arXiv:2609.24147
Reconstructing the phonon distribution of trapped ions out of the Lamb-Dicke regime
Christophe H. Valahu, Prachi Nagpal, Teerawat Chalermpusitarak, Maverick J. Millican, Cameron McGarry, Frank Scuccimarra, Vassili G. Matsos, Hon-Kwan Chan, Ting Rei Tan
Abstract
Characterizing the phonon distribution of trapped-ion motional states is essential for many applications in quantum information processing, but existing methods become ill-conditioned beyond the Lamb-Dicke regime. We overcome this limitation by recasting phonon-state reconstruction as a filter-function inversion problem. Using composite pulses on the carrier and multiple sidebands, we engineer well-conditioned filters in phonon space to characterize pure and mixed states, study heating dynamics, and reconstruct Fock states up to $n=250$.
AI Summary
The paper addresses reconstruction of trapped-ion phonon-number distributions beyond the Lamb–Dicke regime, where nonlinear, nearly degenerate spin–motion couplings make conventional sideband-Rabi measurements poorly conditioned. It formulates measurements as $M_j=\sum_n F_j(n)P(n)$ and designs the filter matrix using composite pulses and multiple sideband orders. Seven-segment phase-modulated pulses create narrowband responses; greedy A-optimal selection chooses measurement settings, and constrained least squares estimates nonnegative populations. This reconstructs number distributions, not motional coherences. Experiments use a single $^{171}$Yb$^+$ ion with a 1.40 MHz radial mode and Lamb–Dicke parameter $\eta\simeq0.085$. A common 100-filter family combining carrier, first-blue-sideband and second-blue-sideband measurements reconstructs target Fock states across $0\leq n<100$, outperforming the tested primitive first-sideband protocol. Separately optimized ten-level windows characterize Fock states up to $N=250$, prepared using sequential sideband pulses with interleaved, postselected spin measurements. Thermal and coherent distributions are also reconstructed. Delay-dependent measurements of $N=10$ and $N=100$ states agree with a calibrated gain/loss model predicting $d\mathbb{E}[n]/dt=\gamma_h$ and initial variance growth $\gamma_h(2N+1)$, with $\gamma_h=0.2\,\mathrm{s}^{-1}$. Performance remains range- and calibration-dependent. Pulses are limited to 3 ms, while higher-order sidebands are restricted by weak coupling and decoherence. Conditioning worsens for broader ranges and higher occupations, reaching $8.6\times10^4$ for $n\in[246,255]$. Quantum projection noise and uncertainty in $\eta$ dominate the highest-occupation reconstruction. Broad thermal/coherent reconstructions require smoothness regularization and retain secondary lobes and discrepancies in reconstructed moments.
Central claim may not hold: The demonstrated filter matrices include condition numbers approaching 10^5, so the paper establishes useful reconstruction improvements rather than a general resolution of ill-conditioning. The n=250 result uses an assumed ten-bin support window [246,255], limiting what it establishes about reconstruction of unknown high-excitation distributions. The supplementary reconstructed moments show substantial discrepancies from several thermal and coherent targets, contradicting the main text's unqualified claim of good agreement with all targets. The claim of reliable reconstruction for arbitrary states over arbitrary phonon ranges lacks a general distinguishability or conditioning guarantee and exceeds the demonstrated finite-range results.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.