Skip to content

New many-body potential for the bond order

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

This Physical Review Letter derives an analytic many-body potential for the bond order in s-valent systems so that local atomic environment directly modulates bond strength. The formulation uses recursion/Green’s-function ideas: bond order is expressed via an integral involving bonding and antibonding Green’s functions, with three- and four-membered ring (path) contributions controlling differences between bonding and antibonding recursion coefficients. The work motivates extending embedded-atom style models with explicit three-body and four-body terms tied to bond-order physics, addressing failures of bulk-fitted potentials on small clusters.

Methods

This Physical Review Letter derives bond order for s-valent systems from a recursion / tridiagonal Hamiltonian picture: bond order is an integral over bonding vs antibonding Green’s functions written as continued fractions, with three-membered-ring and four-membered-ring path contributions entering at successive recursion levels as differences between bonding and antibonding coefficients. A Dyson / perturbation expansion about an average reference Green’s function yields an explicit many-body bond-order expression involving response functions \(Z_{p,n}\) for a reference semi-infinite linear chain; an analytic limiting case uses zero on-site energies and constant hopping \(b\) with Fermi energy fixed by electron count per atom \(N\). The formalism is illustrated for how environment reduces bond order relative to an isolated dimer and how cluster topology (e.g., four-atom clusters) selects among bonding patterns.

Findings

The Letter provides an analytic many-body bond-order potential for s-valent systems in which local atomic environment directly modulates bond strength, addressing failures of bulk-fitted potentials on small clusters. It demonstrates that three- and four-membered-ring terms control differences between bonding and antibonding recursion coefficients at low orders, and it discusses reduced response functions versus electrons per atom (Figure 1 in the Letter). The work positions bond-order concepts as a bridge from tight-binding/recursion intuition toward empirical many-body models used in MD, explicitly citing Tersoff-type and Chelikowsky-style bond-order constructions as related context.

Comparisons / outlook. The discussion contrasts embedding / second-moment rationales with the need for explicit three-body and four-body terms in open tetrahedral semiconductors, and cites prior bond-order fits to Si that successfully span bulk and cluster geometries.

Corpus honesty. This page is grounded in pdf_path and normalized/extracts/2011pettifor-venue-new-many-body_p1-2.txt (PRL letter text); any extended citations beyond the excerpt require the PDF.

Outlook. However, the Letter’s explicit scope is s-valent toy models and analytic structure; quantitative MD benchmarks are not the subject of these pages.

Limitations

  • s-valent focus; transfer to transition metals or polar systems requires extensions not contained in this short Letter.
  • Practical classical potentials still require parametrization choices when implementing bond-order ideas in MD engines.

Relevance to group

Historical bond-order foundation relevant to ReaxFF and Brenner/Tersoff lineages used throughout the corpus.

Citations and evidence anchors

  • reaxff-family
  • Bond-order and empirical many-body potentials (lineage context)