Search arXivSearch

arXiv · 1908.02600

Purifying electron spectra from noisy pulses with machine learning using synthetic Hamilton matrices

Abstract

Photo-electron spectra obtained with intense pulses generated by free-electron lasers through self-amplified spontaneous emission are intrinsically noisy and vary from shot to shot. We extract the purified spectrum, corresponding to a Fourier-limited pulse, with the help of a deep neural network. It is trained on a huge number of spectra, which was made possible by an extremely efficient propagation of the Schrödinger equation with synthetic Hamilton matrices and random realizations of fluctuating pulses. We show that the trained network is sufficiently generic such that it can purify atomic or molecular spectra, dominated by resonant two- or three-photon ionization, non-linear processes which are particularly sensitive to pulse fluctuations. This is possible without training on those systems.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Sajal Kumar Giri, Ulf Saalmann, Jan M. Rost. 2020-03-12. Purifying electron spectra from noisy pulses with machine learning using synthetic Hamilton matrices. https://doi.org/10.1103/physrevlett.124.113201

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

$LS/LSJ$ Hybrid Coupling Framework for Auger Angular Distributions of Experimentally Unresolved Multiplets with Isolated Fine-Structure

We present a hybrid $LS/LSJ$ coupling framework for treating state multiplets in the regime $ΔE_\text{int}\sim\hbar/τ_\text{int} \ggΔE_\text{FS}\ggΓ$, where $τ_\text{int}$ is the collision interaction time, $ΔE_\text{FS}$ the fine-structure splitting, and $Γ$ the natural width. In this regime, the collision interaction is fast compared with fine-structure evolution, whereas the individual $J$ levels are well isolated on the scale of their natural widths. The produced multiplet's alignment is therefore described in $LS$ coupling and then projected onto the individual fine-structure $J$ levels before their subsequent decay, described by $LSJ$ coupling. Applied to the multi-open-shell $1s2s2p\,^4\!P_J$ manifold, which closely satisfies these conditions, the hybrid treatment reveals strong suppression and inversion of the Auger angular-distribution anisotropy relative to the traditional pure-$LS$ treatment. It substantially improves agreement with absolute experimental data while using the same $LS$ production cross sections. This hybrid $LS/LSJ$ framework thus provides the appropriate treatment of state multiplets with isolated fine-structure levels, as it retains the $J$-dependent decay dynamics missing from the traditional pure-$LS$ treatment.

physics.atom-ph

A Protocol for Shielding-Enhanced Loading of Single Polar Molecules into Optical Tweezers

We propose the high-fidelity preparation of single bosonic molecules in optical tweezers starting from small tweezer-trapped molecular ensembles. Our scheme combines a static electric field and a microwave field to generate strong, tunable, anisotropic interactions that shield the molecules against two-body collisional loss. We show that this shielding eliminates all long-range bound states, preventing three-body recombination. This elimination persists for all microwave ellipticities, including the experimentally practical limit of linear polarization. Application of an additional electric field gradient can be used to induce controlled spilling of strongly interacting molecules out of the trap until one remains. With realistic experimental parameters, we estimate that single tweezer-trapped NaCs molecules can be isolated from a pair with fidelities exceeding 99\%, and $> 95\%$ per site across an array. These results establish collisional shielding with electric fields as an effective tool for preparing highly-filled tweezer arrays of polar molecules.

physics.atom-ph

Vector Measurements Using Integrated Radio Frequency Atomic Magnetometers

We demonstrate reconstruction of three-dimensional radio-frequency (RF) magnetic-field vectors using a pair of integrated RF atomic magnetometers operated with orthogonal bias-field orientations. A theoretical and experimental analysis identifies a phase-ambiguity dead band that limits reconstruction when the two sensor responses become nearly identical. Measurements performed in an unshielded laboratory environment demonstrate accurate reconstruction of RF magnetic-field orientations and validate the predicted dependence of reconstruction accuracy on signal imbalance. These results establish integrated RF atomic magnetometers as a compact and sensitive platform for directional RF magnetic-field sensing, particularly at low frequencies, and provide a foundation for portable source-localization and field-mapping applications.

physics.atom-ph