First‑Principles Insights into Transition‑Metal Adsorption on Black Phosphorene: Implications for Catalysis and Spintronics

Abstract

Black phosphorene, a two‑dimensional monolayer of phosphorus, has attracted attention for its direct bandgap, high carrier mobility, and mechanical flexibility. Using spin‑polarized density functional theory, we investigated how 12 transition‑metal (TM) atoms—Fe, Co, Ni, Cu, Ru, Rh, Pd, Ag, Os, Ir, Pt, and Au—interact with phosphorene. All TM‑phosphorene complexes exhibit strong binding (adsorption energies 2–6 eV) and stable geometries, with the metal preferring the hollow site. Notably, Fe‑, Co‑, and Au‑phosphorene systems become magnetic semiconductors with magnetic moments of 2, 1, and 0.96 µ_B, respectively. When oxygen molecules adsorb on these TM‑phosphorene surfaces, most O₂–TM‑phosphorene complexes display significant O–O bond elongation, a key indicator of catalytic activity for CO oxidation. Furthermore, O₂ adsorption can convert several systems into magnetic semiconductors or half‑metallic states, highlighting their potential for spintronic devices. Our findings provide a foundation for designing phosphorene‑based catalysts and spintronic components.

Background

Phosphorene’s puckered honeycomb lattice bestows a direct bandgap (~0.9 eV) and ultrahigh carrier mobility, making it suitable for transistors, batteries, solar cells, photocatalysis, spintronics, and gas sensing. However, its inherent nonmagnetic character limits certain applications. Introducing magnetism via surface adsorption has proven effective in other two‑dimensional systems: graphene, silicene, and arsenene all acquire spin polarization when decorated with transition metals or other adatoms. Prior work on phosphorene focused mainly on 3d metals; the influence of 4d/5d elements and noble metals on both electronic structure and catalytic behavior remains largely unexplored. Our study addresses these gaps by systematically examining 12 transition metals—including ferromagnetic Fe, Co, Ni; diamagnetic Cu; and noble metals Ru, Rh, Pd, Ag, Os, Ir, Pt, Au—adsorbed on black phosphorene. We also evaluate O₂ adsorption on the TM‑phosphorene systems to assess catalytic viability for CO oxidation.

Methods

Spin‑polarized density functional theory calculations were performed with the Vienna Ab Initio Simulation Package (VASP). We employed the generalized gradient approximation (GGA) with the Perdew–Burke–Ernzerhof (PBE) functional, and incorporated van der Waals interactions via the DFT‑D3 scheme. A plane‑wave cutoff of 400 eV and a 3 × 3 × 1 Monkhorst–Pack k‑point grid were used for geometry optimizations, while a denser 7 × 7 × 1 grid was applied for total energy calculations. A 4 × 3 supercell (13.20 × 13.74 Å) with 20 Å vacuum in the z‑direction ensured negligible inter‑layer coupling. Geometry optimizations converged when forces dropped below 0.01 eV/Å and total energy changes were < 1 × 10⁻⁵ eV.

Results and Discussion

Pristine phosphorene
Phosphorene’s tetragonal lattice (a = 3.30 Å, b = 4.58 Å) contains four P atoms per unit cell. The P–P bond lengths are 2.22 Å (horizontal) and 2.26 Å (vertical). Its direct bandgap of 0.89 eV places both the conduction band minimum and valence band maximum at the Γ point, in agreement with prior theoretical studies.

TM adsorption geometry and energetics
All TM atoms favor the hollow (H) site, with adsorption energies ranging from 2 to 6 eV—indicative of robust chemisorption. The TM–P bond lengths (2.11–2.43 Å) and Bader charge analyses (0.07–0.33 e⁻ transferred to phosphorene) confirm strong covalent interactions. TM‑phosphorene complexes are generally stable and retain the phosphorene lattice.

Magnetic and electronic properties
Fe‑, Co‑, and Au‑phosphorene systems exhibit net magnetic moments of 2, 1, and 0.96 µ_B, respectively, while the remaining TM‑phosphorene complexes are nonmagnetic. Spin‑density maps reveal that magnetism originates mainly from the adatom with minor contributions from neighboring P atoms. Band‑structure calculations show that Fe‑, Co‑, and Au‑phosphorene are magnetic semiconductors with bandgaps of 0.38, 0.22, and 0.06 eV, respectively—promising for spintronic applications.

O₂ adsorption and catalytic potential
O₂ molecules adsorb on top of the TM adatom in two distinct configurations depending on the TM species. Except for Pd‑phosphorene, all O₂–TM‑phosphorene systems exhibit significant O–O bond elongation (up to 1.40 Å), a hallmark of effective CO‑oxidation catalysts. Adsorption energies (0.97–3.11 eV) confirm chemisorption for most systems. Bader charge transfer from the TM‑phosphorene to O₂ (−0.09 to −0.68 e⁻) weakens the O–O bond and activates the molecule for catalytic reactions. In the Pt‑phosphorene case, a 0.19 e⁻ transfer fills antibonding orbitals, evidenced by LDOS and charge‑density difference analyses, explaining its superior catalytic performance.

Spin properties of O₂–TM‑phosphorene
O₂ adsorption can induce or enhance magnetism. For Ni, Cu, Rh, Ag, and Ir systems, magnetic moments of 2.00, 1.00, 1.00, 1.14, and 1.00 µ_B arise largely from the paramagnetic O₂ molecule. Fe, Co, and Au complexes show combined contributions from the TM and O₂. Band structures reveal flat bands near the Fermi level, characteristic of localized O₂ states. While most O₂–TM‑phosphorene complexes become magnetic semiconductors, Co‑phosphorene uniquely exhibits half‑metallicity, opening avenues for spin‑polarized transport.

Conclusions

We demonstrated that 12 transition metals bind strongly to the hollow site of black phosphorene, forming stable complexes. Fe, Co, and Au adatoms convert phosphorene into magnetic semiconductors with sizable magnetic moments. O₂ adsorption on these TM‑phosphorene surfaces generally activates the O–O bond, yielding promising catalysts for CO oxidation, except for the Pd case. Additionally, O₂ adsorption can tune the magnetic state—producing magnetic semiconductors or half‑metallicity—making these systems attractive for next‑generation spintronic devices. Our work provides a theoretical framework for engineering phosphorene‑based catalysts and spin‑functional materials.