Illuminating surface atoms in nanoclusters by differential X-ray absorption spectroscopy
Evidence and attribution¶
Authority of statements
Prose sections below (Summary, Methods, Findings, etc.) are curated summaries of the publication identified by doi, title, and pdf_path in the front matter above. They are not new primary claims by this wiki.
For definitive numerical values, reaction schemes, and interpretations, use the peer-reviewed article (and optional records under normalized/papers/ when present)—not this page alone.
Summary¶
Differential EXAFS (D-EXAFS) tracks Ar-induced structural change on silica-supported Pd clusters (~1 nm). Difference analysis indicates Pd–Pd nearest-neighbor distances expand by ~0.104 Å upon Ar adsorption at 77 K. Atomistic MD supports a picture where under-coordinated Pd atoms, kinetically trapped at 77 K, restructure when Ar supplies enough energy to surmount barriers—without a net change in nearest-neighbor coordination number at the level discussed. Senftle, Janik, and van Duin contribute simulation interpretation alongside Frenkel/Rioux spectroscopy.
Methods¶
Local sources: the PDF at papers/Spanjer_PCCP_Pd_cluster_2014.pdf is present in this workspace; opening sections are captured in normalized/extracts/2014spanjers-physical-che-illuminating-surface_p1-2.txt.
Experiment: 3% Pd/SiO\(_2\) catalysts are prepared by strong electrostatic adsorption (SEA) from Pd(NH\(_3\))\(_4\)(NO\(_3\))\(_2\) onto Davisil A60 silica at pH 11, followed by filtration, drying (398 K), and reduction (438 K, 4% H\(_2\)/He, 1 h). Pd K-edge XAS is measured at APS 10-BM in transmission; samples are treated in a custom in situ cell (ESI), reduced, He-purged, cooled to ~77 K, then measured under flowing He (“clean”) and under Ar + He (“after Ar adsorption”) at an Ar partial pressure of order ~100 Torr (stated as yielding roughly one statistical monolayer of Ar on silica/Pd based on BET estimates). Data are processed/analyzed with IFEFFIT as stated.
Simulation. Classical atomistic molecular dynamics—with force field, MD engine, NVT/NPT staging as applicable, timestep, equilibration/production run lengths (ps/ns), thermostat coupling, and periodic (PBC) treatment documented in papers/Spanjer_PCCP_Pd_cluster_2014.pdf and the ESI (N/A for program name and numeric protocol on the indexed p1–2 extract)—is used to discriminate surface-restructuring models consistent with the D-EXAFS difference signal for ~1 nm SiO\(_2\)-supported Pd clusters containing on the order of 10\(^2\)–10\(^3\) atoms in the published models (N/A for exact counts on this page). Pressure/barostat usage follows whichever ensemble the authors specify for each stage (N/A to transcribe numerically here).
Force-field training. N/A: parameters come from the literature parametrization cited in the article; no new FF fit is reported.
Static QM. N/A as headline production QM: the contribution combines XAS with classical MD model discrimination.
Findings¶
D-EXAFS (EXAFS after minus before Ar) isolates surface-weighted structural change on ~1 nm SiO\(_2\)-supported Pd. Difference analysis gives a first-shell Pd–Pd expansion of 0.104 ± 0.005 Å after Ar adsorption at 77 K. MD supports a mechanism in which under-coordinated Pd, kinetically trapped at 77 K, restructure when Ar provides enough energy to cross restructuring barriers, lengthening surface Pd–Pd distances without a net change in overall nearest-neighbor coordination number at the level discussed—supporting the claim that D-EXAFS can illuminate surface atoms where conventional EXAFS averages bulk and surface. TEM, alternative models, and modeling caveats are in the PCCP article and ESI.
Limitations¶
- Projected TEM-like averaging and cluster dispersity affect interpretation; force field choices in MD bound mechanistic detail.
Relevance to group¶
Connects Penn State catalysis/spectroscopy collaboration networks (Senftle, Janik, van Duin) to operando-style structure probes for nanoparticle catalysts.
Citations and evidence anchors¶
- DOI: https://doi.org/10.1039/c4cp02146k