A new arXiv preprint describes a modeled route for turning small trapped ensembles of NaCs molecules into single-molecule sites. Under the modeled spilling conditions, it estimates more than 99% isolation for a single site when the interaction shift exceeds 0.5 kHz, and 95% fidelity per tweezer across an array initially containing two molecules per site. These are model estimates, not measured loading outcomes.
The proposal starts with a small, stochastic number of polar molecules in each optical-tweezer site. Its aim is to remove the extras while leaving one molecule behind in the trap.
The proposed sequence
The modeled protocol first combines an x-polarized microwave field with static electric-field shielding. It then applies a linear electric-field gradient, allowing molecules to spill from the trap. The quantitative NaCs example uses a static field of 2.366 kV/cm.
The shielding analysis uses close-coupling scattering, a calculation that follows multiple molecular collision pathways. It reports no field-linked bound states for arbitrary microwave polarization ellipticity, including the practical linearly polarized limit. In the case represented by ξ = 45°, resonances appear at roughly 0.23 and 0.5 in the ratio of microwave strength to detuning, with an interval between them free of those states.
A calculated shielding window
At a modeled collision energy of 100 nK, with the microwave strength set to 0.4 times the detuning and the detuning set to the Förster energy defect, the extracted result reports an elastic-to-loss rate ratio above 106 when the microwave frequency is below 10 MHz. The supplied notation does not preserve the number's exponent formatting, so its exact scale is uncertain; the result also depends on collision energy and short-range assumptions.
Making one molecule stay
To set the timing of the spill, the model calculates the energy shift produced by two molecules in one trap using numerical imaginary-time evolution on a three-dimensional grid. The maximum computed shift is 0.6372 kHz at a microwave-strength-to-detuning ratio of 0.3; the analysis uses 0.55 kHz as a conservative representative value.
For a doubly occupied tweezer, the isolation calculation treats first-molecule tunneling, remaining-molecule tunneling and two-body loss as three competing exponential processes. With those assumptions, it uses a two-body loss timescale of about 6.063 seconds and finds nearly four orders of magnitude separating the two tunneling times. The model predicts more than 99% single-molecule isolation at the optimal spill time when the interaction shift is above 0.5 kHz.
From one trap to an array
At array level, the model estimates more than 99% fidelity for a single isolated site and 95% fidelity per tweezer across an initially doubly occupied array. For NaCs, the modeled shielding window spans ±2.8 V/cm around the center field value. That tolerance is a calculation-based projection, not a measured operating range.
What remains unresolved
The work remains a theoretical protocol: it reports no experimental implementation, measured molecular lifetime or measured loading fidelity. It also assumes that multi-molecule ensembles have already been prepared, while loss at higher occupancy is estimated with two-body scattering proxies rather than a full many-body calculation.
The spilling analysis assumes uncorrelated sequential tunneling and loss with exponential event times, and the array analysis assumes a constant gradient throughout the procedure. Whether the approach extends beyond the modeled NaCs case remains un demonstrated; experimental validation, fuller many-body calculations, time-dependent gradient control and tests with other dipolar species remain open questions.
Paper data and sources
Original title: A Protocol for Shielding-Enhanced Loading of Single Polar Molecules into Optical Tweezers
Authors: Reuben R. W. Wang, Christian H. Nunez, Conner Williams et al.
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-20
DOI: Not available
Original paper · Full text