[Paper Review] Surface science motivated by heating of trapped ions from the quantum ground state

D. A. Hite, Kyle S. McKay+1 more|arXiv (Cornell University)|Apr 15, 2021

Quantum Information and Cryptography49 references9 citations

TL;DR

This study investigates anomalous nonmonotonic heating in trapped ions caused by electric-field noise from electrode surfaces, linking it to surface potential roughness and contaminant coverage. Using Kelvin probe force microscopy (KPFM) and XPS, it shows that Ar+ ion bombardment alters gold electrode surfaces, reducing hydrocarbon contamination and creating work-function patches; a phenomenological model correlates surface-potential roughness and coverage to nonmonotonic heating rates, explaining previously observed anomalies in ion-trap quantum computing systems.

ABSTRACT

For the past two and a half decades, anomalous heating of trapped ions from nearby electrode surfaces has continued to demonstrate unexpected results. Caused by electric-field noise, this heating of the ions' motional modes remains an obstacle for scalable quantum computation with trapped ions. One of the anomalous features of this electric-field noise is the reported nonmonotonic behavior in the heating rate when a trap is incrementally cleaned by ion bombardment. Motivated by this result, the present work reports on a surface analysis of a sample ion-trap electrode treated similarly with incremental doses of Ar$^+$ ion bombardment. Kelvin probe force microscopy and x-ray photoelectron spectroscopy were used to investigate how the work functions on the electrode surface vary depending on the residual contaminant coverage between each treatment. It is shown that the as-fabricated Au electrode is covered with a hydrocarbon film that is modified after the first treatment, resulting in work functions and core-level binding energies that resemble that of atomic-like carbon on Au. Changes in the spatial distribution of work functions with each treatment, combined with a suggested phenomenological coverage and surface-potential roughness dependence to the heating, appear to be related to the nonmonotonic behavior previously reported.

Motivation & Objective

• To understand the origin of nonmonotonic motional heating in trapped ions during incremental ion-bombardment surface treatments.
• To investigate how surface potential heterogeneity and contaminant coverage affect electric-field noise in ion-trap electrodes.
• To correlate surface science measurements (work function, binding energy, roughness) with experimentally observed nonmonotonic heating rates.
• To develop a phenomenological model linking surface-potential roughness and adsorbate coverage to heating rate anomalies.

Proposed method

• Performed incremental Ar+ ion bombardment on electroplated Au ion-trap electrodes to reduce contaminant coverage.
• Used Kelvin probe force microscopy (KPFM) to map spatial variations in work function and surface-potential roughness.
• Applied x-ray photoelectron spectroscopy (XPS) to measure core-level binding energies and confirm changes in carbon overlayer chemistry.
• Measured frequency-dependent heating rates of trapped ions to determine noise spectral density scaling (1/f^α).
• Fitted a phenomenological model: ˙¯n = C·(1−e−θ/λ)·σφ + bkg, where σφ is RMS surface-potential roughness, θ is coverage, and C, λ, bkg are fit parameters.
• Compared results to the C/Au(110) model system to assess similarity in work function and chemical state.

Experimental results

Research questions

• RQ1 How does incremental ion bombardment alter the work function and surface potential landscape of electroplated Au ion-trap electrodes?
• RQ2 Why does motional heating in trapped ions exhibit nonmonotonic behavior with increasing ion-bombardment dose, contrary to monotonic expectations?
• RQ3 To what extent do surface potential roughness and contaminant coverage correlate with measured heating rates?
• RQ4 How do the surface properties of as-fabricated Au electrodes compare to the C/Au(110) model system in terms of work function and chemical state?
• RQ5 Can a simple phenomenological model based on coverage and surface-potential roughness qualitatively explain the nonmonotonic heating trend?

Key findings

• After the first ion-bombardment treatment, the work function and C 1s binding energy of the Au electrode surface matched those of atomic carbon on Au(110), indicating transformation of hydrocarbon film to atomic-like carbon.
• KPFM revealed surface-potential patch sizes of 50–100 nm with wide distributions, indicating significant spatial inhomogeneity not explained by crystalline orientation alone.
• The power-law exponent α of the electric-field noise spectral density changed from ∼1.5 (high coverage) to ∼0.4 (low coverage), indicating a shift in dominant noise mechanism.
• Anisotropy in heating rates was observed: at 3 MHz, the mode aligned with the sputter beam experienced ∼3.5× higher noise than the perpendicular mode after four treatments.
• The model ˙¯n = C·(1−e−θ/λ)·σφ + bkg, with C = 4 (quanta/s)/meV, λ = 0.2 ML, and bkg = 4 quanta/s, qualitatively reproduced the nonmonotonic heating trend.
• At θ = 0.9 ML, the sample electrode had σφ = 42 meV (model heating rate: 170 quanta/s), while the C/Au(110) model had σφ = 11 meV (model heating rate: 48 quanta/s), showing strong roughness dependence.