ALD/ALE 2024 Session AF2-WeA: Growth and Characterization: In-situ and in-vacuo Analysis, Surface Science of ALD II

Rare-earth nickelates (RNiO₃, with R = rare earth element) have a broad range of unique and tunable physical properties, making them relevant for future applications in electronics, electrocatalysis, solid oxide fuel cells, electrochromic windows and microelectromechanical systems. As a result, several RNiO₃ ALD processes (R = La, Nd) have been reported, and the general approach is to combine two binary ALD processes for R_xO_y and NiO in supercycles.^1-3 As two distinct ALD processes are combined into one ALD process, several non-idealities may occur, such as a larger impurity content than for the binary processes, nucleation effects leading to irreproducibility issues, and unexpected high/low growth leading to deviating compositions.⁴

In this work, we combined the ALD NiCp₂/O₃ and Er(MeCp)₃/O₃ processes at 300°C using the supercycle approach, leading to Er_xNi_yO_z films with low impurity content. Figure 1a shows the growth per cycle (GPC) of the ALD process as a function of the NiO cycle ratio. Ideally, a linear curve should be obtained between the GPC values for the binary processes, but this is clearly not the case. In Figure 1b, it is seen that the Ni atomic concentration as found by in vacuo XPS is also much lower than expected based on the rule of mixtures.⁴

To investigate the origin of these non-idealities, in vacuo ALD-XPS nucleation studies were conducted, and a selection of XPS data is shown in Figure 2. Although an Er signal is visible from the first Er(MeCp)₃-pulse on an ALD NiO substrate, it takes ~ 4 cycles of the NiO process to get a clear Ni signal on an ALD Er₂O₃ surface. In Figure 3, a plot is shown of the Ni and Er atomic percentage as a function of ALD cycles, obtained by quantifying the XPS data. The fact that there is a nucleation delay for NiO growth on Er₂O₃, but not for Er₂O₃ on NiO, explains the non-idealities described above. The present XPS data hint that incomplete surface reactions of the NiCp₂/O₃ process on the carbon-rich ALD Er₂O₃-surface are at the basis of this nucleation delay, with more research currently ongoing to elucidate this.

References:

[1] H. Seim et al. J. Mater. Chem. 7, 449–454 (1997).

[2] H. H. Sønsteby et al. Nat. Commun. 11, 2872 (2020).

[3] Y. Sun et al. ACS Appl. Electron. Mater. 3, 1719−1731 (2021).

[4] A. J. M. Mackus et al. Chem. Mater. 31, 1142–1183 (2019).

The properties of aluminum nitride (AlN), including a wide bandgap (6.2 eV), high dielectric constant (k ∼ 9), high electrical resistivity (ρ ∼ 10¹¹–10¹³ Ω cm), and high thermal conductivity (2.85 W/K cm)¹ make it one of the commonly used materials in microelectronics and optoelectronics. AlN is also used in microelectromechanical systems (MEMS devices) due to its piezoelectric properties. It also presents good miscibility with other nitrides and can be used in ternary materials when combined with Ga, In, Ti and Hf, which increases the range of its potential applications.

Although ALD of AlN from AlMe₃ (TMA) has been widely reported to date, the surface chemistry of AlN ALD from other precursors, particularly Al(NMe₂)₃ (TDMAA) have not yet been reported. We compared AlN ALD with TMA and TDMAA as Al precursors and NH₃ with and without plasma activation as the N precursor in the temperature range from 100 to 400 °C. Using mass spectrometry, we find that the surface chemistry of the TMA-process involves reductive elimination and ligand exchange from the gaseous CH₄ detected both during the TMA- and NH₃ pulses. The TDMAA-process also involves transamination mechanisms from the N(Me)₂ and HN(Me)₂, and CH₄ detected during the NH₃ pulses.

By comparing our experimental results to modeling results² drawn from density functional theory methods, we can deduce a detailed, atomistic surface chemical mechanism of TMA on an NH₂-terminated AlN surface. No literature is available for theoretical studies of AlN ALD from TDMAA, making our surface chemical mechanism for the TDMAA-process less detailed.

Refs.:

1. Kot, M. et al. Comparison of plasma-enhanced atomic layer deposition AlN films prepared with different plasma sources. J. Vac. Sci. Technol. A37, 020913 (2019).
2. Rönnby, K. et al. Surface chemical mechanisms of trimethyl aluminum in atomic layer deposition of AlN. J. Mater. Chem. C, 11, 13935-13945 (2023).

Hafnium oxide is a promising material to be used in many applications such as high-k dielectrics or in advanced metal-oxide-semiconductor devices. The accurate knowledge of the deposition is crucial for creating pure, defect free, interfaces as well as optimising the reaction process timing and decreasing material consumption. In the past few years, ambient pressure X-ray photoelectron spectroscopy (APXPS) [1] has been used to study the ALD of HfOx on various surfaces [2,3]. These studies have focused on revealing intricate details within the first few half-cycles and have discovered new surface species, inter- and intramolecular reactions, and bimolecular reaction pathways.

The previous APXPS studies have used water as the co-reactant, but in this work, we turn our focus to ozone. We show how the experimental APXPS setup was modified to incorporate an ozone generator, and the results from the XPS analysis. The measurements were conducted in such a way that we were able to separate the contributions from oxygen and ozone to the growth in the co-reactant half-cycle.

The data indicates that oxygen alone creates very strong changes on the surface and contributes to the growth of the layers removing some ligands from the Hf precursor. Each co-reactant pulse was initiated in pure oxygen and the ozone was only started after a short delay. Deposition temperature was varied in a large range which shows strong affects on the reaction mechanism of the first few half-cycles. The study shows the importance of in situ experiments and separating the process into its constituent components in order to understand them better.

[1.] E. Kokkonen, M. Kaipio, H.-E. Nieminen, F. Rehman, V. Miikkulainen, M. Putkonen, M. Ritala, S. Huotari, J. Schnadt, and S. Urpelainen, Ambient Pressure X-Ray Photoelectron Spectroscopy Setup for Synchrotron-Based in Situ and Operando Atomic Layer Deposition Research, Review of Scientific Instruments 93, 013905 (2022).

[2.] G. D’Acunto et al., Bimolecular Reaction Mechanism in the Amido Complex-Based Atomic Layer Deposition of HfO ₂, Chem. Mater. 35, 529 (2023).

[3.] R. Timm et al., Self-Cleaning and Surface Chemical Reactions during Hafnium Dioxide Atomic Layer Deposition on Indium Arsenide, Nat Commun 9, 1412 (2018).