Optically Levitated Nanoparticles

• Optically levitated nanoparticles are dielectric or metallic particles (50–500 nm) trapped in vacuum using focused laser beams, serving as a platform for precision metrology and quantum optomechanics.
• Advanced feedback control techniques, including electrical, parametric, and cavity-optomechanical methods, enable effective cooling and near-ground-state occupation of both translational and rotational modes.
• Innovative trap engineering and dual-wavelength schemes facilitate detailed studies of dipolar scattering, nonlinear dynamics, and tunable nanoparticle interactions for advanced sensing applications.

Optically levitated nanoparticles are dielectric or (less commonly) metallic particles, typically with dimensions of 50–500 nm, that are trapped and manipulated in vacuum using the gradient force of a tightly focused laser beam. In the Rayleigh regime—when the particle radius is much smaller than the wavelength—the particle acts as an induced electric dipole, allowing for a versatile mechanical oscillator platform that is nearly free from thermal, clamping, and substrate-related dissipation. This system has become vital for force and torque sensing, precision metrology, quantum optomechanics, and studies of macroscopic quantum phenomena, with an extensive toolbox spanning trap engineering, state readout and control, and ultra-sensitive environmental coupling.

1. Principles of Optical Trapping and Mechanical Motion

A tightly focused laser beam (“optical tweezer”) generates a three-dimensional potential for a dielectric nanoparticle through the gradient force acting on its induced dipole (Jin et al., 2021). For a sphere of radius $r$ and permittivity $\varepsilon$, the key forces are

• Gradient force: $$\mathbf{F}_{\mathrm{grad}} = \frac{1}{2} \mathrm{Re}[\alpha] \nabla |\mathbf{E}|^2$$
• Scattering (radiation pressure) force: $$F_{\mathrm{scat}} = \frac{k^4 |\alpha|^2}{6\pi\varepsilon_0^2 c} I$$

The polarizability $\alpha$ in the Rayleigh regime is given by

$$\alpha = 4\pi\varepsilon_0 r^3 \frac{\varepsilon - 1}{\varepsilon + 2}$$

where $I$ is the optical intensity and $c$ is the speed of light. High numerical aperture (NA) objectives (e.g., NA = 0.9–0.95) provide steep intensity gradients and enable strong three-dimensional confinement, with trap frequencies for a 100–150 nm silica sphere in the range of 30–300 kHz.

In high vacuum (pressures ≪1 mbar), the gas damping rate drops drastically, and internal heating/cooling, photonic recoil, and feedback forces become dominant. The optical potential near the focus is approximately harmonic, $$U(\mathbf{r}) \approx \frac{1}{2} m (\omega_x^2 x^2 + \omega_y^2 y^2 + \omega_z^2 z^2)$$, with sub-nanometer thermal oscillation amplitudes achievable under high trap stiffness (Jin et al., 2021, Diehl et al., 2018).

2. Scattering, Imaging, and Dipolar Radiation Control

The Rayleigh-scattering regime simplifies the optical response to that of a point dipole with well-defined orientation. Dually, the system enables precise studies of dipole scattering patterns, anisotropy, and emission in free space (Jin et al., 2021). Dual-wavelength schemes, with a trapping beam (e.g., 1064 nm) and an orthogonal probe (e.g., 532 nm), allow independent control of trapping and excitation.

By adjusting the probe polarization, the induced dipole $$\mathbf{p} = \varepsilon_0 \alpha \mathbf{E}_{\mathrm{inc}}$$ can be rotated in the laboratory frame. Far-field detection in both image and Fourier (k)-space captures the angular and polarization-resolved dipolar emission, with the measured intensity following

$$I(\theta, \phi) = \frac{3}{8\pi r^2} I_{\mathrm{tot}} [1 - (\hat{\mathbf{p}} \cdot \hat{\mathbf{r}})^2]$$

and the k-space image directly encoding the dipole orientation and emission anisotropy.

A key platform advantage is absence of particle–substrate interaction, yielding a pristine mapping of the 4π angular scattering distribution. Vectorial phenomena, such as polarization vortices (detected when the dipole is aligned along the objective axis), and the observation of Kerker conditions—zero forward or backward scattering by electric–magnetic dipole interference—can be studied directly with full orientation and wavelength control (Jin et al., 2021).

3. Cooling, Feedback, and Control Strategies

Feedback control is fundamental for stabilizing optically levitated nanoparticles in the ultra-high-Q regime. Three major classes are implemented:

• Electrical feedback: Coulomb force is applied by external electrodes. 3D feedback cooling via an optimal linear–quadratic regulator (LQR) reaches sub-Kelvin temperatures for all center-of-mass modes and, in quantum-limited regimes, can achieve phonon-number occupation ⟨n⟩ < 1 (Kremer et al., 2024).
• Parametric feedback/cold-damping: The optical trap stiffness is modulated in real time according to the measured position/velocity, producing enhanced cooling rates and enabling sub-quanta ground-state occupations (Tebbenjohanns et al., 2021).
• Cavity–optomechanical and coherent-scattering cooling: A high-finesse cavity allows for coherent exchange between mechanical motion (both translation and rotation) and the optical field. Elliptic coherent scattering enables simultaneous cooling of all six degrees of freedom, with translational temperatures in the 100 μK regime and librations at ~5 mK (Pontin et al., 2022).

The feedback efficiency is ultimately limited by the detection quantum efficiency and feedback bandwidth. With state-of-the-art methods, occupation numbers as low as 0.65 have been realized for the center-of-mass mode in cryogenic, cavity-free “free-space” traps (Tebbenjohanns et al., 2021), allowing for studies of macroscopic quantum mechanics and matter-wave interference.

4. Multi-Mode and Rotational Dynamics

Nonspherical (ellipsoidal or dumbbell) nanoparticles exhibit rotational and torsional optomechanics in addition to translation:

• Torsional vibration: A nanoscale ellipsoid with tensor polarizability in a polarized Gaussian tweezer experiences an angular restoring torque, with typical eigenfrequencies Ωθ an order of magnitude higher than CoM motions (Ωθ/2π ~ 1 MHz vs Ω_CoM/2π ~ 0.1–0.2 MHz). This enables resolved-sideband ground-state cooling of torsional modes and torque sensitivities at the $10^{-29}~{\rm N}\cdot{\rm m}/\sqrt{\rm Hz}$ level (Hoang et al., 2016).
• GHz rotations: Nanodumbbells can be driven to GHz mechanical rotation by circularly polarized trapping fields, probing quantum friction and Casimir torques at sub-micrometer separations with torque sensitivity below $10^{-25}~{\rm N}\cdot{\rm m}/\sqrt{\rm Hz}$ (Ju et al., 2023).
• Six-dimensional (full translational + rotational) cooling: Simultaneous control of all six degrees of freedom is achieved via elliptic scattering into a cavity, demonstrating orientational stability at sub-μrad accuracy (Pontin et al., 2022).

Arrays can now be assembled with site-selective control, enabling on-demand synthesis of compound objects (e.g., nanodumbbells) and studies of coupled-oscillator physics (Yan et al., 2022).

5. Coupling, Collective Effects, and Advanced Trap Engineering

Advances in optical trap engineering permit the assembly and coherent control of interacting particle systems:

• Tunable coupling via third-particle “coupler”: Introducing a third optically trapped nanoparticle enables phase- and position-tunable dipole–dipole coupling between two otherwise uncoupled particles, controllable via their optical phase and polarization. This establishes a pathway toward programmable coupled mechanical arrays, synthetic exceptional points, and complex interaction Hamiltonians (Wu et al., 2024).
• Metasurface-based traps: Application of dielectric metasurfaces provides on-chip, multi-focal potential landscapes with high-NA and high efficiency (31%), enabling bistable and double-well topologies for trapping and manipulating multiple nanoparticles over hours with full electronic control of well depth and separation (Sun et al., 2024).
• Nonlinear and stochastic dynamics: The nonlinear, Duffing-type response at higher motional energies enables studies of stochastic bistability, Kramers escape, and stochastic resonance, supporting force and displacement sensing at the zeptonewton scale and dynamic processes such as amplitude-based memory and threshold detection (Ricci et al., 2017, Ge et al., 2016).

Coherent modulations of the trapping potential also realize classical analogs of two-level atom physics—Autler–Townes splitting, Rabi flopping, and Bloch-sphere dynamics, including PT-symmetric Hamiltonians and phonon lasing in coupled modes (Frimmer et al., 2017, Sharma et al., 2024).

6. Surface Proximity, Charge Control, and Thermal Interactions

Nanoparticles can be positioned at controlled, sub-wavelength distances from surfaces, facilitating precision studies of Casimir–Polder forces, patch-potential effects, and particle-surface optomechanical coupling. Interferometric back-focal-plane imaging quantifies nanoparticle–surface separations with nanometer-scale precision (Diehl et al., 2018).

Charge neutrality is paramount: any net charge introduces decoherence via fluctuating Coulomb forces, reducing coherence times and masking subtle environmental couplings. Single-electron accuracy in charge control is established by in-situ glow-discharge and lock-in force detection protocols (Frimmer et al., 2017). Charge neutrality enables free-fall trap-to-trap experiments without sensitivity to stray electric fields, opening new avenues for matter-wave interferometry and quantum-state engineering (Mattana et al., 17 Jul 2025).

Advanced schemes further enable thermometric inference of internal nanoparticle temperature via displacement detection of dipole–thermal-image coupling in the presence of infrared-reflective surfaces, revealing quantum limits to decoherence and internal state control (Agrenius et al., 2022).

7. Materials Platform and Future Directions

The choice of nanoparticle material determines both fundamental limits and application scope:

• Silica (SiO₂): Widespread for low absorption and high optical stability. Capable of sub-Kelvin cooling and high mechanical Q.
• Silicon carbide (SiC): Recently demonstrated for stable optical levitation with active fluorescent quantum defects (carbon antisite-vacancy pairs), preserving single-photon emission in vacuum and offering prospects for hybrid quantum–spin–mechanical systems (Alavi et al., 25 Apr 2025).
• Diamond (nanodiamonds): Nitrogen-vacancy (NV) centers and related defects are attractive for spin–optomechanical integration, but high absorption can cause thermal instability in vacuum. This is mitigated by core–shell engineering and doughnut-beam trapping to minimize optical heating (Zhou et al., 2016).

Next-generation directions include ground-state cooling without optical cavities; quantum-limited displacement and torque sensing at the 10^{-21}–10^{-29} N·m/√Hz regime; programmable assembly of large-scale coupled arrays for quantum simulation; and hybrid devices interfacing spin, charge, optical, and mechanical degrees of freedom.

---

References:

• "Imaging the dipole scattering of an optically levitated dielectric nanoparticle" (Jin et al., 2021)
• "Optical potential mapping with a levitated nanoparticle at sub-wavelength distances from a membrane" (Diehl et al., 2018)
• "All electrical cooling of an optically levitated nanoparticle" (Kremer et al., 2024)
• "Controlling the net charge on a nanoparticle optically levitated in vacuum" (Frimmer et al., 2017)
• "Torsional optomechanics of a levitated nonspherical nanoparticle" (Hoang et al., 2016)
• "PT Symmetry, induced mechanical lasing and tunable force sensing in a coupled-mode optically levitated nanoparticle" (Sharma et al., 2024)
• "Simultaneous cooling of all six degrees of freedom of an optically levitated nanoparticle by elliptic coherent scattering" (Pontin et al., 2022)
• "Near-field GHz rotation and sensing with an optically levitated nanodumbbell" (Ju et al., 2023)
• "Interaction Between an Optically Levitated Nanoparticle and Its Thermal Image: Internal Thermometry via Displacement Sensing" (Agrenius et al., 2022)
• "Tunable on-chip optical traps for levitating particles based on single-layer metasurface" (Sun et al., 2024)
• "Quantum control of a nanoparticle optically levitated in cryogenic free space" (Tebbenjohanns et al., 2021)
• "On-demand assembly of optically-levitated nanoparticle arrays in vacuum" (Yan et al., 2022)
• "Optically levitated nanoparticle as a model system for stochastic bistable dynamics" (Ricci et al., 2017)
• "Force sensing with an optically levitated charged nanoparticle" (Hempston et al., 2017)
• "Trap-to-trap free falls with an optically levitated nanoparticle" (Mattana et al., 17 Jul 2025)
• "Feedback-induced Bistability of an Optically Levitated Nanoparticle: A Fokker-Planck Treatment" (Ge et al., 2016)
• "A levitated nanoparticle as a classical two-level atom" (Frimmer et al., 2017)
• "Coupler enabled tunable dipole-dipole coupling between optically levitated nanoparticles" (Wu et al., 2024)
• "Optical levitation of fluorescent silicon carbide nanoparticles in vacuum" (Alavi et al., 25 Apr 2025)
• "Optical Levitation of Nanodiamonds by Doughnut Beams in Vacuum" (Zhou et al., 2016)