Classical Limit of the Moyal Bracket

The Moyal bracket is the phase-space representation of the quantum commutator. For Weyl symbols $A_W$ and $B_W$,

$$\left( \frac{1}{i\hbar} [A,B] \right)_W = \{A_W,B_W\}_M.$$

Its leading derivative term is the classical Poisson bracket:

$$\{A,B\}_M = \{A,B\}_{\mathrm{PB}} + O(\hbar^2).$$

This compact equation is easy to overread. The correction is small only when $A$ and $B$ are smooth on phase-space scales whose product is large compared with $\hbar$. Quantum evolution can create progressively finer structure, so an approximation valid initially can fail later.

Moyal Bracket owns the exact bracket and its algebra. Phase-Space Dynamics owns the full Wigner evolution and standard worked models. This page owns the asymptotic comparison: the small parameter, correction estimates, exact quadratic exception, and failure modes.

Moyal Bracket Expansion

For one canonical pair $(x,p)$, define the bidifferential operator $\Lambda$ by

$$A\Lambda B = \frac{\partial A}{\partial x} \frac{\partial B}{\partial p} - \frac{\partial A}{\partial p} \frac{\partial B}{\partial x}.$$

The star product can be written

$$A\star B = A \exp\left( \frac{i\hbar}{2}\Lambda \right) B,$$

and the Moyal bracket is

$$\{A,B\}_M = \frac{1}{i\hbar} \left( A\star B-B\star A \right).$$

Only odd powers of $\Lambda$ survive the antisymmetrization:

$$\{A,B\}_M = \frac{2}{\hbar} A \sin\left( \frac{\hbar}{2}\Lambda \right) B.$$

Expanding the sine gives

$$\begin{aligned} \{A,B\}_M &= A\Lambda B - \frac{\hbar^2}{24} A\Lambda^3B \\ &\quad+ \frac{\hbar^4}{1920} A\Lambda^5B - \cdots. \end{aligned}$$

Equivalently,

$$\{A,B\}_M = \sum_{n=0}^{\infty} \frac{(-1)^n}{(2n+1)!} \left( \frac{\hbar}{2} \right)^{2n} A\Lambda^{2n+1}B.$$

The cubic bidifferential term is

$$\begin{aligned} A\Lambda^3B &= A_{xxx}B_{ppp} - 3A_{xxp}B_{ppx} \\ &\quad+ 3A_{xpp}B_{pxx} - A_{ppp}B_{xxx}. \end{aligned}$$

Subscripts denote partial derivatives. This formula shows why a power of $\hbar$ is not enough to estimate a term: each correction also contains higher derivatives.

The derivative series may be formal or asymptotic rather than globally convergent. Its useful regime is determined by symbol regularity and phase-space scales. When fine structure is important, the exact star-product or operator formulation can be safer than truncating the series.

Leading Poisson Bracket Term

The first term is

$$A\Lambda B = \{A,B\}_{\mathrm{PB}}.$$

For Wigner evolution,

$$\frac{\partial W}{\partial t} = \{H_W,W\}_M,$$

so the leading approximation is the classical Liouville equation

$$\frac{\partial W_{\mathrm{cl}}}{\partial t} = \{H_{\mathrm{cl}},W_{\mathrm{cl}}\}_{\mathrm{PB}}.$$

The notation $W_{\mathrm{cl}}$ must be interpreted carefully. A quantum Wigner function can be negative, and evolving it with the Poisson bracket does not turn it into a classical probability density. The approximation concerns the generator of phase-space transport.

Suppose the relevant symbols vary on characteristic coordinate and momentum scales $L_x$ and $L_p$. Schematically,

$$\partial_x \sim \frac{1}{L_x}, \qquad \partial_p \sim \frac{1}{L_p}.$$

Then

$$A\Lambda B \sim \frac{AB}{L_xL_p},$$

while

$$\hbar^2A\Lambda^3B \sim \frac{\hbar^2AB} {L_x^3L_p^3}.$$

The ratio of the first correction to the Poisson term is therefore of order

$$\epsilon_M^2, \qquad \epsilon_M = \frac{\hbar}{L_xL_p}.$$

The classical approximation requires

$$\epsilon_M\ll1,$$

not the dimensionful statement “$\hbar$ is small.” The phase-space area $L_xL_p$ must be large compared with the quantum action scale $\hbar$.

For several degrees of freedom, the same idea applies to every resolved canonical direction. Anisotropic problems can be semiclassical along some phase-space directions and strongly quantum along others.

Quantum Corrections

For

$$H_W(x,p) = \frac{p^2}{2m} + V(x),$$

the Wigner–Moyal equation is

$$\begin{aligned} \frac{\partial W}{\partial t} &= - \frac{p}{m} \frac{\partial W}{\partial x} + V'(x) \frac{\partial W}{\partial p} \\ &\quad- \frac{\hbar^2}{24} V'''(x) \frac{\partial^3W}{\partial p^3} + O(\hbar^4). \end{aligned}$$

The first row is the classical Liouville generator. Away from zeros of the leading force term, a local diagnostic is

$$R_2(x,p) = \frac{\hbar^2}{24} \left| \frac{V'''(x)}{V'(x)} \frac{\partial_p^3W}{\partial_pW} \right|.$$

The leading truncation is plausible where $R_2\ll1$ and where later terms are also smaller. Near zeros of $V'$ or $\partial_pW$, this pointwise ratio is not useful; norm estimates or direct comparison of the complete generators should be used instead.

Define local scales schematically by

$$\frac{1}{L_V^2} \sim \left| \frac{V'''}{V'} \right|, \qquad \frac{1}{P_W^2} \sim \left| \frac{\partial_p^3W} {\partial_pW} \right|.$$

Then

$$R_2 \sim \frac{1}{24} \left( \frac{\hbar}{L_VP_W} \right)^2.$$

This estimate separates two sources of quantum corrections:

• the Hamiltonian is nonlinear on a short coordinate scale $L_V$;
• the Wigner function has fine momentum structure on a short scale $P_W$.

Either can spoil the approximation.

For the quartic potential

$$V(x) = \frac{\lambda}{4}x^4,$$

one has

$$V'(x)=\lambda x^3, \qquad V'''(x)=6\lambda x.$$

The equation terminates after the first correction:

$$\begin{aligned} \frac{\partial W}{\partial t} &= - \frac{p}{m} \frac{\partial W}{\partial x} + \lambda x^3 \frac{\partial W}{\partial p} \\ &\quad- \frac{\hbar^2\lambda x}{4} \frac{\partial^3W}{\partial p^3}. \end{aligned}$$

Although the series terminates, the correction need not be small. Its size still depends on the momentum derivatives of the evolving Wigner function.

Quadratic Hamiltonians as Special Cases

Let the phase-space vector be

$$z = \begin{pmatrix} x\\ p \end{pmatrix},$$

and consider a general time-dependent quadratic Weyl symbol

$$H_W(z,t) = \frac12 z^{\mathsf T}G(t)z + g(t)^{\mathsf T}z + c(t).$$

Every derivative of $H_W$ of order three or higher vanishes. Since each correction $H_W\Lambda^{2n+1}W$ with $n\geq1$ differentiates $H_W$ at least three times,

$$\{H_W,W\}_M = \{H_W,W\}_{\mathrm{PB}}$$

exactly, for every Wigner function for which the operations are defined.

The resulting flow is affine symplectic. If $\Phi_{t,t_0}$ is the associated classical phase-space map, then

$$W(z,t) = W\left( \Phi_{t,t_0}^{-1}z, t_0 \right).$$

This includes free-particle shear, harmonic-oscillator rotation, squeezing, and linear driving. It explains why Gaussian states remain Gaussian under quadratic dynamics.

The exact equality of generators does not make the state classical:

• a negative Wigner function remains a signed quasiprobability under the invertible flow;
• uncertainty and purity constraints remain quantum;
• measurement statistics still follow the Born rule;
• noncommuting operator products still require the star product.

Quadratic dynamics is an exact classical-looking transport law for a quantum representation, not a conversion of quantum states into classical ensembles.

When the Classical Approximation Fails

Fine interference structure

A superposition of wave packets separated by $\Delta x$ can produce momentum fringes with scale

$$\delta p \sim \frac{\hbar}{\Delta x}.$$

Then

$$\partial_p^nW \sim \left( \frac{\Delta x}{\hbar} \right)^n W$$

within the oscillatory region. The explicit powers of $\hbar$ in the Moyal series are offset by derivatives that grow as inverse powers of $\hbar$. Small $\hbar$ alone does not suppress the correction.

Singular or rapidly varying Hamiltonians

Hard walls, discontinuities, singular potentials, and sharply varying fields may not admit the smooth derivative expansion used above. One should return to the exact star product, integral kernel, matching conditions, or operator evolution rather than differentiate a nonsmooth symbol formally.

Tunneling and nonanalytic effects

Tunneling amplitudes can scale as

$$e^{-S/\hbar}.$$

Such terms are smaller than every power of $\hbar$ but are not captured by any finite truncation of a power series in $\hbar$. A perturbative Moyal expansion and a nonperturbative semiclassical analysis answer different questions.

Long-time evolution

Even if $\epsilon_M$ is initially small, classical flow can stretch and fold a phase-space distribution until its gradients become large. In a chaotic region with Lyapunov scale $\lambda$, a typical resolved length can shrink roughly as $e^{-\lambda t}$. A corresponding correspondence time has the logarithmic form

$$t_E \sim \frac{1}{\lambda} \ln\left( \frac{S_{\mathrm{cl}}}{\hbar} \right),$$

up to system- and observable-dependent constants. This Ehrenfest-time estimate warns that a fixed small $\hbar/S_{\mathrm{cl}}$ does not guarantee uniform accuracy for arbitrarily long times.

Quantum Chaos Preview compares this phase-space breakdown scale with spectral, periodic-orbit, transport, and OTOC diagnostics.

Classicality is more than bracket correspondence

Replacing the Moyal bracket by the Poisson bracket does not by itself explain definite measurement outcomes, suppress interference, or produce a positive distribution. Coarse graining can hide fine Wigner oscillations, and environmental decoherence can suppress selected coherences in a reduced state, but those are additional physical operations. They are not algebraically identical to taking the leading Moyal term. Decoherence as a Classical-Limit Bridge compares the three mechanisms directly.

Practical Validity Checklist

Before truncating the Moyal expansion, check:

1. Which phase-space scales of $H_W$ and $W$ enter the observable?
2. Is every relevant action product $L_xL_p$ large compared with $\hbar$?
3. Are the symbols smooth enough for the required derivatives?
4. Is the leading Poisson term nonzero in the region used for a relative estimate?
5. Does evolution generate shorter scales before the final time?
6. Are nonperturbative effects such as tunneling relevant?
7. Is the claim about transport, observables, or full state reconstruction?
8. Can the truncated result be checked against exact quantum evolution or a converged numerical calculation?

The Semiclassical Limit page supplies the broader action-scale viewpoint. Stationary Phase and the Classical Limit gives the complementary path-integral mechanism.

Common Mistakes

• Writing $\hbar\to0$ without defining a dimensionless action ratio.
• Dropping Moyal corrections because their coefficients contain $\hbar^2$ while ignoring large derivatives of $W$.
• Treating a pointwise correction ratio as meaningful where its leading denominator vanishes.
• Assuming the Moyal derivative series converges for every symbol.
• Generalizing the exact quadratic result to weakly anharmonic dynamics without an error estimate.
• Concluding that Poisson transport makes a negative Wigner function a probability density.
• Expecting a finite power series in $\hbar$ to reproduce tunneling exponentials.
• Claiming correspondence uniformly for arbitrarily long chaotic evolution.
• Confusing the Moyal classical limit with decoherence, coarse graining, or measurement.

Cross-Links

• Why Phase Space in Quantum Mechanics?
• Wigner Function
• Star Product
• Moyal Bracket
• Phase-Space Dynamics
• Gaussian States and Wigner Functions
• Poisson Brackets
• Semiclassical Limit Overview
• Semiclassical Limit
• Stationary Phase and the Classical Limit
• Quantum Chaos Preview

References

• J. E. Moyal, “Quantum mechanics as a statistical theory,” Proceedings of the Cambridge Philosophical Society 45, 99–124 (1949).
• H. J. Groenewold, “On the principles of elementary quantum mechanics,” Physica 12, 405–460 (1946).
• M. Hillery, R. F. O’Connell, M. O. Scully, and E. P. Wigner, “Distribution functions in physics: fundamentals,” Physics Reports 106, 121–167 (1984).
• M. V. Berry and N. L. Balazs, “Evolution of semiclassical quantum states in phase space,” Journal of Physics A 12, 625–642 (1979).
• R. G. Littlejohn, “The semiclassical evolution of wave packets,” Physics Reports 138, 193–291 (1986).
• C. K. Zachos, D. B. Fairlie, and T. L. Curtright, eds., Quantum Mechanics in Phase Space, World Scientific, 2005.

Exercises

1. Expand the sine definition of the Moyal bracket through order $\hbar^2$.

Solution

Start with

$$\{A,B\}_M = \frac{2}{\hbar} A \sin\left( \frac{\hbar}{2}\Lambda \right) B.$$

Using

$$\sin u = u-\frac{u^3}{3!}+O(u^5),$$

one finds

$$\begin{aligned} \{A,B\}_M &= \frac{2}{\hbar} A \left[ \frac{\hbar}{2}\Lambda - \frac{1}{3!} \left( \frac{\hbar}{2}\Lambda \right)^3 + O(\hbar^5) \right] B \\ &= A\Lambda B - \frac{\hbar^2}{24} A\Lambda^3B + O(\hbar^4). \end{aligned}$$

Since $A\Lambda B=\{A,B\}_{\mathrm{PB}}$, the leading term is the Poisson bracket.

2. Derive the first Moyal correction for $H=p^2/(2m)+V(x)$.

Solution

The order-$\hbar^2$ term is

$$- \frac{\hbar^2}{24} H\Lambda^3W.$$

The kinetic term is quadratic in $p$, so its third derivatives vanish. The potential depends only on $x$, so the only surviving cubic term is

$$H\Lambda^3W = V'''(x) \frac{\partial^3W}{\partial p^3}.$$

Hence

$$\begin{aligned} \frac{\partial W}{\partial t} &= - \frac{p}{m} \frac{\partial W}{\partial x} + V'(x) \frac{\partial W}{\partial p} \\ &\quad- \frac{\hbar^2}{24} V'''(x) \frac{\partial^3W}{\partial p^3} + O(\hbar^4). \end{aligned}$$

3. Obtain the dimensionless correction scale from characteristic lengths.

Solution

If each $x$ derivative contributes a scale $L_x^{-1}$ and each $p$ derivative contributes $L_p^{-1}$, then

$$A\Lambda B \sim \frac{AB}{L_xL_p}.$$

The cubic term scales as

$$\hbar^2A\Lambda^3B \sim \frac{\hbar^2AB} {L_x^3L_p^3}.$$

Their ratio is

$$\frac{\hbar^2}{L_x^2L_p^2} = \left( \frac{\hbar}{L_xL_p} \right)^2.$$

Thus the dimensionless expansion parameter is

$$\epsilon_M = \frac{\hbar}{L_xL_p}.$$

4. Prove that every quadratic Hamiltonian gives exact Poisson evolution of the Wigner function.

Solution

Every correction beyond the Poisson bracket contains

$$H_W\Lambda^{2n+1}W, \qquad n\geq1.$$

The total derivative order acting on $H_W$ is $2n+1\geq3$. A quadratic polynomial has no derivatives of order three or higher, so every correction vanishes:

$$H_W\Lambda^{2n+1}W = 0, \qquad n\geq1.$$

Therefore

$$\{H_W,W\}_M = \{H_W,W\}_{\mathrm{PB}}$$

exactly. The conclusion concerns the transport equation, not whether $W$ is nonnegative or the state is classical.

5. Explain why interference fringes can invalidate power counting based only on explicit powers of $\hbar$.

Solution

For packets separated by $\Delta x$, the Wigner interference term can oscillate in momentum on the scale

$$\delta p \sim \frac{\hbar}{\Delta x}.$$

Each momentum derivative then contributes roughly

$$\partial_p \sim \frac{\Delta x}{\hbar}.$$

An apparently small term such as

$$\hbar^2 \partial_p^3W$$

can scale as

$$\hbar^2 \left( \frac{\Delta x}{\hbar} \right)^3 W = \frac{(\Delta x)^3}{\hbar}W.$$

The derivatives offset the explicit $\hbar^2$. The correct estimate must include the phase-space scales of the evolving state.