Low Temperature Growth of High Purity, Low Resistivity Copper Films by Atomic Layer Deposition

Dey

The Journal of Chemical Physics

2017

Original citationElliott Reaction cycles for the atomic layer deposition (ALD) of metals are presented, based on the incomplete data that exist about their chemical mechanisms, particularly from density functional theory (DFT) calculations. ALD requires self-limiting adsorption of each precursor, which results from exhaustion of adsorbates from previous ALD pulses and possibly from inactivation of the substrate through adsorption itself. Where the latter reaction does not take place, an "abbreviated cycle" still gives self-limiting ALD, but at a much reduced rate of deposition. Here, for example, ALD growth rates are estimated for abbreviated cycles in H 2 -based ALD of metals. A wide variety of other processes for the ALD of metals are also outlined and then classified according to which a reagent supplies electrons for reduction of the metal. Detailed results on computing the mechanism of copper ALD by transmetallation are summarized and shown to be consistent with experimental growth rates. Potential routes to the ALD of other transition metals by using complexes of non-innocent diazadienyl ligands as metal sources are also evaluated using DFT. Published by AIP Publishing.

Section: Figmentioning

confidence: 99%

Classification of processes for the atomic layer deposition of metals based on mechanistic information from density functional theory calculations

Dey

The Journal of Chemical Physics

2017

“…6 Kalutarage et al compared twostep and three-step processes using the ALD reaction of Cu(dmap) 2 with BH 3 (NHMe 2 ) and separately with BH 3 (NHMe 2 ) and HCO 2 H. 9 They showed that the two-step process requires a Cu seed layer, and affords a growth rate of about 0.13 Å/cycle within the 130−160 °C ALD window.…”

Section: Introductionmentioning

confidence: 99%

Kinetics and Coverage Dependent Reaction Mechanisms of the Copper Atomic Layer Deposition from Copper Dimethylamino-2-propoxide and Diethylzinc

2016

Chem. Mater.

Original citationMaimaiti, Yasheng; Elliott, Simon D. (2016) Just Accepted "Just Accepted" manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides "Just Accepted" as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. "Just Accepted" manuscripts appear in full in PDF format accompanied by an HTML abstract. "Just Accepted" manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). "Just Accepted" is an optional service offered to authors. Therefore, the "Just Accepted" Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the "Just Accepted" Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these "Just Accepted" manuscripts. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 2 mechanism. During both reaction mechanisms, ligand diffusion and reordering are generally endothermic processes, which may result in residual ligands blocking the surface sites at the end of the Et 2 Zn pulse, and in residual Zn being reduced and incorporated as an impurity. We also find that the nearby ligands play a cooperative role in lowering the activation energy for formation and desorption of by-products, which explains the advantage of using organometallic precursors and reducing agents in Cu ALD. The ALD growth rate estimated for the mechanism is consistent with the experimental value of 0.2 Å/cycle. The proposed reaction mechanisms provide insight into ALD processes for copper and other transition metals.

“…12 Later, a thin film of very pure Cu has been deposited using a three-step ALD process which entails the sequential reactions of Cu(dmap) 2 , formic acid and hydrazine (N 2 H 4 ) at 120 °C. 6 Recently, low temperature Cu ALD has also been demonstrated using a two-step process of Cu(dmap) 2 and borane dimethylamine [BH 3 (NHMe 2 )]…”

Section: Introductionmentioning

confidence: 99%

Precursor Adsorption on Copper Surfaces as the First Step during the Deposition of Copper: A Density Functional Study with van der Waals Correction

2015

J. Phys. Chem. C

Copper dimethylamino-2-propoxide [Cu(dmap) 2 ] is used as a precursor for low-temperature atomic layer deposition (ALD) of copper thin films. Chemisorption of the precursor is the necessary first step of ALD, but it is not known in this case whether there is selectivity for adsorption sites, defects, or islands on the substrate. Therefore, we study the adsorption of the Cu(dmap) 2 molecule on the different sites on flat and rough Cu surfaces using PBE, PBE-D3, optB88-vdW, and vdW-DF2 methods. We found the relative order of adsorption energies for Cu(dmap) 2 on Cu surfaces is E ads (PBE-D3) > E ads (optB88-vdW) > E ads (vdW-DF2) > E ads (PBE). The PBE and vdW-DF2 methods predict one chemisorption structure, while optB88-vdW predicts three chemisorption structures for Cu(dmap) 2 adsorption among four possible adsorption configurations, whereas PBE-D3 predicts a chemisorbed structure for all the adsorption sites on Cu(111). All the methods with and without van der Waals corrections yield a chemisorbed molecule on the Cu( 332) step and Cu(643) kink because of less steric hindrance on the vicinal surfaces. Strong distortion of the molecule and significant elongation of Cu−N bonds are predicted in the chemisorbed structures, indicating that the ligand−Cu bonds break during the ALD of Cu from Cu(dmap) 2 . The molecule loses its initial square-planar structure and gains linear O−Cu−O bonding as these atoms attach to the surface. As a result, the ligands become unstable and the precursor becomes more reactive to the coreagent. Charge redistribution mainly occurs between the adsorbate O−Cu−O bond and the surface. Bader charge analysis shows that electrons are donated from the surface to the molecule in the chemisorbed structures, so that the Cu center in the molecule is partially reduced.