Levitation of a YIG sphere using a magnetic Paul trap - towards strongly coupled quantum magno-mechanics

AI summary

Overview

Research area: Quantum magnomechanics and levitated optomechanics — hybrid quantum systems in which the quantized spin-wave excitations (magnons) of a magnetic crystal are coupled to its centre-of-mass motion.

Technical level: Intermediate. The paper combines a classical stability analysis (vector Mathieu equations, Floquet theory) with an experimental room-temperature levitation demonstration and a linearized quantum-noise calculation of magnon–phonon cooperativity. Readers should be comfortable with the Mathieu equation, normal-mode/PSD language, and standard optomechanical linearization.

Scope (one sentence): The paper reports the room-temperature levitation of a 0.2 mm-diameter yttrium iron garnet (YIG) sphere in a planar magnetic Paul trap, a six-degree-of-freedom stability analysis of that trap, and a numerical estimate showing that the magnon–centre-of-mass cooperativity can exceed unity under feasible driving conditions.

What This Paper Is About

Magnetic Paul traps had previously been used only with hard magnets, because the existing classical theory of trapping assumes a magnetization of fixed magnitude and fixed orientation relative to the body. The authors ask whether the same trap architecture can levitate a soft magnet such as YIG — the material of choice for magnonics — by applying a bias field larger than the saturation magnetization, and whether the resulting system could reach the strong-coupling regime between magnons and mechanical motion.

The goal is twofold: demonstrate stable room-temperature trapping of a small YIG sphere experimentally, and quantify how strongly magnons could couple to the sphere's centre-of-mass motion given realistic damping and drive parameters.

Key Contributions

1. Extension of magnetic Paul trapping from hard to soft magnets. The authors argue that a soft anisotropic magnet with saturated magnetization behaves like a hard magnet in the classical limit, provided the bias field satisfies (B_0 > B_s), and they realize this experimentally with YIG.

2. Full six-degree-of-freedom stability analysis including librational coupling. Unlike an ion, a levitated magnet has tilting degrees of freedom. The translational radial motion and the librational motion form a coupled vector Mathieu equation, solved numerically via Floquet theory to produce the stability diagram in the (a_z)–(q_z) plane.

3. Room-temperature levitation and motional characterization. Continuous trapping of a 0.2 mm-diameter YIG sphere for at least an hour, with measured Q-factors of order 25 and secular frequencies of 15.8 Hz and 17.2 Hz, plus micromotion and sideband structure in the motional spectrum.

4. Numerical estimate of enhanced magnon–mechanical cooperativity. Using a driven-magnon model, the authors estimate a single-magnon coupling of (g_{mb}/2\pi \sim 17) µHz and show that the cooperativity (C_{mb}) can exceed unity for feasible drive powers, despite large magnon damping.

Main Findings

• Coupling shifts the stability region: The librational–translational coupling term (c) (which can be tuned via the AC field frequency and magnetic gradient) changes the trap's stability region. The experimental operating point ((q_z = 0.48), (a_z \simeq 0), (a_s \simeq 8600), (c \simeq 0.5)) is marked with a brown diamond in Fig. 1(a) and lies inside the stable region.

• Soft-magnet levitation is viable: With two N52-grade neodymium disc magnets in Helmholtz configuration (2 inch diameter, 1/2 inch thickness, surface separation 37.6 mm, measured minimum field 0.247 T), a planar PCB trap with 1 mm and 2 mm radius electrodes (35 µm thick, 0.3 mm wide, 0.7 mm hole), and AC currents of 0.594 A and 1.223 A RMS at 140 Hz, the authors trapped a 0.2 mm-diameter YIG sphere (saturation magnetization 0.178 T) at room temperature.

• Measured secular frequencies and damping: For trapping currents (I_{1ac} = 0.48) A RMS, (I_{2ac} = 1) A RMS, biases (I_{1dc} = 56) mA and (I_{2dc} = 113) mA at (\Omega/2\pi = 140) Hz, the power spectral density shows secular peaks at (f_x = 15.8) Hz and (f_y = 17.2) Hz, with Lorentzian fits giving (Q_x = 25) and (Q_y = 18).

• Micromotion and sidebands resolved: The spectrum contains the micromotion peak at (f_m = 140) Hz and sidebands at (f_m \pm f_x) and (f_m \pm f_y), which the authors assign to the closely spaced x and y modes. Additional harmonics appear but are not attributed to secular motion because they are too narrow for a room-pressure experiment.

• Ring-down confirmation of Q: For (I_{1ac} = 0.587) A RMS, (I_{2ac} = 1.175) A RMS, (I_{1dc} = 59) mA, (I_{2dc} = 116) mA at 140 Hz, modulating the trapping amplitude at 19.3 Hz with 1% depth and fitting the decay exponentially yields a Q of 29 (figure caption) and Q = 31 (main text) — the paper reports both values and states the latter is close to the Lorentzian estimate.

• Z-mode not measured: The authors state that the z-mode could not be measured due to the limited sensitivity of the optical detection along the z-axis.

• Cooperativity above unity: Assuming (\kappa_m/2\pi = 1) MHz, (\kappa_b/2\pi \sim 0.4) mHz, sphere diameter (d = 0.2) mm, (Q = 10^3), gradient (\xi = 252) mT/m, spin density (\rho_s = 4.3\times10^{27},\mathrm{m^{-3}}), (\gamma_e/2\pi = 28) GHz/T, (T = 4) K, (B_{sat} = 0.254) T and (\omega_b/2\pi = 62) Hz, the cooperativity (C_{mb} = G_{mb}^2/(\kappa_m\kappa_b)) reaches moderately high values above 1 for zero detuning as drive power increases. The drive power is capped to preserve the Holstein–Primakoff approximation (only a small fraction of spins excited).

• Trapping duration and stability: Continuous trapping was maintained for at least an hour; the experimental operating point is confirmed to sit inside the theoretically stable region.

Methodology in Plain English

The researchers first built a classical model of a magnet sitting in a superposition of magnetic fields: a large static uniform bias field, a static linear gradient that counteracts gravity, a static field curvature, and a time-varying (AC) field curvature. Assuming the bias field dominates, the magnetization stays fixed in magnitude and direction relative to the crystal, so the same equations used for hard magnets apply.

Writing Newton's equations for translation and Euler's equations for tilt, and rescaling coordinates and time, they obtained a Mathieu equation for the vertical (z) motion, and a coupled pair of equations for the radial motion and the tilt angle (s). That pair is cast as a vector Mathieu equation with a (2\times2) matrix (A) containing the coupling (c). They solve it with Floquet theory: they write the solution as a sum of exponentials with characteristic exponents (\Xi_j), substitute into the equation to obtain a recursive matrix relation, then numerically solve (\det M = 0) and require (\mathrm{Im},\Xi = 0) for stability. The trap is stable where the stable regions of z, r and s overlap.

For the experiment, two permanent disc magnets provide the static fields; a custom PCB with two concentric current-carrying electrodes provides both the AC trapping curvature (from counter-circulating currents at 140 Hz) and the DC bias currents used to compensate unwanted static curvature. The sphere is confined in a 0.7 mm hole between glass coverslips. Motion is read out optically: laser light is focused onto the sphere and the transmitted/backscattered light falls on a photodetector, whose voltage is spectrally analysed. A ring-down measurement — modulating the trap current at 19.3 Hz with 1% depth, then switching the modulation off and fitting the exponential decay — provides an independent Q measurement.

Finally, they model the quantum system: a harmonic mechanical mode coupled to a magnon mode through the magnetic gradient, with a microwave drive added. Linearizing around the driven steady state gives an enhanced coupling (G_{mb} = \sqrt{n},g_{mb}) that grows with the steady-state magnon occupation (n), found by solving a cubic steady-state equation. The ratio of this coupling squared to the product of magnon and phonon decay rates is the cooperativity, plotted against drive power.

Why This Matters

The work shows that a soft magnetic material with strong magnon–photon coupling can be levitated with a chip-scale magnetic trap at room temperature, without cryogenics. That combination matters because magnons remain quantum-coherent at room temperature, and a levitated magnet offers a massive, isolated mechanical oscillator — a platform for testing quantum mechanics with macroscopic objects, for precision inertial sensing, and for storing and transducing quantum information.

Real-world applications (as framed in the paper):

• Passive levitation of massive objects for advanced inertial sensors.
• Generation of non-classical macroscopic motional states.
• Table-top tests of low-energy gravity with quantum mechanics.
• Quantum information processing, quantum interconnects and quantum memories, using the long-lived mechanical mode for storage.

Industry relevance: Chip-scale planar magnetic traps are compatible with printed-circuit-board fabrication, which points toward integrable room-temperature devices rather than dilution-refrigerator-based systems. The magnon–mechanical interface is relevant to quantum transducers, microwave-to-optical conversion, and hybrid architectures combining magnons with superconducting qubits and microwave cavities.

Future Directions

• Cooling the centre-of-mass motion. The demonstrated mechanical Q-factors of order 25 at room pressure are far from the (Q = 10^3) assumed in the quantum estimate, so feedback cooling or vacuum operation is needed to approach the quantum regime.
• Detecting the z-mode. Better optical detection along the trap axis would fill the one gap in the experimental characterization and allow a full comparison with the stability calculation.
• Experimental verification of strong cooperativity. The prediction (C_{mb} > 1) is numerical; demonstrating it requires a driven, low-damping magnomechanical system.
• Managing librational–translational coupling. Since the analysis shows coupling (c) reshapes the stability region, controlling or exploiting this coupling (via AC frequency and field gradient) is an open design question for scaling the trap to other magnet geometries.

Target Audience

Researchers in quantum magnonics, levitated optomechanics and hybrid quantum systems; experimentalists interested in room-temperature mechanical oscillators with spin-wave interfaces; and graduate students or device engineers looking for a concrete, chip-fabricated design for magnet levitation with a worked stability analysis. Readers focused on purely quantum experiments will need the supplementary material for the derivation of the equations of motion.

Authors’ abstract

Magnetic levitation offer passive levitation of massive objects for use in advanced inertia sensors, for the generation of non-classical macroscopic motional states, and towards the table-top testing of low energy gravity with quantum mechanics. Magnons, a quanta of spin wave, couple to many physical quantities and strongly to electromagnetic fields, even at room temperature. In this work we demonstrate the stable trapping of a small YIG sphere using a magnetic Paul trap. We present a classical stability analysis of a magnetic Paul trap and show the stability diagram for all mechanical degrees of freedom. We show that coupling between librational and translational modes changes the stability region. We experimentally levitate the soft magnet yttrium iron garnet at room temperature obtaining Q-factors of $\sim25$ and secular frequencies $15.8$ Hz and $17.2$ Hz. We provide numerical estimates of the achievable enhanced coupling between the center-of-mass motion and excited magnon modes, with a cooperativity above unity despite strong mechanical damping, indicating potential applications in quantum information processing, quantum interconnects and quantum memories.