Papers
Topics
Authors
Recent
Search
2000 character limit reached

Quantum Echoes: Time-Reversal in BEC

Updated 5 July 2026
  • Quantum Echoes Experiment is a controlled time-reversal protocol for mesoscopic Bose–Einstein condensates that distinguishes coherent superpositions from decoherence through engineered Hamiltonian inversion.
  • It employs periodic shaking and an interaction sign flip to reverse many-body dynamics, effectively refocusing damped Rabi-like oscillations.
  • Numerical validation shows that restoring population imbalance and reducing number fluctuations serve as practical indicators of mesoscopic coherence.

A Quantum Echoes Experiment is an effective time-reversal protocol for a mesoscopic Bose–Einstein condensate in a periodically shaken double well, proposed to distinguish coherent many-body superpositions from incoherent statistical mixtures by refocusing the dynamics after an apparent collapse of NN-particle Rabi-like oscillations (Weiss, 2011). In this setting, the “echo” is not a phenomenological revival but the controlled reversal of an effective many-body Hamiltonian, implemented by periodic shaking and a sign change of the interaction. The proposal is motivated by a central diagnostic problem: damped population oscillations and large number fluctuations can arise either from unitary evolution into mesoscopic superpositions or from decoherence, and static observables alone do not separate these possibilities.

1. Mesoscopic double-well platform

The physical system is a mesoscopic Bose–Einstein condensate in a double-well potential, typically with N100N \sim 100–$1000$ atoms. In the two-mode approximation, one localized mode is retained in each well, and the undriven system is described by the two-site Bose–Hubbard Hamiltonian

H^0=J(c^1c^2+c^2c^1)+U2j=12n^j(n^j1).\hat{H}_0 = -J \left( \hat{c}_1^\dagger \hat{c}_2 + \hat{c}_2^\dagger \hat{c}_1 \right) + \frac{U}{2} \sum_{j=1}^2 \hat{n}_j (\hat{n}_j - 1).

Here JJ is the tunneling matrix element, UU is the on-site interaction strength, n^j=c^jc^j\hat{n}_j=\hat{c}_j^\dagger \hat{c}_j, and total particle number N^=n^1+n^2\hat{N}=\hat{n}_1+\hat{n}_2 is conserved. The Hilbert-space dimension is N+1N+1, with Fock basis n1,n2|n_1,n_2\rangle such that N100N \sim 1000 (Weiss, 2011).

Periodic shaking is introduced as a time-dependent bias between the wells,

N100N \sim 1001

with shaking amplitude N100N \sim 1002 and angular frequency N100N \sim 1003. The main observable is the normalized population imbalance,

N100N \sim 1004

Starting from N100N \sim 1005, weak interactions produce coherent oscillations of population between the wells. These are N100N \sim 1006-particle Rabi-like oscillations: collective many-body tunneling that resembles single-particle Rabi dynamics in the observable N100N \sim 1007.

2. Apparent damping and the superposition problem

For finite interactions, the oscillations can appear damped even though the dynamics remains strictly unitary under N100N \sim 1008. For N100N \sim 1009 and $1000$0, numerical solutions show oscillating $1000$1 with decaying amplitude on timescales before any significant revival. This apparent damping is not a decoherence effect; it is associated with collapse of the coherent oscillation and the build-up of nontrivial many-body correlations and mesoscopic superpositions (Weiss, 2011).

The number-fluctuation diagnostic is

$1000$2

with $1000$3. For pure states, $1000$4 coincides with a quantum Fisher information with respect to phase shifts generated by $1000$5. Numerically, $1000$6 grows significantly as the apparent damping sets in, indicating the formation of states with large number fluctuations and metrological relevance. A paradigmatic reference point is the NOON state,

$1000$7

which exhibits maximal number fluctuations.

The many-body state is visualized with atomic coherent states,

$1000$8

where $1000$9 encodes the mean population imbalance and H^0=J(c^1c^2+c^2c^1)+U2j=12n^j(n^j1).\hat{H}_0 = -J \left( \hat{c}_1^\dagger \hat{c}_2 + \hat{c}_2^\dagger \hat{c}_1 \right) + \frac{U}{2} \sum_{j=1}^2 \hat{n}_j (\hat{n}_j - 1).0 the relative phase. A single atomic coherent state is a product state and satisfies H^0=J(c^1c^2+c^2c^1)+U2j=12n^j(n^j1).\hat{H}_0 = -J \left( \hat{c}_1^\dagger \hat{c}_2 + \hat{c}_2^\dagger \hat{c}_1 \right) + \frac{U}{2} \sum_{j=1}^2 \hat{n}_j (\hat{n}_j - 1).1. By contrast, at H^0=J(c^1c^2+c^2c^1)+U2j=12n^j(n^j1).\hat{H}_0 = -J \left( \hat{c}_1^\dagger \hat{c}_2 + \hat{c}_2^\dagger \hat{c}_1 \right) + \frac{U}{2} \sum_{j=1}^2 \hat{n}_j (\hat{n}_j - 1).2 in the driven example with H^0=J(c^1c^2+c^2c^1)+U2j=12n^j(n^j1).\hat{H}_0 = -J \left( \hat{c}_1^\dagger \hat{c}_2 + \hat{c}_2^\dagger \hat{c}_1 \right) + \frac{U}{2} \sum_{j=1}^2 \hat{n}_j (\hat{n}_j - 1).3, the wave function is highly non-Gaussian in the H^0=J(c^1c^2+c^2c^1)+U2j=12n^j(n^j1).\hat{H}_0 = -J \left( \hat{c}_1^\dagger \hat{c}_2 + \hat{c}_2^\dagger \hat{c}_1 \right) + \frac{U}{2} \sum_{j=1}^2 \hat{n}_j (\hat{n}_j - 1).4 representation, resembles several well-separated lobes in phase space, and has H^0=J(c^1c^2+c^2c^1)+U2j=12n^j(n^j1).\hat{H}_0 = -J \left( \hat{c}_1^\dagger \hat{c}_2 + \hat{c}_2^\dagger \hat{c}_1 \right) + \frac{U}{2} \sum_{j=1}^2 \hat{n}_j (\hat{n}_j - 1).5. The central misconception addressed by the proposal is that damped H^0=J(c^1c^2+c^2c^1)+U2j=12n^j(n^j1).\hat{H}_0 = -J \left( \hat{c}_1^\dagger \hat{c}_2 + \hat{c}_2^\dagger \hat{c}_1 \right) + \frac{U}{2} \sum_{j=1}^2 \hat{n}_j (\hat{n}_j - 1).6 together with large fluctuations does not, by itself, certify a coherent mesoscopic superposition: an incoherent statistical mixture can reproduce similar static number statistics.

3. Effective time reversal and the echo protocol

The proposal uses high-frequency periodic driving to replace the time-dependent Hamiltonian by an effective time-independent one,

H^0=J(c^1c^2+c^2c^1)+U2j=12n^j(n^j1).\hat{H}_0 = -J \left( \hat{c}_1^\dagger \hat{c}_2 + \hat{c}_2^\dagger \hat{c}_1 \right) + \frac{U}{2} \sum_{j=1}^2 \hat{n}_j (\hat{n}_j - 1).7

with renormalized tunneling

H^0=J(c^1c^2+c^2c^1)+U2j=12n^j(n^j1).\hat{H}_0 = -J \left( \hat{c}_1^\dagger \hat{c}_2 + \hat{c}_2^\dagger \hat{c}_1 \right) + \frac{U}{2} \sum_{j=1}^2 \hat{n}_j (\hat{n}_j - 1).8

where H^0=J(c^1c^2+c^2c^1)+U2j=12n^j(n^j1).\hat{H}_0 = -J \left( \hat{c}_1^\dagger \hat{c}_2 + \hat{c}_2^\dagger \hat{c}_1 \right) + \frac{U}{2} \sum_{j=1}^2 \hat{n}_j (\hat{n}_j - 1).9 is the zeroth-order Bessel function. For the parameters used, the high-frequency regime is roughly JJ0. Because JJ1 can change sign, the tunneling term can be reversed by switching the shaking amplitude. The paper gives the explicit pair

JJ2

for which

JJ3

with opposite signs (Weiss, 2011).

To reverse the full effective Hamiltonian, the protocol also flips the interaction sign JJ4. The resulting idealized sequence is

JJ5

Then the time-evolution operator is such that at JJ6 the net propagator is the identity and the system returns to its initial state. This is the many-body echo. In practice, the effective-Hamiltonian picture is approximate, the switching cannot be perfectly instantaneous, and the initial phase of the drive matters. The switching time should be chosen close to a maximum of JJ7 to minimize non-adiabatic excitations. The authors also test smooth switching,

JJ8

with analogous switching for JJ9, and still obtain a clear echo.

4. Discriminating superpositions from statistical mixtures

The diagnostic logic is dynamical rather than static. If the apparent damping arises from coherent many-body evolution, then reversing the effective Hamiltonian refocuses the state and restores the initial population imbalance. If the damping arises from decoherence, then environmental entanglement and loss of phase coherence are not undone by changing the sign of UU0 and UU1; the echo is then much weaker (Weiss, 2011).

The experimentally relevant quantity at the end of the sequence is

UU2

A large UU3 indicates that the intermediate state at UU4 belonged to a coherent unitary evolution. A small UU5 indicates that the apparent damping likely arose from decoherence and statistical mixing. The proposal therefore addresses the central ambiguity of mesoscopic interference experiments: it distinguishes reversible collapse from irreversible decoherence by testing refocusability.

The same logic also separates mesoscopic superpositions from generic product-state dynamics. Although an atomic coherent state itself never has UU6, time evolution of various initial product states under the designed protocol can produce large UU7. The paper therefore compares the target state to a scan over initial atomic coherent states UU8. None of these product states produces a return UU9 comparable to the mesoscopic superposition. The characteristic combined signature is: large fluctuations at n^j=c^jc^j\hat{n}_j=\hat{c}_j^\dagger \hat{c}_j0, n^j=c^jc^j\hat{n}_j=\hat{c}_j^\dagger \hat{c}_j1; very small fluctuations after the echo, n^j=c^jc^j\hat{n}_j=\hat{c}_j^\dagger \hat{c}_j2; and large n^j=c^jc^j\hat{n}_j=\hat{c}_j^\dagger \hat{c}_j3 near unity.

5. Numerical validation and experimental blueprint

The proposal is tested by solving the full time-dependent n^j=c^jc^j\hat{n}_j=\hat{c}_j^\dagger \hat{c}_j4-particle Schrödinger equation,

n^j=c^jc^j\hat{n}_j=\hat{c}_j^\dagger \hat{c}_j5

in the Fock basis with dimension n^j=c^jc^j\hat{n}_j=\hat{c}_j^\dagger \hat{c}_j6. Time integration uses the Shampine–Gordon routine, and most examples use n^j=c^jc^j\hat{n}_j=\hat{c}_j^\dagger \hat{c}_j7 (Weiss, 2011). Several parameter sets are highlighted. Under the undriven Hamiltonian, n^j=c^jc^j\hat{n}_j=\hat{c}_j^\dagger \hat{c}_j8, n^j=c^jc^j\hat{n}_j=\hat{c}_j^\dagger \hat{c}_j9, and initial state N^=n^1+n^2\hat{N}=\hat{n}_1+\hat{n}_20 give the apparent damping of N^=n^1+n^2\hat{N}=\hat{n}_1+\hat{n}_21. With driving N^=n^1+n^2\hat{N}=\hat{n}_1+\hat{n}_22 and N^=n^1+n^2\hat{N}=\hat{n}_1+\hat{n}_23, the state at N^=n^1+n^2\hat{N}=\hat{n}_1+\hat{n}_24 is a mesoscopic superposition with N^=n^1+n^2\hat{N}=\hat{n}_1+\hat{n}_25.

For the echo, the Hamiltonian parameters are switched at N^=n^1+n^2\hat{N}=\hat{n}_1+\hat{n}_26 from N^=n^1+n^2\hat{N}=\hat{n}_1+\hat{n}_27, N^=n^1+n^2\hat{N}=\hat{n}_1+\hat{n}_28 to N^=n^1+n^2\hat{N}=\hat{n}_1+\hat{n}_29, N+1N+10, so that N+1N+11. The population imbalance then shows a clear revival near N+1N+12, long before any natural revival under N+1N+13. The echo remains visible even for lower driving frequency N+1N+14, and it survives continuous switching according to the hyperbolic-tangent protocol. For realistic parameters N+1N+15, the time-reversal sequence recovers a large fraction of the initial population imbalance, with N+1N+16.

As an experimental blueprint, the protocol requires a double-well platform, periodic shaking with tunable amplitude N+1N+17, and interaction control sufficient to implement N+1N+18. The full sequence is explicit: prepare all atoms in one well N+1N+19; apply periodic shaking with n1,n2|n_1,n_2\rangle0 for time n1,n2|n_1,n_2\rangle1; switch to n1,n2|n_1,n_2\rangle2; continue for another interval n1,n2|n_1,n_2\rangle3; and measure n1,n2|n_1,n_2\rangle4 and n1,n2|n_1,n_2\rangle5. The paper also notes the main limitations: decoherence is not explicitly modeled, the two-mode approximation may receive corrections at very high driving frequencies, and technical limits on speed and precision of switching remain relevant.

6. Position within echo physics

The cold-atom proposal is explicitly akin to spin echo, Loschmidt echoes, and dynamical decoupling: in each case, reversibility or its failure is used as a probe of coherence. In the double-well Bose–Einstein condensate, the echo is not merely a recovery of a mean field but a refocusing of many-body unitary dynamics after controlled inversion of an effective Hamiltonian (Weiss, 2011).

Later theory on imperfect many-body echoes places this interpretation on a broader footing. For macroscopic observables n1,n2|n_1,n_2\rangle6, if the backward evolution is implemented with inaccuracies n1,n2|n_1,n_2\rangle7, the relative echo signal obeys

n1,n2|n_1,n_2\rangle8

and the echo peak scales as

n1,n2|n_1,n_2\rangle9

under generic many-body conditions (Dabelow et al., 2020). This later result does not alter the cold-atom construction; it clarifies its scope. A Quantum Echoes Experiment demonstrates reversibility of coherent collapse only insofar as the effective sign flip is accurate and decoherence remains negligible over the full duration. A plausible implication is that the double-well protocol should be understood as an experimentally accessible Loschmidt-echo-type discriminator of mesoscopic coherence, rather than as a universal reversal of all sources of irreversibility.

Topic to Video (Beta)

No one has generated a video about this topic yet.

Whiteboard

No one has generated a whiteboard explanation for this topic yet.

Follow Topic

Get notified by email when new papers are published related to Quantum Echoes Experiment.