We compare the growth behavior of nm-thin ruthenium layers using N 2 /NH 3 -and N 2 /H 2 -based plasmas during plasma enhanced atomic layer deposition for two ruthenium precursors. For bis(ethylcyclopentadienyl)ruthenium, we have found a large incubation time on titanium nitride when using N 2 /NH 3 plasma. With N 2 /H 2 plasma the incubation was reduced and tunable with the Ru dose per pulse. In contrast to the incubation, the steady state growth does not depend significantly on the plasma ambient given that there are H 2 or NH 3 species. For (methylcyclopentadienylpyrrolyl)ruthenium, no significant incubation was observed for either plasmas, which is attributed to the presence of the pyrrolyl group.Recently, ruthenium films obtained by atomic layer deposition (ALD) have attracted a lot of interest as enabling material in future technology nodes for several applications such as dynamic random access memory (DRAM) and interconnect metallization. 1 For DRAM metal-insulator-metal capacitors (MIMCAP), Ru and/or RuO 2 are candidates to replace TiN electrodes in metal-insulator-metal capacitor structures. Ru has the advantage, when compared to other metals, that it remains conducting when oxidized and hence does not contribute to the equivalent oxide thickness (EOT) of the capacitor.Ultrathin, smooth, conformally deposited Ru films are needed for future 3D devices structures. This is a challenge given the multiple reports on nucleation issues for ALD Ru on various substrates. Thermal ALD of Ru using O 2 as a co-reactant is studied most frequently. 2,3,4 Alternatively, a plasma such as N 2 /H 2 or NH 3 is used as co-reactant. The latter is called plasma enhanced ALD (PE-ALD). In general, the use of plasma during or before Ru ALD has been shown to significantly reduce the incubation. 5,6,7 The most reported Ru precursor for PE-ALD Ru is bis(ethylcyclopentadienyl)-ruthenium or Ru(EtCp) 2 (references see Table I). However, the initial growth behavior as well as and the growth-per-cycle (GPC) values (Table I) vary strongly from report to report. This is likely due to the extreme sensitivity of the ALD Ru process to the substrate preparation and/or process conditions such as reactor temperature, reactor pressure, precursor dose and plasma conditions. The impact of the plasma, usually N 2 /H 2 or NH 3 based, on the nucleation and the growth, is not yet reported. Besides, the direct comparison of the PE-ALD Ru process for Ru(EtCp) 2 and one of the more recently reported Ru percursors like methylcyclopentadienylpyrrolylruthenium (MCPRu) 3,7 has not been described in literature.In this letter, we report on the deposition of PE-ALD Ru from Ru(EtCp) 2 and MCPRu using similar process conditions. For both precursors, we compare the impact of N 2 /NH 3 and N 2 /H 2 plasma on the nucleation and growth of PE-ALD Ru.Ru thin films were deposited by PE-ALD in a 300 mm ALD chamber with direct plasma capability. The plasma is generated between the substrate and the showerhead using a high frequency generator. The ALD chamber is attached to...

No abstract

Charging currents of electrons and ions to a spherical dust grain in a uniform magnetized dusty plasma have been examined. It is found that the external magnetic field reduces the charging currents, thereby decreasing the dust charge fluctuation damping of a low-frequency electrostatic wave in a dusty plasma.

Jeans instability in a homogeneous cold quantum dusty plasma in the presence of the ambient magnetic field and the quantum effect arising through the Bohm potential has been examined using the quantum magnetohydrodynamic model. It is found that the Jeans instability is significantly reduced by the presence of the dust-lower-hybrid wave and the ion quantum effect. The minimum wavenumber for Jeans stability depends clearly on ion quantum effect and the dust-lower-hybrid frequency also.