997
On the Growth of Silver Thin Films By Atmospheric-Plasma Spatial ALD

Wednesday, October 14, 2015: 15:00
Phoenix East (Hyatt Regency)

ABSTRACT WITHDRAWN

Silver nanoparticles and thin-films are gaining attention for plasmonics applications in photovoltaics[1] and optical lithography[2], amongst others. However, the growth of ultra-thin noble metals films is generally challenging due to their high surface energy that hinders the wetting on the substrate. In this work we investigated the influence of substrates with different surface energy on the nucleation and growth evolution of silver thin films deposited by atmospheric-pressure plasma enhanced spatial Atomic Layer Deposition (ALD) in a rotary reactor(Fig.1)[3].

The ultra-thin Ag films were deposited using Ag(fod)(PEt3), triethylphosphine (6,6,7,7,8,8,8-heptafluoro-2,2-dimethyl-3,5-octanedionate silver(I), as the precursor and an H2/N2 plasma generated by a surface dielectric barrier discharge (SDBD) source as the co-reactant. Silver thin films were deposited at 120 °C with a rotation speed of 40 rpm. Ag(fod)(PEt3) was evaporated into the deposition chamber using Ar as carrier gas flowing at 100 sccm, at the same temperature as that of the reactor. The H2/N2 co-reactant gas mixture was composed from 800 sccm H2 and 600 sccm N2. Two substrates were used: crystalline Si (100) and borosilicate glass, with and without sputter-deposited molybdenum as a wetting layer.

Molybdenum films are known to have high surface energy (3 Jm-2)[4]. Thus Ag is expected to wet a Mo film better than a bare Si surface. To verify this, the differences in the growth mode of Ag Mo wetting layers were investigated on the basis of electrical, optical and morphological properties as analyzed by four-point probe (4PP), spectroscopic ellipsometry (SE) and SEM imaging (Fig. 2), respectively. Preliminary results show that the coalescence of the Ag film is more facile on 80 nm Mo/Si than on bare Si. In the case of Ag thin film growth on an 80 nm Mo wetting layer on top of Si, the Ag sheet resistance derived from 4PP measurements shows that percolation is readily achieved. Going from bare Si substrate to Mo thin film substrate, the sheet resistance drops from ~10 Ω/□ to ~3.5 Ω/□. In the case of a bare Si surface for different exposure times the measured sheet resistance, the increasing contribution of the Drude tail in the SE analysis and the SEM inspection all indicate that bare Si delays the onset of layer-by-layer growth of closed Ag thin films, leading to a Volmer-Weber island growth mode. Furthermore, the effect of the wetting layer has been investigated in greater detail, by varying its thickness from a 10 nm non-coalesced Mo film to an 80 nm closed one.