Skip to content

Modeling and in situ probing of surface reactions in atomic layer deposition

Corpus note

pdf_path points to papers/Zheng_Hong_ACS_AMI_ALD_2017_proof.pdf, an ACS author proof. The version-of-record narrative with stable figure numbering is on 2017yuanxia-zheng-acs-am7b01618 (and 2017zheng-venue-research). See NON_PRIMARY_ARTICLE_PAPER_SLUGS.md for proof-duplicate policy.

Summary

The underlying ACS AMI article investigates atomic layer deposition (ALD) of alumina from trimethylaluminum and water on germanium surfaces that are either hydrogen-terminated or oxidized, a stack relevant to high-κ gate stacks and interface trap control in semiconductor processing. Zheng et al. combine ReaxFF reactive molecular dynamics with in situ spectroscopic ellipsometry (alongside AFM, XPS, and impedance measurements) to study trimethylaluminum / water atomic layer deposition of alumina on Ge(100) substrates prepared in hydrogen-terminated versus oxidized states. The scientific goal is to connect early-cycle surface chemistry—nucleation, suboxide intermixing, and trap-related electrical response—to atomistic models of precursor adsorption, ligand elimination, and oxygen insertion. This wiki slug holds the proof PDF bytes; the canonical prose for the same DOI 10.1021/acsami.7b01618 lives on [[2017yuanxia-zheng-acs-am7b01618]].

Methods

Molecular dynamics (reactive). This slug tracks the ACS author proof bytes (papers/Zheng_Hong_ACS_AMI_ALD_2017_proof.pdf); the ReaxFF molecular dynamics story matches [[2017yuanxia-zheng-acs-am7b01618]]: TMA/H₂O ALD on H–Ge(100) versus oxidized Ge(100) inside periodic supercells with explicit atom counts, timestep (fs), thermostat/barostat settings, NVT/NPT staging, temperature (K) setpoints, and equilibration/production duration (ps/ns) as published in the version-of-record. Electric fields and metadynamics/umbrella enhanced sampling are not highlighted here.

Experiments. In situ ellipsometry, AFM, XPS, and impedance protocols mirror the canonical article; use [[2017yuanxia-zheng-acs-am7b01618]] for figure-quality references.

Force-field fitting. N/A — identical published Al/Ge/O/H ReaxFF consumption as the canonical page.

Static QM / DFT. N/A — not the primary engine for the coupled workflow summarized here.

Review scope. N/A — proof PDF duplicate for cataloguing alternate hashes.

Findings

Outcomes and mechanisms. Reactive trajectories again tie delayed nucleation on H–Ge to sluggish ligand elimination, whereas oxidized templates undergo self-cleaning that seeds an Al₂O₃/GeO_x intermixed layer moderating oxygen diffusion into Ge.

Comparisons. Ellipsometry/AFM/XPS/impedance benchmarks on the proof PDF should be cross-checked against [[2017yuanxia-zheng-acs-am7b01618]] before citing panel letters or Supporting Information pointers.

Sensitivity / design levers. Surface termination (hydrogen vs native oxide) remains the central lever controlling early-cycle growth transients.

Limitations / outlook. Proof files may contain layout queries or non-final typography—confirm against the issue PDF before archival citation.

Corpus honesty. Prefer [[2017yuanxia-zheng-acs-am7b01618]] for reader-facing science; keep this slug for duplicate PDF provenance only.

Limitations

Proof PDFs may include layout artifacts, color query marks, or non-final panel labels. Pagination and Supporting Information numbering must be verified against [[2017yuanxia-zheng-acs-am7b01618]] before external citation. Figure lettering in proofs sometimes differs from XML typesetting, so SI cross-references should be checked against the publisher PDF when building bibliographies.

Confidence rationale: med—duplicate proof; science summarized by pointer to VOR.

Reader notes (navigation)