Angle-resolved x-ray photoelectron spectroscopy of ultrathin Al2O3 films grown by atomic layer deposition

Abstract

Thin (45 nm) and ultrathin (4.5–1 nm) Al2O3 layers deposited on HF-stripped Si or thin SiO2 surfaces by atomic layer deposition were studied by angle-resolved x-ray photoelectron spectroscopy, before and after rapid thermal annealing (RTA) at 800 °C for 15 min in N2 or annealing in a conventional furnace under ultrahigh vacuum (UHV) (p=10−6 mbar) and N2 at the same temperature. Samples were characterized in terms of chemical defects and interfacial regrowth upon annealing. Chemical defects as Al–OH groups are evidenced from the O 1s spectra even after RTA. Interfacial regrowth, estimated from the Si 2p components intensity, was found (1) to occur upon every kind of annealing, (2) to be dependent on the initial Al2O3 thickness (the thicker the layer the greater the regrowth), and (3) to be significantly reduced on SiO2 surfaces as thin as 0.5 nm. In agreement with previous studies, we also consistently found the presence of an oxidized Si component at the Al2O3/substrate interface centered at 102 eV for which an assignment to silicate groups (Si–O–Al) is proposed, on the basis of other x-ray photoelectron spectroscopy parameters. Given the influence of UHV annealing in interfacial regrowth, Al–OH defects may play an important role in interface oxidation.

Surface preparation and post thermal treatment effects on interface properties of thin Al2O3 films deposited by ALD

“…Similar results were observed with thermally grown SiO 2 substrate [11]. Taking into account the previous results that a bare SiO 2 film before Al 2 O 3 deposition does not show any silicate component [11], the silicate is thought to be formed during Al 2 O 3 deposition by reaction of Al-OH groups and Si [12,17]. After annealing an increase of SiO 2 component (Si 4þ ) is obviously observed.…”

“…Instead, it consists of two sub-layers: SiO x and a thin layer of Alsilicate at the SiO 2 -Al 2 O 3 interface. By quantitative analyse of O 1s core-level spectra interfacial layer thickness could be obtained [11] and it is coherent with those measured by HRTEM shown in Table 1.…”

“…Photoelectrons were detected at a take-off angle 75°to have maximum sensitivity to interfacial layer between Si and Al 2 O 3 layer. More detailed description and results on ARXPS measurements can be found elsewhere [11]. Current-voltage (I-V ) characteristics were obtained at room temperature and also in different measuring temperatures to investigate the conduction mechanisms which occur through the Al 2 O 3 films.…”

Valence band offset at Al2O3/In0.17Al0.83N interface formed by atomic layer deposition

“…The O 1s II component was found to be a surface-related peak that decreased in intensity as h increased, which was in good agreement with the previous report. 6 The energy difference between these components was 1.4 eV. It is highly likely that the O 1s I component corresponds to Al-O-Al bonds and the O 1s II component corresponds to Al-O-H bonds, which is in agreement with other reports.…”

“…It is highly likely that the O 1s I component corresponds to Al-O-Al bonds and the O 1s II component corresponds to Al-O-H bonds, which is in agreement with other reports. 6,7 In addition, the separation between O 1s I and the valence band maximum (VBM), E V , for the 16-nm-thick Al 2 O 3 layer was constant and independent of h, which indicates that this component can be used to determine the bulk constant parameter of the ALD Al 2 O 3 layer. Therefore, analysis of the band offset was carried out by using the O 1s I component.…”