Picometer control of a levitating milligram gravity sensor

This paper demonstrates the simultaneous linear feedback cooling of two translational modes of a milligram magnetically levitated gravity sensor to below 2 picometers and 10 millikelvin, utilizing a high-Q superconducting trap and extreme vibration isolation to pave the way for future quantum ground-state experiments.

The Gist

Imagine you have a tiny, heavy marble (about the weight of a small paperclip) floating in mid-air. It's not floating on water or air currents; it's floating because it's being pushed away by a super-strong magnetic field inside a special metal trap. This is the "levitating milligram gravity sensor" described in the paper.

The scientists wanted to see how perfectly still they could make this floating marble. Why? Because to study the weird rules of quantum physics (the rules that govern the very small) on something as heavy as a marble, you have to stop it from jiggling almost completely. If it jiggles too much, the quantum effects get lost in the noise.

Here is how they did it, broken down into simple concepts:

1. The Floating Marble and the "Super-Silent" Room

The marble is a small magnet. It floats inside a "Type I superconducting trap." Think of this trap as a magical bowl made of a special metal (tantalum) that, when cooled to near absolute zero, repels the magnet so strongly that the magnet never touches the sides.

To keep the marble from shaking, the entire experiment is placed inside a "dry dilution refrigerator" (a giant, ultra-cold cooler). But cold isn't enough; the building itself vibrates (from traffic, pumps, etc.). So, the scientists built a multi-layered suspension system.

• The Analogy: Imagine the experiment is a delicate chandelier hanging inside a house. To stop the chandelier from shaking when someone walks by, they don't just hang it on a string. They hang it on a series of heavy springs and massive weights, which are themselves hanging on even bigger springs, all sitting on a 25-ton concrete block in the basement. This setup is so good at stopping vibrations that it blocks out 99.999999999% of the shaking energy at the frequencies that matter.

2. The "Eyes" and the "Hands"

The scientists needed to see the marble move and then stop it.

• The Eyes (Detection): They used a device called a SQUID (Superconducting Quantum Interference Device). This is an incredibly sensitive "eye" that can detect the tiniest change in the magnetic field caused by the marble moving. It's so sensitive it can see the marble move by less than the width of a single atom (picometers).
• The Hands (Feedback): When the marble starts to wiggle, the "eye" tells a computer. The computer instantly sends a signal to a "piezo actuator" (a tiny motor that can vibrate very precisely). This motor shakes the entire trap in the exact opposite direction of the marble's wobble.
• The Analogy: Imagine you are trying to balance a broomstick on your hand. If the stick leans left, you move your hand left to catch it. But here, the "hand" (the trap) is moving so precisely and quickly that it cancels out the "wind" (vibrations) trying to push the "stick" (the magnet) over. This is called feedback cooling.

3. The Result: A Stillness Beyond Imagination

By using this "catch and counter-move" technique, the scientists managed to calm the marble down to a state of near-perfect stillness.

• The Scale: They reduced the marble's movement to less than 2 picometers. To visualize this: A human hair is about 50,000 to 100,000 picometers wide. They made the marble move less than 1/25,000th of the width of a single hair.
• The Temperature: In physics, "temperature" for a single object often just means "how much it is jiggling." They cooled the motion of the marble to below 10 millikelvin (that's 0.01 degrees above absolute zero).

4. Why This Matters (According to the Paper)

The paper states that this setup is a "gravity sensor." Because the marble is heavy (for a quantum experiment) and so still, it can detect tiny changes in gravity.

The main achievement of this paper is proving that you can take a relatively heavy object (a milligram is huge in the quantum world) and cool it down to a state where it is almost perfectly still, using a combination of:

1. Super isolation (stopping outside vibrations).
2. Super detection (seeing the tiniest movements).
3. Active feedback (pushing back against the movement instantly).

The authors conclude that while they haven't reached the "quantum ground state" (the absolute lowest possible energy level) yet, they are very close. They believe that with a few more improvements—like better vibration isolation and even quieter sensors—they could eventually freeze this floating marble so completely that it would start behaving like a quantum object, bridging the gap between the heavy world we live in and the tiny, strange world of quantum mechanics.

In short: They built a super-stable, super-cold cradle for a floating magnet and used a high-speed "anti-wobble" system to make it so still that it barely moves at all, proving it's possible to prepare a heavy object for quantum experiments.

Technical Summary

Technical Summary: Picometer Control of a Levitating Milligram Gravity Sensor

Problem and Motivation
The pursuit of large-mass quantum experiments aims to bridge the gap between microscopic quantum systems and macroscopic general relativity. A critical prerequisite for observing quantum mechanical effects in mechanical systems is cooling a particle close to its mechanical quantum ground state. While ground-state cooling has been achieved in micromechanical membranes, trapped ions, and optically levitated nanoparticles, these platforms face specific limitations. Membranes suffer from clamping losses; trapped ions are limited to atomic scales; and optically levitated nanoparticles, while promising, possess masses in the femtogram range, which is insufficient to generate measurable gravitational interactions.

Magnetic levitation offers a pathway to larger masses without internal heating or clamping losses. However, the relatively large masses in magnetic levitation result in lower resonance frequencies compared to optical systems, making the damping of mechanical vibrations more challenging. This work addresses the challenge of simultaneously achieving high-quality factor (Q) levitation, state-of-the-art vibration isolation, and low-noise position detection to reduce the motion of a milligram-scale particle to the picometer regime, a necessary step toward future quantum ground-state cooling.

Methodology
The experiment utilizes a passive superconducting levitation platform. The core sensor consists of a composite object with an estimated mass of 0.35 mg, formed by stacking three cubic neodymium iron boron magnets (0.25 mm side length) and attaching a 0.20 mm glass sphere to break rotational symmetry. This assembly is levitated via Meissner repulsion within a type-I superconducting trap made of tantalum.

The detection and control system operates as follows:

1. Sensing: A superconducting pick-up coil detects the motion of the levitated magnet via induced currents. This coil is part of a detection circuit coupled to a DC SQUID (Superconducting Quantum Interference Device) amplifier.
2. Signal Processing: The SQUID signal is processed by a lock-in amplifier, which calculates a linear feedback signal.
3. Actuation: The feedback signal is sent to a piezoelectric actuator attached to the trap housing. By applying a phase-shifted tone to the trap, the system exerts a feedback force proportional to the particle's velocity, effectively increasing the damping of specific resonant modes.
4. Isolation: The entire setup is housed in a dry dilution refrigerator. To mitigate environmental noise, the experiment is mounted on a multi-stage mass-spring system suspended from a 25-metric ton concrete block. This system provides 110–130 dB of vibrational attenuation in the 50–70 Hz frequency range.

The researchers focused on two translational modes (identified as the x- and y-modes) with resonance frequencies of approximately 50.6 Hz and 68.0 Hz. The motion was calibrated using a calibration transformer to convert SQUID flux measurements into absolute displacement units (V/m).

Key Results
The study successfully demonstrated simultaneous linear feedback cooling of the two translational modes:

• Mode 3 (y-mode): Resonance frequency of 50.59 Hz with a Q factor of $3.8 \times 10^6$. The mode temperature was reduced to 7.1 mK, corresponding to an RMS amplitude of 1.6 pm.
• Mode 4 (x-mode): Resonance frequency of 67.98 Hz with a Q factor of $5.5 \times 10^6$. The mode temperature was reduced to 6.6 mK, corresponding to an RMS amplitude of 1.2 pm.

These cooling levels correspond to phonon numbers of $2.9 \times 10^6$ and $2.0 \times 10^6$ for modes 3 and 4, respectively. The experiment also noted the presence of "upconverted" noise peaks in the spectrum, attributed to the non-linear mixing of the vibration isolation system's modes with the particle's motion.

Significance and Future Outlook
The paper establishes that combining high-Q magnetic levitation, extensive vibration isolation, and SQUID-based readout can achieve picometer-level control over a milligram-scale object. The authors state that this setup serves as a benchmark for the system's performance, demonstrating that all necessary experimental conditions can be fulfilled simultaneously.

While the current phonon numbers are far from the quantum ground state (which would require $\sim 1$ phonon), the authors argue that the system is a promising platform for combining gravitational interaction and quantum mechanical effects. They identify specific pathways to reach the ground state:

1. Improved Vibration Isolation: Reducing force noise and eliminating the upconversion of low-frequency mechanical resonances.
2. Higher Q Factors: Requiring even better isolation to minimize damping.
3. Enhanced Detection: Increasing the coupling between the pick-up coil and the magnet to lower detection noise.
4. SQUID Optimization: Improving the SQUID's energy resolution to approach the quantum limit, as current back-action noise dominates over thermal noise at certain coupling levels.

The authors conclude that with these improvements, cooling a magnetically levitated particle to its quantum ground state is theoretically and experimentally feasible, paving the way for future large-mass quantum gravity experiments.