Levitated nano-trampoline resonators for magnetic field sensing

Technical Summary: Levitated Nano-Trampoline Resonators for Magnetic Field Sensing

Problem Statement
The detection of extremely weak forces is a central challenge in precision measurement, with applications ranging from biomedical imaging to fundamental physics tests. While nanomechanical resonators offer exceptional sensitivity due to their small effective mass and high resonance frequencies, their ultimate performance is fundamentally constrained by energy dissipation (coupling to the thermal bath). To utilize these resonators for magnetic field sensing, they typically require functionalization—such as attaching magnetostrictive materials or magnetic elements. However, such approaches inevitably degrade the resonator's quality factor ($Q$), constrain material choices, and increase fabrication complexity, thereby limiting achievable sensitivity. Furthermore, existing levitation techniques (optical, electric, superconducting) often rely on complex trapping schemes, small masses, or cryogenic environments, while purely diamagnetically levitated systems are restricted to low frequencies due to intrinsically low mechanical resonance frequencies.

Methodology
The authors introduce a hybrid levitated nanomechanical resonator that decouples magnetic transduction from mechanical amplification. The system consists of two distinct components:

1. Macroscopic Proof Mass: A $10.2 \times 10.2 \times 1.2$ mm$^3$ pyrolytic graphite (PG) plate is diamagnetically levitated above an array of alternating NdFeB permanent magnets. This plate acts as a free-floating proof mass that couples strongly to external magnetic fields without requiring active feedback or power.
2. High-Q Nanomechanical Amplifier: A chip containing a silicon nitride (SiN) square membrane (100 nm thick, side lengths 0.2–1.4 mm) is placed on top of the levitated graphite. The SiN membrane serves as a high-$Q$ resonator (up to $6 \times 10^6$) that mechanically couples to the graphite's motion.

Working Principle
When an external oscillating magnetic field ($B_s$) is applied, it interacts with the diamagnetic graphite in the presence of the static magnetic field gradient ($\nabla B_m$) of the permanent magnets. This generates a perturbative force ($F_s \approx \frac{\chi}{\mu_0} B_s \nabla B_m$) that drives the motion of the graphite plate ($y_g$). This motion is mechanically transferred to the SiN membrane. While the graphite's motion is too weak to be directly detected at radio frequencies, the high-$Q$ nanomechanical resonator provides resonant amplification of this motion. The resulting displacement of the membrane is detected via an optical interferometer with balanced detection in a vacuum chamber ($<5 \times 10^{-7}$ mbar) to suppress air damping.

Key Results

• Dynamic Characterization: The SiN membrane resonators exhibited multiple mechanical modes. For a representative device with a side length of $\sim 460$ $\mu$m, the first three modes were identified, with frequency ratios ($f_{12}/f_{11}$ and $f_{21}/f_{11}$) matching analytical expectations for high-stress square membranes.
• Magnetic Response: The system demonstrated a clear response to external radio-frequency magnetic fields. Resonance peaks appeared only when the magnetic drive was active, confirming that the response originated from the levitated graphite plate. The system operated over a broadband frequency range from 300 kHz to 1.6 MHz by varying membrane sizes.
• Sensitivity: Operating at room temperature without magnetic shielding, the system achieved a peak magnetic field sensitivity of 4.5 pT/$\sqrt{\text{Hz}}$ at a resonance frequency of 443 kHz with a mechanical $Q$ of $6 \times 10^6$.
• Noise Analysis: The sensitivity was determined to be limited by thermomechanical noise. The measured displacement noise spectrum exceeded the measurement imprecision, confirming that the resonator motion was dominated by thermal fluctuations. The minimum detectable magnetic field amplitude for device D6 was 45 pT.
• Stability: Allan deviation analysis over a 5-minute interval showed that relative amplitude fluctuations decreased with increasing drive amplitude, achieving a minimum fluctuation of 0.1% at $\tau \approx 0.2$ s.

Significance and Claims
The paper claims that this hybrid architecture establishes a new class of high-sensitivity magnetometers that operate at room temperature without magnetic shielding. By physically separating the magnetic transduction (diamagnetic graphite) from the mechanical amplification (high-$Q$ SiN resonator), the system avoids the need to functionalize the ultrahigh-$Q$ resonator with magnetic materials, thereby preserving its low dissipation.

The authors state that the current sensitivity is limited by thermomechanical noise and that further improvements in the mechanical $Q$ factor (e.g., reaching $10^9$) could enable sensitivities on the order of 10 fT/$\sqrt{\text{Hz}}$. This would place the performance of such levitated nanomechanical magnetometers on par with state-of-the-art atomic magnetometers and superconducting quantum interference devices (SQUIDs), but with the distinct advantage of operating under ambient conditions. The work addresses the central challenge of coupling high-$Q$ resonators to external signals without compromising their mechanical dissipation, offering a modular pathway for precision sensing.