Repeated weak measurements: watching quantum correlations evolve

This paper introduces a minimally invasive protocol using repeated weak measurements to directly observe the time evolution of many-body quantum correlations without external perturbation, successfully demonstrating the technique by measuring the Van Hove function and dynamical structure factor in an atomic Bose-Einstein condensate.

The Gist

Imagine you are trying to watch a delicate soap bubble float across a room. If you shine a bright, harsh spotlight on it to see where it is, the light pressure might pop the bubble or push it off course. You've disturbed the very thing you wanted to watch. This is the classic problem in quantum physics: measuring a system often breaks it.

For decades, scientists have had to choose between "strong" measurements (which give clear answers but destroy the quantum state) or "weak" measurements (which are gentle but give very fuzzy, noisy data).

This paper introduces a clever new trick: using two weak measurements in a row to watch quantum ripples evolve without ever popping the bubble.

Here is the story of how they did it, explained with everyday analogies.

1. The Setup: A Quantum "Cloud"

The scientists used a Bose-Einstein Condensate (BEC). Think of this not as a gas, but as a single, giant "super-atom" cloud where thousands of atoms act in perfect unison, like a synchronized dance troupe. In this cloud, if one atom moves, the whole cloud ripples. These ripples are called phonons (sound waves).

2. The Problem: The "Flashbulb" Effect

Usually, to see these ripples, you'd hit the cloud with a laser pulse (a "strong measurement"). But this is like taking a photo with a flashbulb: it freezes the moment, but the flash itself might startle the dancers, changing their routine. You can't see how they naturally move because you interrupted them.

3. The Solution: The "Whisper" and the "Echo"

The team developed a method using two weak measurements. Imagine trying to hear a whisper in a noisy room.

• Measurement 1 (The Whisper): Instead of a flashbulb, they use a very dim, gentle light. It's so weak that it barely disturbs the cloud. It's like whispering, "Hey, are you there?" to the cloud.

  • The Catch: Because the light is so dim, the answer is very fuzzy. You get a lot of static noise. It's like hearing a whisper through a bad radio connection.
  • The Magic: Even though the measurement is fuzzy, it does leave a tiny, gentle nudge on the cloud (called backaction). It's like the whisper itself slightly pushes the dancers.

• Measurement 2 (The Echo): A tiny fraction of a second later, they whisper again to the same cloud.

  • This second whisper picks up two things:
    1. The natural ripples that were already there.
    2. The ripples created by the first whisper's gentle nudge.

4. The Detective Work: Correlating the Noise

Here is the genius part. If you look at just one experiment, the data is just static noise. It looks like random snow on an old TV.

But, the scientists repeated this experiment 128 times. They took all the "static" from the first whisper and all the "static" from the second whisper and cross-correlated them.

• The Analogy: Imagine 128 people trying to hear a faint echo in a cave. Individually, they just hear wind noise. But if they all compare their recordings, they realize: "Wait, every time we heard a specific type of wind noise in the first recording, we heard a specific echo in the second recording."
• By mathematically matching the noise patterns, the random static cancels out, and the true signal (the movement of the ripples) emerges clearly.

5. What They Saw: The Van Hove Function

By doing this, they could watch the ripples travel across the cloud in real-time.

• They saw a ripple start at one spot.
• They watched it split into two waves moving in opposite directions.
• They measured exactly how fast these waves moved (the speed of sound in the quantum cloud).

They created a map called the Van Hove function, which is essentially a "movie" of how quantum correlations spread through the system.

6. The "Post-Selection" Trick (The Quantum Amplifier)

The paper also uses a concept called Quantum Weak Values. This is like a magic filter.

• Normally, you average all 128 experiments.
• But the scientists said, "Let's only keep the data where the first whisper was positive (a specific type of noise)."
• By throwing away half the data and keeping only the "positive" whispers, the signal from the second measurement gets amplified. It's like tuning a radio to a specific frequency to make a faint station loud and clear.
• This allowed them to isolate the specific effect of the "nudge" (backaction) from the rest of the noise, proving that the measurement itself was creating the ripples they were watching.

Why Does This Matter?

This is a huge deal for a few reasons:

1. No More "Breaking" the System: We can now watch quantum systems evolve naturally without smashing them with heavy-handed measurements.
2. Universal Tool: This isn't just for cold atoms. This technique could work on superconducting qubits (computer chips), trapped ions, or even light particles.
3. New Physics: It opens the door to studying "non-equilibrium" physics—how quantum systems behave when they are chaotic or changing rapidly, rather than just sitting still.

In Summary:
The scientists figured out how to watch a quantum system dance by whispering to it twice. By listening carefully to the "static" in their whispers and comparing them, they could see the dance steps clearly without ever stepping on the dancers' toes. They turned the "noise" of measurement into a powerful tool for discovery.

Technical Summary

1. Problem Statement

Experimental access to the dynamical properties of many-body quantum systems is traditionally limited by measurement backaction. Standard "strong" (projective) measurements fully collapse the wavefunction, erasing the system's initial state and preventing the observation of subsequent time evolution. Consequently, obtaining dynamical correlation functions (such as the Van Hove function or Dynamical Structure Factor) usually requires:

• Applying an external perturbation (drive) to an equilibrium system.
• Measuring the system's response.
• Repeating the process for different parameters to build a statistical ensemble.

This "active paradigm" is invasive and complex. The authors seek a method to measure dynamical correlations without external perturbation, leveraging the measurement process itself to probe the system while minimizing information loss.

2. Methodology

The authors propose and demonstrate a protocol based on time-separated pairs of weak quantum measurements on an atomic Bose–Einstein Condensate (BEC).

• System: A $^{87}$Rb BEC ($\approx 2.0 \times 10^5$ atoms) confined in a crossed dipole trap.
• Measurement Technique: Phase-Contrast Imaging (PCI). This is a dispersive, weak measurement technique where a far-detuned probe laser acquires a phase shift proportional to the local atomic density. The phase shift is converted to intensity via an on-axis homodyne interferometer.
• The Protocol:
  1. First Measurement ($M_1$) at $t=0$: A weak measurement of the atomic density is performed. This yields a noisy outcome $n_x = \langle \hat{n}_x \rangle + \delta n_x$. Crucially, the measurement induces quantum backaction, locally updating the many-body wavefunction and creating excitations correlated with the measurement noise.
  2. Time Evolution ($\delta t$): The system evolves freely for a time delay $\delta t$.
  3. Second Measurement ($M_2$) at $t=\delta t$: A second weak measurement is performed at a displaced position $x + \delta x$.
• Data Analysis Strategies:
  • Cross-Correlation Function (CCF): The ensemble-averaged correlation between the noise of the first measurement and the noise of the second measurement directly yields the Van Hove function $G(\delta x, \delta t)$, independent of the measurement strength.
  • Quantum Weak Values (QWV): Inspired by Aharonov's post-selection, the authors filter the second measurement outcomes based on the sign of the first measurement's noise (e.g., keeping only cases where $\delta n_x > 0$). This post-selection isolates the contribution of quantum backaction, yielding a signal proportional to the measurement strength $\phi$ and the Van Hove function. This allows for the amplification of the backaction signal.

3. Key Contributions

• Passive Dynamical Spectroscopy: The work demonstrates a "many-body analogue of quantum nonlinear spectroscopy" where dynamical correlations are extracted without applying external drive fields or perturbations.
• Direct Measurement of Backaction: By using QWV post-selection, the authors isolate and quantify the specific role of quantum backaction in driving system dynamics, distinguishing it from pre-existing thermal fluctuations.
• Generalizability: The method is shown to be applicable to any pair of observables amenable to weak measurement, extending beyond the limitations of strong measurements.
• Robustness: The technique works effectively even in partially condensed, higher-temperature BECs where the condensate fraction is low.

4. Key Results

• Observation of Sound Propagation:
  • The authors measured the Van Hove function $G(\delta x, \delta t)$, which revealed the propagation of density fluctuations.
  • At short delays ($\delta t = 0.45$ ms), correlations were localized near $\delta x = 0$.
  • At longer delays ($\delta t = 3.0$ ms), the correlations split into two distinct peaks moving at the speed of sound ($c \approx 1.31$ mm/s), visually demonstrating the ballistic spread of Bogoliubov excitations.
• Dynamical Structure Factor (DSF):
  • By performing a 2D Fourier transform of the measured Van Hove function, the authors reconstructed the Dynamical Structure Factor $S(k, \omega)$.
  • The resulting spectrum matched the theoretical Bogoliubov dispersion relation, confirming the method's ability to map the collective excitation spectrum of the many-body system.
• QWV Amplification:
  • The study confirmed that the QWV signal amplitude grows linearly with the first measurement's strength ($\phi$), whereas the CCF signal amplitude remains constant.
  • Post-selection (discarding a fraction $f_d$ of data) amplified the signal slope, validating the theoretical prediction of weak value amplification.
  • The Signal-to-Noise Ratio (SNR) was optimized by balancing the amplification gain against the loss of data statistics.
• Temperature Dependence:
  • Experiments were conducted at different temperatures (varying condensate fraction $R_c$ from 0.88 to 0.49).
  • The measured speed of sound decreased as $R_c$ decreased, consistent with theoretical models for finite-temperature trapped BECs ($c \propto R_c^{1/5}$).
  • The QWV protocol successfully extracted quantitative correlation data even in the partially condensed regime.

5. Significance

• Fundamental Physics: The work provides a new window into non-equilibrium quantum dynamics, showing that measurement backaction is not merely a nuisance but a controllable resource for probing and potentially engineering quantum states.
• Technological Impact:
  • Quantum Sensing: The method offers a route to high-precision measurements of correlation functions without disturbing the system's equilibrium state via strong drives.
  • Quantum Control: The ability to use global weak measurements with real-time feedback suggests a path toward stabilizing far-from-equilibrium states and creating measurement-induced phases in large-scale quantum systems (e.g., superconducting qubit arrays or trapped ions).
  • Broad Applicability: The technique is platform-agnostic and can be applied to various quantum systems (photonic lattices, spin systems) to probe correlations between any weakly observable quantities (magnetization, spin density, momentum).

In summary, Altuntaş and Spielman have successfully replaced the traditional "perturb-and-measure" paradigm with a "measure-and-correlate" approach, utilizing the interplay between weak measurement, backaction, and unitary evolution to directly observe the time evolution of quantum correlations.