Local density of states in amorphous Ni-P alloys

Abstract

The electronic density of states around the P and Ni atoms in amorphous alloys NixP1-x for different x(0.08<x<0.26) is investigated using X-ray emission. The Ni d band of the alloy is shifted to higher binding energies by 0.2 eV with respect to that of the pure Ni. The P 3s band becomes more localized and strongly bonded at a binding energy of -13.1 eV. The P 3p band is hybridized with the Ni d band and forms bonding and antibonding bands. The antibonding p band is located at +6.4 eV above the Fermi level, whereas the bonding p band is at-5.9 eV. The d densities of states around the P atoms show special features with respect to those in pure P. These new P d states (at -2.5 eV binding energy, including 0.3 d electrons) come from Ni d states. An amount of 0.3-0.4 electrons (per P atom) is transferred from the Ni atoms to P. This charge transfer corresponds to 0.1 electron per Ni atom. The experimental results have been confirmed by two types of local density calculation using either a cluster approximation or an infinite periodic lattice model where a definite number of Ni atoms are substituted by P impurities. Both computational methods characterize well the measured local and partial density of states.

Effect of phosphorus concentration on the electronic structure of nanocrystalline electrodeposited Ni–P alloys: an XPS and XAES investigation

“…This agrees with earlier soft X-ray studies on the density of states in amorphous Ni-P alloys [13,14] which did not detect any changes in Ni Lα distribution in implanted or electrodeposited Ni-P alloys with respect to pure nickel. Phosphorus concentration (8 < P < 26 at.%) did not influence either the Ni2p main peaks or the Ni Lα distribution.…”

“…Phosphorus concentration (8 < P < 26 at.%) did not influence either the Ni2p main peaks or the Ni Lα distribution. [14] Only a slight shift of the Ni2p3/2 peak (ca +0.2 eV) was observed with respect to pure nickel. [13,14] More recent studies on metallic nickel, NiS and NiAs compounds [21] as well as on Ni-B, Ni-P and Ni-S amorphous alloys confirm that the Ni2p peak-binding energies remain constant, though both satellite binding energy and satellite intensity vary [21,22] (Table 4).…”

“…Compared to elemental red phosphorus, the phosphorus in Ni-P alloys, [26] as well as in other amorphous alloys, [25] exhibits a slightly negative charge and (compared to elemental phosphorus) the binding energy of phosphorus P2p3/2 in Ni-P alloys is ca 0.3-0.4 eV lower, in agreement with the literature data. [13,14,25] Thus it can be concluded that a weak charge transfer (ca 0.3-0.4 electrons per P atom) from nickel to phosphorus occurs in Ni-P alloys, leading to a partially covalent bond. The charge transfer corresponds to +0.1 electron per Ni atom as shown above.…”

“…The slight shift of around 0.1 eV in the Ni2p3/2 core levels of pure nickel and Ni-P alloys may be caused by a small-charge transfer from nickel to phosphorus in the alloy. [13,14] Early studies on the electronic structure of Ni-P alloys based on Compton profiles and XPS valence band structure showed [15] that for a Ni 18.9 at.% P alloy there were no significant differences between electrodeposited or melt spun amorphous alloys. It was further concluded that the 3d band in this alloy is not completely filled and close to that of pure crystalline nickel.…”

“…[13,14,16] The phosphorus DOS is more sensitive to the alloying process, [13,14] a core-level shift of ca 0.5 eV to lower energy corresponding to the charge transfer from nickel to phosphorus (charge transfer is higher on the minority carrier). [14] Several theoretical models have been developed for determining the electronic structure of the Ni-P system, [17 -19] for different phosphorus concentrations (15-25 at.%), [17] Ni-P clusters and supercells [18] and a tightbinding-bond approach with molecular dynamics simulation. [19] Considering the charge transfer, the partial DOS of the metalloids and the specific-binding interaction between the metalloid and transition metal, discrepancies exist between the spectroscopic data and model calculations.…”