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 A

Unitary and Hermitian phase operators for the electromagnetic field

Yee Kin Tsui, Michael F. Reid

Published Jul 1, 1992• 11 citations
Publisher

Abstract

Pegg and Barnett [Phys. Rev. A 39, 1665 (1989)] have proposed a formalism for a Hermitian optical phase operator, constructed from the so-called phase states. In this paper we give an alternative derivation. We begin by defining a unitary phase operator ${e}^{i\stackrel{^}{\ensuremath{\varphi}}}$ from the relationship $\stackrel{^}{a}={e}^{i\stackrel{^}{\ensuremath{\varphi}}}R(\stackrel{^}{N})$ ($\stackrel{^}{a}$ is an annihilation operator for a single-mode boson field), but borrow the idea of a finite- (but arbitrarily large) dimensional space from Pegg and Barnett. Finally, we comment on an experiment to determine quantum phase uncertainty. Our analysis uses the operator ${e}^{i\stackrel{^}{\ensuremath{\varphi}}}$, which we feel is a more natural choice than $\stackrel{^}{\ensuremath{\varphi}}$. The use of ${e}^{i\stackrel{^}{\ensuremath{\varphi}}}$ also avoids the multivalued problems associated with $\stackrel{^}{\ensuremath{\varphi}}$.

DOI
10.1103/physreva.46.549

Related papers

Hermitian Phase Operator of Quantized Light Fields. A Fundamental Problem of Quantum Optics.

Hiroshi Iwasawa

For a long time, about 60 years, the phase as an observable of the quantized light fields has r emained a fundamental problem with considerable difficulties on quantum mechanics and also q uantum optics.Recently, however, the hermitian phase operator and its eigenstates in a finite H ilbert space are found by Pegg and Barnett.In this paper, we provide a historical review and t he background of the problem on the phase operator, explore the Pegg-Barnett formalism, and s how the recent developements of this fields and its application to various light fields.Further problems on this formalism and other approaches are also discussed.

The Review of Laser Engineering•Jan 1, 1993•0 citations

Phase operators on Hilbert space

John A. Vaccaro

A simple formalism is introduced for describing the quantum phase of a single-mode field (or, equivalently, a harmonic oscillator) using a particular infinite-dimensional Hilbert space ${\mathit{H}}_{\mathit{s}\mathit{y}\mathit{m}}$. This space is spanned by vectors which represent states of the single field mode containing non-negative numbers of photons. It has a symmetry in the sense that it contains vectors in the neighborhood of both the vacuum state and the infinite photon number state. A striking property of ${\mathit{H}}_{\mathit{s}\mathit{y}\mathit{m}}$ is that it supports Hermitian and unitary phase operators. Indeed the Pegg-Barnett unitary phase operators are shown to converge strongly on the Hilbert space ${\mathit{H}}_{\mathit{s}\mathit{y}\mathit{m}}$ and so the correponding limit points are ordinary operators on ${\mathit{H}}_{\mathit{s}\mathit{y}\mathit{m}}$. This can be compared with the situation for the infinite-dimensional Hilbert space H conventionally used in quantum optics on which the Pegg-Barnett unitary phase operators converge weakly, in general.

Physical Review A•Apr 1, 1995•25 citations

The Phase Operator

Stephen M. Barnett, John A. Vaccaro

More particularly, the sets {|n〉 : n = 0, 1, · · · s} and {|θm〉 : m = 0, 1, · · · s} are two mutually unbiased bases for s as |〈n|θm〉|2 = 1/(s + 1) for n, m = 0, 1, · · · , s. The exponential phase operator is defined in terms of the phase state basis as ei φˆθ = s∑ m=0 eiθm |θm〉〈θm|, (2.4) or equivalently, in terms of the energy eigenstates as |n〉〈n + 1| + ei(s+1)θ0 |s〉〈0|. (2.5) The subscript θ serves as a reminder that the values of phase variable θm lie in the range from θ0 to θ0 + 2π . This construction ensures that the phase states are the eigenstates of ei φˆθ , i.e., ei φˆθ |θm〉 = eiθm |θm〉. (2.6) The exponential phase operator is unitary, because n=0 |n〉〈n| = 1ˆ s (2.7) where 1ˆ s is the identity operator on s . The Hermitian phase operator representing the phase angle itself was introduced in (Pegg and Barnett 1988b Paper 2.2) and (Barnett and Pegg 1989 Paper 2.3) as simply φˆθ = s∑ m=0 θm|θm〉〈θm|. (2.8) The operator on the left hand sides of Eqs. (2.4) and (2.5) is the exponential of i φˆθ .Moreover, the operator corresponding to an arbitrary function of the phase angle f (θ) was introduced in Barnett and Pegg (1992 Paper 3.3) and also discussed in Pegg and Barnett (1997a Paper 2.6) as f (φˆθ ) = s∑ m=0 f (θm)|θm〉〈θm|. (2.9) It follows that the Hermitian conjugate of ei φˆθ is the operator (ei φˆθ )† = s∑ m=0 e−iθm |θm〉〈θm| = e−i φˆθ . (2.10) This construction on the finite dimensional spaces elegantly sidesteps the problems faced with the infinite dimensional Hilbert space discussed in the previous chapter. Its simplicity is particularly appealing. The question remains, however, of how to deal with states such as | f 〉 = ∑∞n=0 cn|n〉 which reside in the infinite dimensional Hilbert space. For each value of s, there exists a phase operator defined on s ; this leads naturally to sequences of operators, such as e.g. φˆθ |s=S, φˆθ |s=S+1, φˆθ |s=S+2 · · · which are phase angle operators defined on the sequence of spaces S, S+1, S+2, · · · for arbitrarily large integer S. The space s contains a state | f 〉s which approximates | f 〉 to arbitrary precision for a sufficiently large value of s, such that the modulus of the overlap 〈 f | f 〉s is as close to unity as desired. Thus the issue of including states such as | f 〉 entails dealing with such sequences as s tends to infinity.

CRC Press eBooks•Apr 27, 2007•0 citations

Phase operator for the photon field and an index theorem

Kazuo Fujikawa

An index relation dim ker ${\mathit{a}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$a-dim ker ${\mathrm{aa}}^{\ifmmode^\circ\else\textdegree\fi{}}$=1 is satisfied by the creation and annihilation operators ${\mathit{a}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$ and a of a harmonic oscillator, where dim ker ${\mathit{a}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$a, for example, stands for the number of normalizable basis vectors ${\mathit{u}}_{\mathit{n}}$ that satisfy ${\mathit{a}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$${\mathit{au}}_{\mathit{n}}$=0. A Hermitian phase operator, which inevitably leads to dim ker ${\mathit{a}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$a-dim ker ${\mathrm{aa}}^{\ifmmode^\circ\else\textdegree\fi{}}$=0, cannot be consistently defined. If one considers an (s+1)-dimensional truncated theory, a Hermitian phase operator of Pegg and Barnett [Europhys. Lett. 6, 483 (1988)] which carries a vanishing index can be defined. However, for arbitrarily large s, we show that the vanishing index of the Hermitian phase operator of Pegg and Barnett causes a substantial deviation from minimum uncertainty in a characteristically quantum domain with small average photon numbers. We also mention an interesting analogy between the present problem and the chiral anomaly in gauge theory which is related to the Atiyah-Singer index theorem. It is suggested that the phase operator problem related to the above analytic index may be regarded as an alternative class of quantum anomaly. From an anomaly viewpoint, it is not surprising that the phase operator of Susskind and Glogower [Physics 1, 49 (1964)], which carries a unit index, leads to an anomalous identity and an anomalous commutator.

arXiv:hep-th/9411066•Physical Review A•Oct 1, 1995•33 citations

Fourier theory, uncertainty relations, and quantum phase

Jeffrey H. Shapiro, Scott Roger Shepard, N. C. Wong

The problem of phase in quantum mechanics is an old one but one that has received increasing attention of late. For a single-mode radiation field with annihilation operator â there have been two recent formulations for its quantum phase, Pegg and Bamett1 have created a Hermitian phase operator ϕ ^ s by truncating the state space associated with â to the subspace spanned by the number kets {|n〉:0 ≤ n ≤ S}.

•May 21, 1990•1 citation

Comment on ‘‘Phase properties of the quantized single-mode electromagnetic field’’

Xin Ma, William Rhodes

We show that the approach by Pegg and Barnett [Phys. Rev. A 39, 1665 (1989)] to construct a phase operator through a limiting procedure does not give a satisfactory description of the phase properties of the quantized electromagnetic field. We refute the claim that their approach predicts the same results of physical experiments on phase properties as the conventional Hilbert space approach. Their procedure could only be used as a definition of the phase, but it is not a good definition because the ``phase'' states they used to construct their ``phase'' operator are not of precisely defined phase contrary to their assertion.

Physical Review A•Mar 1, 1991•11 citations
Pegg and Barnett [Phys. Rev. A 39, 1665 (1989)] have proposed a formalism for a Hermitian optical phase operator, constructed from the so-called phase states. In this paper we give an alternative derivation. We begin by defining a unitary phase operator {e}^{i\stackrel{^}{\ensuremath{\varphi}}} from the relationship \stackrel{^}{a}={e}^{i\stackrel{^}{\ensuremath{\varphi}}}R(\stackrel{^}{N}) (\stackrel{^}{a} is an annihilation operator for a single-mode boson field), but borrow the idea of a finite- (but arbitrarily large) dimensional space from Pegg and Barnett. Finally, we comment on an experiment to determine quantum phase uncertainty. Our analysis uses the operator {e}^{i\stackrel{^}{\ensuremath{\varphi}}}, which we feel is a more natural choice than \stackrel{^}{\ensuremath{\varphi}}. The use of {e}^{i\stackrel{^}{\ensuremath{\varphi}}} also avoids the multivalued problems associated with \stackrel{^}{\ensuremath{\varphi}}.
For a long time, about 60 years, the phase as an observable of the quantized light fields has r emained a fundamental problem with considerable difficulties on quantum mechanics and also q uantum optics.Recently, however, the hermitian phase operator and its eigenstates in a finite H ilbert space are found by Pegg and Barnett.In this paper, we provide a historical review and t he background of the problem on the phase operator, explore the Pegg-Barnett formalism, and s how the recent developements of this fields and its application to various light fields.Further problems on this formalism and other approaches are also discussed.
A simple formalism is introduced for describing the quantum phase of a single-mode field (or, equivalently, a harmonic oscillator) using a particular infinite-dimensional Hilbert space Hsym{\mathit{H}}_{\mathit{s}\mathit{y}\mathit{m}}Hsym​. This space is spanned by vectors which represent states of the single field mode containing non-negative numbers of photons. It has a symmetry in the sense that it contains vectors in the neighborhood of both the vacuum state and the infinite photon number state. A striking property of Hsym{\mathit{H}}_{\mathit{s}\mathit{y}\mathit{m}}Hsym​ is that it supports Hermitian and unitary phase operators. Indeed the Pegg-Barnett unitary phase operators are shown to converge strongly on the Hilbert space Hsym{\mathit{H}}_{\mathit{s}\mathit{y}\mathit{m}}Hsym​ and so the correponding limit points are ordinary operators on Hsym{\mathit{H}}_{\mathit{s}\mathit{y}\mathit{m}}Hsym​. This can be compared with the situation for the infinite-dimensional Hilbert space H conventionally used in quantum optics on which the Pegg-Barnett unitary phase operators converge weakly, in general.
More particularly, the sets {|n〉 : n = 0, 1, · · · s} and {|θm〉 : m = 0, 1, · · · s} are two mutually unbiased bases for s as |〈n|θm〉|2 = 1/(s + 1) for n, m = 0, 1, · · · , s. The exponential phase operator is defined in terms of the phase state basis as ei φˆθ = s∑ m=0 eiθm |θm〉〈θm|, (2.4) or equivalently, in terms of the energy eigenstates as |n〉〈n + 1| + ei(s+1)θ0 |s〉〈0|. (2.5) The subscript θ serves as a reminder that the values of phase variable θm lie in the range from θ0 to θ0 + 2π . This construction ensures that the phase states are the eigenstates of ei φˆθ , i.e., ei φˆθ |θm〉 = eiθm |θm〉. (2.6) The exponential phase operator is unitary, because n=0 |n〉〈n| = 1ˆ s (2.7) where 1ˆ s is the identity operator on s . The Hermitian phase operator representing the phase angle itself was introduced in (Pegg and Barnett 1988b Paper 2.2) and (Barnett and Pegg 1989 Paper 2.3) as simply φˆθ = s∑ m=0 θm|θm〉〈θm|. (2.8) The operator on the left hand sides of Eqs. (2.4) and (2.5) is the exponential of i φˆθ .Moreover, the operator corresponding to an arbitrary function of the phase angle f (θ) was introduced in Barnett and Pegg (1992 Paper 3.3) and also discussed in Pegg and Barnett (1997a Paper 2.6) as f (φˆθ ) = s∑ m=0 f (θm)|θm〉〈θm|. (2.9) It follows that the Hermitian conjugate of ei φˆθ is the operator (ei φˆθ )† = s∑ m=0 e−iθm |θm〉〈θm| = e−i φˆθ . (2.10) This construction on the finite dimensional spaces elegantly sidesteps the problems faced with the infinite dimensional Hilbert space discussed in the previous chapter. Its simplicity is particularly appealing. The question remains, however, of how to deal with states such as | f 〉 = ∑∞n=0 cn|n〉 which reside in the infinite dimensional Hilbert space. For each value of s, there exists a phase operator defined on s ; this leads naturally to sequences of operators, such as e.g. φˆθ |s=S, φˆθ |s=S+1, φˆθ |s=S+2 · · · which are phase angle operators defined on the sequence of spaces S, S+1, S+2, · · · for arbitrarily large integer S. The space s contains a state | f 〉s which approximates | f 〉 to arbitrary precision for a sufficiently large value of s, such that the modulus of the overlap 〈 f | f 〉s is as close to unity as desired. Thus the issue of including states such as | f 〉 entails dealing with such sequences as s tends to infinity.
An index relation dim ker ${\mathit{a}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$a-dim ker ${\mathrm{aa}}^{\ifmmode^\circ\else\textdegree\fi{}}$=1 is satisfied by the creation and annihilation operators ${\mathit{a}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$ and a of a harmonic oscillator, where dim ker ${\mathit{a}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$a, for example, stands for the number of normalizable basis vectors un{\mathit{u}}_{\mathit{n}}un​ that satisfy ${\mathit{a}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$aun{\mathit{au}}_{\mathit{n}}aun​=0. A Hermitian phase operator, which inevitably leads to dim ker ${\mathit{a}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$a-dim ker ${\mathrm{aa}}^{\ifmmode^\circ\else\textdegree\fi{}}$=0, cannot be consistently defined. If one considers an (s+1)-dimensional truncated theory, a Hermitian phase operator of Pegg and Barnett [Europhys. Lett. 6, 483 (1988)] which carries a vanishing index can be defined. However, for arbitrarily large s, we show that the vanishing index of the Hermitian phase operator of Pegg and Barnett causes a substantial deviation from minimum uncertainty in a characteristically quantum domain with small average photon numbers. We also mention an interesting analogy between the present problem and the chiral anomaly in gauge theory which is related to the Atiyah-Singer index theorem. It is suggested that the phase operator problem related to the above analytic index may be regarded as an alternative class of quantum anomaly. From an anomaly viewpoint, it is not surprising that the phase operator of Susskind and Glogower [Physics 1, 49 (1964)], which carries a unit index, leads to an anomalous identity and an anomalous commutator.
We show that the approach by Pegg and Barnett [Phys. Rev. A 39, 1665 (1989)] to construct a phase operator through a limiting procedure does not give a satisfactory description of the phase properties of the quantized electromagnetic field. We refute the claim that their approach predicts the same results of physical experiments on phase properties as the conventional Hilbert space approach. Their procedure could only be used as a definition of the phase, but it is not a good definition because the ``phase'' states they used to construct their ``phase'' operator are not of precisely defined phase contrary to their assertion.