition, the new formulas for P are applied to hypervirial relations. In the Appendix, Charles Goebel obtains a similar expression for the momentum operators by replacing the metric scale function by θg where θ is a step function, unity inside and zero outside the range of definition of the generalized coordinates. The differentiation of θ then produces the delta functions.
Based on (i) the spectral resolution definition of the momentum operator, (ii) the linearity of correspondence between physical observables and quantum Hermitian operators, (iii) the definition of conjugate coordinate-momentum variables in classical mechanics, and (iv) the invariance of the classical Hamiltonian to canonical transformations (transformations that change from one set of conjugate variables to another), we show that the $[\stackrel{\ifmmode \hat{}\else \^{}\fi{}}{x},\stackrel{\ifmmode \hat{}\else \^{}\fi{}}{p}]$ coordinate-momentum commutator must have the value ihihih where hhh is a real nonzero number (which we can identify with the experimentally determined $\ensuremath{\hbar}$). The results are then extended to include all generalized coordinates and their conjugate momenta as well as the Cartesian special-relativistic case.
Quantum mechanics was built up on a foundation of analogy with the Hamiltonian theory of classical mechanics. This is because the classical notion of canonical coordinates and momenta was found to be one with a very simple quantum analogue, as a result of which the whole of the classical Hamiltonian theory, which is just a structure built up on this notion, could be taken over in all its details into quantum mechanics. Now there is an alternative formulation for classical dynamics, provided by the Lagrangian. This requires one to work in terms of coordinates and velocities instead of coordinates and momenta. The two formulations are, of course, closely related, but there are reasons for believing that the Lagrangian one is the more fundamental. In the first place the Lagrangian method allows one to collect together all the equations of motion and express them as the stationary property of a certain action function. (This action function is just the time-integral of the Lagrangian.) There is no corresponding action principle in terms of the coordinates and momenta of the Hamiltonian theory. Secondly the Lagrangian method can easily be expressed relativistically, on account of the action function being a relativistic invariant; while the Hamiltonian method is essentially non-relativistic in form, since it marks out a particular time variable as the canonical conjugate of the Hamiltonian function. For these reasons it would seem desirable to take up the question of what corresponds in the quantum theory to the Lagrangian method of the classical theory. A little consideration shows, however, that one cannot expect to be able to take over the classical Lagrangian equations in any very direct way. These equations involve partial derivatives of the Lagrangian with respect to the coordinates and velocities and no meaning can be given to such derivatives in quantum mechanics. The only differentiation process that can be carried out with respect to the dynamical variables of quantum mechanics is that of forming Poisson brackets and this process leads to the Hamiltonian theory. 1 We must therefore seek our quantum Lagrangian theory in an indirect way. We must try to take over the ideas of the classical Lagrangian theory, not the equations of the classical Lagrangian theory.
A path-integral formulation of quantum mechanics is investigated which is closely related to that of Feynman. It differs from Feynman's formulation in that it involves the Hamiltonian function of the canonically conjugate coordinates and momenta. The classical limit yields the variational principle: $\ensuremath{\delta}ff(p\ifmmode\ddagger\else\textdaggerdbl\fi{}\ensuremath{-}N)dt=0$. A path-integral formula is also obtained for the energy eigenstate projection operator associated with the time-independent Schr\"odinger equation. The classical limit of the projection operator formula yields a modified form of the well-known variational principle for the phase-space orbit of given energy. Relativistically covariant Hamiltonian variational principles are analyzed and lead naturally to a relativistic scalar wave equation which involves a proper time variable which is canonically conjugate to the mass in the same manner as the ordinary time variable is conjugate to the energy in nonrelativistic quantum theory.
Since the conformation of physical systems is often advantageously described with the help of generalized (i.e. curvilinear) coordinates, the following questions are raised and general answers given to them: (i) What are the components of the momentum operators, those of the adjoint momentum operators and those of the hermitian momentum operators? (ii) What is the general form of the hamiltonian operator? (iii) What do the general forms of the momentum and hamiltonian operators become when expressed in terms of quasi-momentum operators (e.g. angular momentum component operators)? (iv) How are the answers to the above questions affected by an arbitrary choice of normalization convention for the total wave-function? A complete set of formulae (whatever the independent choices of coordinates, quasi-momentum operators and normalization) is given and various ‘historical’ formulae are shown to be particular instances of the general formulae proposed.
Using a scheme proposed earlier we set up Hamiltonian path integral quantization for a particle in two dimensions in plane polar coordinates.This scheme uses the classical Hamiltonian, without any O(ℏ2)O(\hbar^2)O(ℏ2) terms, in the polar varivables. We show that the propagator satisfies the correct Schrödinger equation.