This work presents the first report of thermal atomic layer deposition (ALD) of NiOx using two nickel precursors - Alanis (TM) and [Ni(ipki)2] - in combination with two different oxygen sources (H2O and O3), over a temperature range of 75-250 degrees C. The Alanis/O3 composition exhibited one of the highest growth rates per cycle (1.1-1.4 & Aring;) and a broad ALD window between 100 and 200 degrees C. In contrast, the Alanis/H2O and [Ni(ipki)2]/O3 combinations yielded lower growth rates of 0.74 & Aring; at 150 degrees C and 0.40 & Aring; at 250 degrees C, respectively. Comprehensive structural, morphological, optical, and chemical characterisation revealed that the choice of precursor combination and the reaction temperature significantly impact the film composition, thereby strongly influencing its suitability for various applications. Notably, those parameters closely determined the photoelectrochemical performance and the stability of the Si/NiOx-based photoanode towards the oxygen evolution reaction (OER). Photoelectrodes fabricated with Alanis/O3 at 200 degrees C demonstrate stability exceeding 24 hours and exhibit a remarkable OER onset potential of 1.15 V vs. RHE for a photocurrent density of 1 mA cm-2.

We investigate the use of first principles thermodynamics based on periodic density functional theory (DFT) to examine the gas-surface chemistry of an oxidized ruthenium surface reacting with hydrogen gas. This reaction system features in the growth of ultrathin Ru films by atomic layer deposition (ALD). We reproduce and rationalize the experimental observation that ALD of the metal from RuO4 and H2 occurs only in a narrow temperature window above 100 °C, and this validates the approach. Specifically, the temperature-dependent reaction free energies are computed for the competing potential reactions of the H2 reagent, and show that surface oxide is reduced to water, which is predicted to desorb thermally above 113 °C, exposing bare Ru that can further react to surface hydride, and hence deposit Ru metal. The saturating coverages give a predicted growth rate of 0.7 Å per cycle of Ru. At lower temperatures, free energies indicate that water is retained at the surface and reacts with the RuO4 precursor to form an oxide film, also in agreement with experiment. The temperature dependence is obtained with the required accuracy by computing Gibbs free energy corrections from phonon calculations within the harmonic approximation. Surface phonons are computed rapidly and efficiently by parallelization on a cloud architecture within the Schrödinger Materials Science Suite. We also show that rotational and translational entropy of gases dominate the free energies, permitting an alternative approach without phonon calculations, which would be suitable for rapid pre-screening of gas-surface chemistries.

Advanced packaging is a key aspect in the drive toward next generation semiconductor applications. Polymers are critical packaging materials and as new uses are explored, demands grow on improved use of existing high performance polymers and development of new polymers meeting the requirements for new electronic technologies. Atomic scale simulation provides a key tool in meeting this development demand. From refractive index to glass transition temperature, atomic scale simulation connects the behavior of the polymer monomers and chains to the properties and performance useful for design and manufacturing. High throughput screening of 10K’s of chemistries and increased understanding of best design directions become accessible with application of atomistic simulation. This talk will describe automated workflows for thermal, mechanical, chemical, and electrical properties that connect to the performance of polymeric electronic packaging materials. The combination of molecular simulation engines including GPU based molecular dynamics with Desmond will be highlighted, with automated workflows for consistent and validated evaluation of properties. Finally, example cases with polymers such as polyimides and epoxy resins will showcase how the methods can be applied to electronic packaging solutions.

In this study, we introduce and alter electrical conduction into para-aramid (aka Kevlar) fibers by utilizing vapor phase infiltration (VPI) to introduce the metal oxide semiconductors ZnO and In2O3, thereby creating a functional hybrid material. VPI leads to an inverted, matrix doping of the introduced semiconductor by the surrounding polymer. The initial insulating para-aramid contains benzenoid rings in its backbone. During infiltration, some of these benzenoid rings transform into quinoid rings, causing a change in the charge density within the polymer chain. This alteration directly affects the bands, photoluminescence, and conductivity of the resulting hybrid fibers. Our findings demonstrate a substantial reduction in electrical resistance of the hybridized para-aramid, surpassing the pristine state by approximately 6 orders of magnitude. Remarkably, exposure to visible light notably enhances the photoconductivity of the ZnO-infiltrated para-aramid, reducing the electrical resistance by 7 orders of magnitude compared to pristine para-aramid. Additionally, we introduce In2O3 as an alternative inorganic compound for VPI, expanding the process's applicability. The In2O3-infiltrated para-aramid exhibits even lower electrical resistance, achieving an impressive reduction by 12 orders of magnitude compared to the pristine state, highlighting the advantages of VPI with In2O3.

GaN is an enabling material for light emitting diodes, advanced radio frequency, and power semiconductor devices. However, fabrication of GaN devices often relies on harsh etch processes, which can leave an etch damage layer, limiting final device performance. In this work, an isotropic atomic layer etching (ALE) process involving SF6 plasma and trimethylaluminium [Al(CH3)3] is presented for the controlled etching of GaN, which reduces oxygen and carbon contamination while smoothing the surface. The ALE chemistry was first examined with density functional theory. A comparison between proposed thermal and plasma-driven reactions is made by implementing Natarajan–Elliott analysis, highlighting that the plasma process is a good candidate for GaN ALE. Saturation was experimentally confirmed for both ALE half-cycles at 150 and 300 °C, with etch rates of 0.31 ± 0.01 and 0.40 ± 0.02 nm/cycle, respectively. Analysis of the films post-ALE shows that the RMS roughness of the films decreases from 2.6 ± 0.1 to 1.9 ± 0.1 nm after 25 nm of etching at 300 °C, in agreement with a previously developed curvature-dependent smoothing model. Taken together, this ALE process enables accurate GaN thickness tuning, surface cleaning, and surface smoothing, allowing for further development of GaN devices.

The spontaneous etching of boron oxide (B2O3) by hydrogen fluoride (HF) gas is important during thermal atomic layer etching after BCl3 converts the surface of various metal oxides to a B2O3 layer. In this study, the chemical vapor etching (CVE) of B2O3 by HF was experimentally monitored using Fourier transform infrared (FTIR) spectroscopy and quadrupole mass spectrometry (QMS). The spontaneous etching of B2O3 by HF gas was also analyzed using density functional theory (DFT). B2O3 films were grown using B2O3 atomic layer deposition with BCl3 and H2O as the reactants at 40 & DEG;C. FTIR spectroscopy then observed the CVE of B2O3 by HF at 150 & DEG;C. B2O3 etching was monitored by the loss of absorbance for B-O stretching vibration in B2O3 films. FTIR spectroscopy studies also observed B-F stretching vibrations from BFx species on the B2O3 surface after HF exposures. In addition, the QMS analysis was able to identify the etch products during the spontaneous etching of B2O3 by HF gas at 150 & DEG;C. The QMS studies observed the main volatile etch products as BF3, BF2(OH), and H2O. Additional volatile etch products were also detected including B3O3F3 and other boroxine ring compounds. The DFT predictions were consistent with the spontaneous etching of B2O3 by HF gas. DFT confirmed that CVE was likely because the energetics of the spontaneous etching reaction B2O3(s) + 6HF(g) & RARR; 2BF(3)(g) + 3H(2)O(g) were more favorable than the self-limiting reaction B2O3(s) + 6HF(g) & RARR; 2BF(3)(s) + 3H(2)O(g). The spontaneous etching of B2O3 was predicted at temperatures above -163 & DEG;C for an HF reactant pressure of 0.2 Torr and BF3 and H2O product pressure of 0.01 Torr.

A combined computational and experimental study is employed to understand the competition between self-limiting (SL) and chemical vapor etch (CVE) reactions to design an atomic layer etch (ALE) process. The pulses in an ALE process have to be self-limiting; i.e., the reactions should reach saturation after sufficient pulse time. By comparing the reaction free energies of corresponding SL and CVE reactions using density functional theory (DFT), the temperature and pressure conditions can be predicted that favor the SL or CVE reactions. The etching of TiO2 when exposed to HF gas is utilized as a test case. Simulations reveal that when TiO2 is exposed to reactant HF at a pressure of 0.2 Torr, the SL reaction removing H2O at 0.01 Torr and fluorinating the surface is preferred up to 87 degrees C (360 K). At higher temperatures, continuous removal of TiO2 by CVE occurs according to the reaction TiO2 + HF -> TiF4 + H2O subject to kinetic activation barriers. Experimental results from in situ Fourier transform infrared (FTIR) spectroscopy and quadrupole mass spectrometry (QMS) are compared with the theoretical predictions. In good agreement with theory, the FTIR spectroscopy studies revealed an onset of spontaneous etching (CVE) at temperatures around 80-90 degrees C. In addition, the QMS analysis observed TiF4 and H2O as the etch products, further validating the calculations. The calculations also predicted that an increase in the reactant gas pressure would enhance etching at high temperatures. The low computational cost of this theoretical approach allows for rapid screening of etch reagents and prediction of the temperature/pressure windows where the reactions will be in the SL or CVE regimes.

Thermally activated chemical vapor‐phase etching of titanium nitride (TiN) is studied by utilizing either alternate exposures of niobium pentafluoride (NbF 5 ) and carbon tetrachloride (CCl 4 ) or by using CCl 4 alone. Nitrogen (N 2 ) gas purge steps are carried out in between every reactant exposure. Titanium nitride is etched in a non‐self‐limiting way by NbF 5 –CCl 4 based binary chemistry or by CCl 4 at temperatures between 370 and 460 °C. Spectroscopic ellipsometry and a weight balance are used to calculate the etch per cycle. For the binary chemistry, an etch per cycle of ≈0.8 Å is obtained for 0.5 and 3 s long exposures of NbF 5 and CCl 4 , respectively at 460 °C. On the contrary, under the same conditions, the etch process with CCl 4 alone gives an etch per cycle of about 0.5 Å. In the CCl 4 ‐only etch process, the thickness of TiN films removed at 460 °C varies linearly with the number of etch cycles. Furthermore, CCl 4 alone is able to etch TiN selectively over other materials such as Al 2 O 3 , SiO 2 , and Si 3 N 4 . X‐ray photoelectron spectroscopy and bright field transmission electron microscopy are used for studying the post‐etch surfaces. To understand possible reaction products and energetics, first‐principles calculations are carried out with density functional theory. From thermochemical analysis of possible reaction models, it is found that NbF 5 alone cannot etch TiN while CCl 4 alone can etch it at high temperatures. The predicted byproducts of the reaction between the CCl 4 gas molecules and TiN surface are TiCl 3 and ClCN. Similarly, TiF 4 , NbFCl 3 , and ClCN are predicted to be the likely products when TiN is exposed to both NbF 5 and CCl 4 . A more favorable etch reaction is predicted when TiN is exposed to both NbF 5 and CCl 4 (Δ G = −2.7 eV at 640 K) as compared to exposure to CCl 4 only (Δ G = −2 eV at 640 K) process. This indicates that an enhanced etch rate is possible when TiN is exposed alternately to both NbF 5 and CCl 4 , which is in close agreement with the experimental results.

Thermal atomic layer etching (ALEt) of amorphous Al2O3 was performed by alternate exposures of niobium pentafluoride (NbF5) and carbon tetrachloride (CCl4). The ALEt of Al2O3 is observed at tempera...

HfO2 and ZrO2 are two high-k materials that are important in the downscaling of semiconductor devices. Atomic-level control of material processing is required for the fabrication of thin films of these materials at nanoscale device sizes. Thermal atomic layer etching (ALE) of metal oxides, in which up to one monolayer of the material can be removed, can be achieved by sequential self-limiting (SL) fluorination and ligand-exchange reactions at elevated temperatures. However, to date, a detailed atomistic understanding of the mechanism of thermal ALE of these technologically important oxides is lacking. In this paper, we investigate the hydrogen fluoride (HF) pulse in the first step in the thermal ALE process of HfO2 and ZrO2 using first-principles simulations. We introduce Natarajan-Elliott analysis, a thermodynamic methodology, to compare reaction models representing the self-limiting (SL) and continuous spontaneous etching (SE) processes taking place during an ALE pulse. Applying this method to the first HF pulse on HfO2 and ZrO2, we found that thermodynamic barriers impeding continuous etching are present at ALE-relevant temperatures. We performed explicit HF adsorption calculations on the oxide surfaces to understand the mechanistic details of the HF pulse. A HF molecule adsorbs dissociatively on both oxides by forming metal-F and O-H bonds. HF coverages ranging from 1.0 +/- 0.3 to 17.0 +/- 0.3 HF/nm(2) are investigated, and a mixture of molecularly and dissociatively adsorbed HF molecules is present at higher coverages. Theoretical etch rates of -0.61 +/- 0.02 angstrom/cycle for HfO2 and -0.57 +/- 0.02 angstrom/cycle for ZrO2 were calculated using maximum coverages of 7.0 +/- 0.3 and 6.5 +/- 0.3 M-F bonds/nm(2), respectively (M = Hf, Zr).

Atomic layer deposition (ALD) of TiO2 thin films on a Si substrate has been investigated using titanium isopropoxide (TTIP) and tetrakis(dimethylamino)titanium (TDMAT) in combination with water. The deposition rate and the chemical stability of the films are significantly different depending on the Ti precursor and process temperature (T-ALD). When the films are annealed a significant thickness shrinkage is reported for the first time on TiO2. A comprehensive analysis of the films with X-ray photoelectron spectroscopy, Fourier transform infrared, ellipsometry, and porosimetry demonstrates that some precursor ligands are incorporated (most likely as isopropanol) when ALD is performed at low temperature (i.e., T-ALD < 200 degrees C) using TTIP. The trapped ligand molecules can be removed by annealing but make the film porous and thus have a detrimental effect on the dielectric properties. Higher-quality nonporous films are grown by using TTIP at T-ALD >= 200 degrees C or by using TDMAT. It is shown that measuring the refractive index is a simple, nondestructive, and reliable way to determine film quality. Numerical simulations of ligand coverage show that the measured growth rates are consistent with a self-limiting ALD mechanism albeit with partial incorporation of ligands from TTIP at low temperature (T-ALD < 200 degrees C), which renders part of the surface inactive toward growth. Aside from this, the higher growth rate of TDMAT is due to more desorption of ligands during the Ti precursor pulse. The overall decrease in the growth rate with temperature is related quantitatively to decreasing coverage of hydroxyl groups on TiO2. Comparing the TTIP and TDMAT processes in this way reveals new aspects of the gas-surface chemistry during self-limiting ALD and how this affects film morphology and electrical properties.

Atomic layer deposition (ALD) of boron-containing films has been mainly studied for use in two-dimensional materials and for B doping of Si. Furthermore, lithium-containing borates show great promise as solid electrolyte coatings for enhanced energy storage. In this work, we examine trimethyl borate (TMB) in combination with O-2 plasma as a precursor for ALD of B-containing films, targeting the growth of B2O3. It is found that after initial growth on a SiO2 or Al2O3 surface, a rapid decrease in the rate of growth during subsequent ALD cycles occurs, indicating surface inhibition during continued growth. Density functional theory (DFT) cluster calculations in combination with in situ Fourier transform infrared spectroscopy (FTIR) demonstrated that the growth is governed by two different mechanisms depending on the Lewis acidity of the surface: chemisorption on an Al-OH- and Si-OH-terminated surface and physisorption on more acidic B-OH surface sites. The growth could be maintained in a mixed process, by reactivating the surface through single exposures to trimethyl aluminum (TMA) and O-2 plasma and thus resetting the surface to Al-OH, on which TMB chemisorption is energetically more favorable. Surprisingly, this process did not result in B2O3 (or Al-doped B2O3) films but instead in B- and H-doped Al2O3 films. Moreover, rather than a uniform boron distribution, the Al2O3 films grown from this process contain a large amount of hydrogen, up to 17 at. % under certain processing conditions, and displayed non-uniform depth distributions of boron and hydrogen with a degree of control over the doping distribution based on the deposition conditions. Finally, the mechanism for the atypical growth mode is proposed on the basis of in situ FTIR and ellipsometry measurements and DFT calculations and was attributed to subsurface reactions of the TMA with the B-OH films grown by TMB-O-2 plasma. This makes the process an interesting, albeit atypical, ALD process that allows for a quasi-continuous tuning of the B concentration in the top region of high-purity Al2O3 films.

A method to obtain full mass over charge (m/z), time-resolved quadruple mass spectrometry (QMS) spectra of an atomic layer deposition (ALD) cycle is proposed. This method allows one to circumvent the limitations of traditional approaches for obtaining QMS information in ALD as all m/z values can be simultaneously screened for the formation of reaction products in an efficient way. As a proof of concept, this method was applied to the trimethylaluminum (TMA)-water process. This process has been studied extensively over the past decades. Besides the expected formation of CH4, formation of gaseous HOAI(CH3)(2) during the water pulse is observed, revealing a secondary reaction pathway for the water. The reaction energy and Gibbs free energy for different reactions are investigated computationally using density functional theory calculations and confirm that the secondary reaction pathway is thermodynamically allowed for certain surface conditions.

The fabrication of next-generation semiconductor devices has created a need for low-temperature (≤400 °C) deposition of highly-conformal (>95%) SiO2, SiNx, and SiC films on high-aspect-ratio nanostructures. To enable the growth of these Si-based dielectric films, semiconductor manufacturers are transitioning from chemical vapor deposition to atomic layer deposition (ALD). Currently, SiO2 films deposited using ALD are already being integrated into semiconductor device manufacturing. However, substantial processing challenges remain for the complete integration of SiNx films deposited by ALD, and there are no known processes for ALD of SiC at temperatures that are compatible with semiconductor device manufacturing. In this focused review, the authors look at the status of thermal and plasma-assisted ALD of these three Si-based dielectric films. For SiO2 ALD, since low-temperature processes that deposit high-quality films are known, the authors focus primarily on the identification of surface reaction mechanisms using chlorosilane and aminosilane precursors, as this provides a foundation for the ALD of SiNx and SiC, two material systems where substantial processing challenges still exist. Using an understanding of the surface reaction mechanisms, the authors describe the underlying reasons for the processing challenges during ALD of SiNx and SiC and suggest methodologies for process improvement. While both thermal and plasma-assisted SiNx ALD processes have been reported in the literature, the thermal NH3-based ALD processes require processing temperatures >500 °C and large NH3 doses. On the other hand, plasma-assisted SiNx ALD processes suffer from nonuniform film properties or low conformality when deposited on high-aspect-ratio nanostructures. In the SiNx section, the authors provide a broad overview of the currently known thermal and plasma-assisted SiNx ALD processes using chlorosilane, trisilylamine, and aminosilane precursors, describe the process shortcomings, and review the literature on precursor reaction pathways. The authors close this section with suggestions for improving the film properties and conformality. In the case of SiC, the authors first outline the limitations of previously reported SiC ALD processes and highlight that unlike SiO2 and SiNx plasma-assisted ALD, no straightforward pathway for low-temperature plasma-assisted growth is currently apparent. The authors speculate that low-temperature ALD of SiC may require the design of completely new precursors. Finally, they summarize the progress made in the ALD of C-containing SiNx and SiO2 films, which may provide many of the benefits of SiC ALD in semiconductor manufacturing. In closing, through this review, the authors hope to provide the readers with a comprehensive knowledge of the surface reactions mechanisms during ALD of Si-based dielectrics, which would provide a foundation for future precursor and process development.

HfO2 and ZrO2 are two high-k materials that are important in the down-scaling of semiconductor devices. Atomic level control of material processing is required for fabrication of thin films of these materials at nanoscale device sizes. Thermal Atomic Layer Etch (ALE) of metal oxides, in which up to one monolayer of the material can be removed, can be achieved by sequential self-limiting fluorination and ligand-exchange reactions at elevated temperatures. However, to date a detailed atomistic understanding of the mechanism of thermal ALE of these technologically important oxides is lacking. In this paper, we investigate the hydrogen fluoride pulse in the first step in the thermal ALE process of HfO2 and ZrO2 using first principles simulations. We introduce Natarajan-Elliott analysis, a thermodynamic methodology, to compare reaction models representing the self-limiting (SL) and continuous spontaneous etch (SE) processes taking place during an ALE pulse. Applying this method to the first HF pulse on HfO2 and ZrO2 we found that thermodynamic barriers impeding continuous etch are present at ALE relevant temperatures. We performed explicit HF adsorption calculations on the oxide surfaces to understand the mechanistic details of the HF pulse. A HF molecule adsorbs dissociatively on both oxides by forming metal-F and O-H bonds. HF coverages ranging from 1.0 0.3 to 17.0 0.3 HF/nm2 are investigated and a mixture of molecularly and dissociatively adsorbed HF molecules is present at higher coverages. Theoretical etch rates of -0.61 0.02 Å /cycle for HfO2 and -0.57 0.02 Å /cycle ZrO2 were calculated using maximum coverages of 7.0 0.3 and 6.5 0.3 M-F bonds/nm2 respectively (M = Hf, Zr).

Thermal atomic layer etch, the reverse of atomic layer deposition, uses a cyclic sequence of plasma-free and solvent-free gas surface reactions to remove ultrathin layers of material with a high degree of control. A theoretical investigation of the hydrogen fluoride pulse in the thermal atomic layer etch of monoclinic alumina has been performed using density functional theory calculations. From experiments, it has been suggested that the HF pulse forms a stable and nonvolatile layer of AlF3 on an alumina surface. Consistent with this, the desorption of an AlF3 molecule from an HF-saturated surface was computed to be energetically unfavorable. HF molecules adsorbed on the alumina surface by forming hydrogen bonds and either remained intact or dissociated to form Al-F and O-H species. At higher coverages, a mixture of molecularly and dissociatively adsorbed HF molecules in a hydrogen-bonded network was observed. Binding energies converged as the coverage of dissociated F became saturated, consistent with a self-limiting reaction. The formation of H2O molecules in the HF pulse was found to be endoergic with an energy barrier of at least +0.9 eV, but their subsequent desorption was computed to cost as little as +0.2 eV. On the basis of a model of the saturated Al-F surface, the theoretical maximum of the etch rate was estimated to be 0.57 0.02 angstrom/cycle (-20.0 +/- 0.8 ng cm(-2) cycle(-1)), which matches the range of maximum experimental values. The actual etch rate will, however, be dependent on the specific reagent used in the subsequent step of the atomic layer etch cycle.

Atomic layer deposition (ALD) has emerged as an important technique for thin-film deposition in the last two decades. Zinc oxide thin films, usually grown via diethylzinc (DEZ) and water process, have seen much interest both in application and in theoretical research. The surface processes related to the growth of the thin film are not entirely understood, and the conceptual picture of the ALD process has been contradicted by recent experiments where ligands from the zinc pulse persist on the surface even after extended water pulse exposures. In this work, we investigate the overall growth of the zinc oxide thin films grown via DEZ/H2O process by modeling the surface chemistry using first-principles kinetic Monte Carlo for the first time. The kinetic Monte Carlo allows us to implement density functional theory calculations conducted on the zinc oxide (100) surface into a kinetic model and extract data directly comparable to experimental measurements. The temperature-dependent growth profile obtained from our model is in good qualitative agreement with the experimental data. The onset of thin-film growth is offset from the experimental data because of the underestimation of the reaction barriers within density functional theory. The growth per cycle of the deposited film is overestimated by 18% in the kinetic model. Mass gain during an ALD cycle is in qualitative agreement with the experimental quartz-crystal microbalance data. The main mass gain within an ALD cycle is obtained during the DEZ pulse and mass change during the water pulse is negligible. The cause of low film growth at low temperatures is due to the high reaction barriers for ethyl-elimination during the water pulse. This kinetic barrier results in low film growth as no new DEZ can adsorb to the ethyl saturated surface. At elevated temperatures, ethyl-elimination becomes accessible, resulting in the ideal layer-by-layer growth of the film. However, a large fraction of ethyl-ligands persist on the surface after each ALD cycle even at high temperatures. This results in ethyl-ligands being encapsulated into the film lattice. This is likely due to an incomplete set of reaction pathways, and it is likely that some yet unidentified process is responsible for the elimination of the ethyl-ligands from the surface as the deposition process progresses.

Understanding the mechanism of SiC chemical vapor deposition (CVD) is an important step in investigating the routes toward future atomic layer deposition (ALD) of SiC. The energetics of various silicon and carbon precursors reacting with bare and H-terminated 3C-SiC (011) are analyzed using ab initio density functional theory (DFT). Bare SiC is found to be reactive to silicon and carbon precursors, while H-terminated SiC is found to be not reactive with these precursors at 0 K. Furthermore, the reaction pathways of silane plasma fragments SiH3 and SiH2 are calculated along with the energetics for the methane plasma fragments CH3 and CH2. SiH3 and SiH2 fragments follow different mechanisms toward Si growth, of which the SiH3 mechanism is found to be more thermodynamically favorable. Moreover, both of the fragments were found to show selectivity toward the Si-H bond and not C-H bond of the surface. On the basis of this, a selective Si deposition process is suggested for silicon versus carbon-doped silicon oxide surfaces.

Plasma-enhanced atomic layer deposition (ALD) of metal oxides is a rapidly gaining interest especially in the electronics industry because of its numerous advantages over the thermal process. However, the underlying reaction mechanism is not sufficiently understood, particularly regarding saturation of the reaction and densification of the film. In this work, we employ first principles density functional theory (DFT) to determine the predominant reaction pathways, surface intermediates and by-products formed when constituents of O2-plasma or O3 adsorb onto a methylated surface typical of TMA-based alumina ALD. The main outcomes are that a wide variety of barrierless and highly exothermic reactions can take place. This leads to the spontaneous production of various by-products with low desorption energies and also of surface intermediates from the incomplete combustion of –CH3 ligands. Surface hydroxyl groups are the most frequently observed intermediate and are formed as a consequence of the conservation of atoms and charge when methyl ligands are initially oxidized (rather than from subsequent re-adsorption of molecular water). Anionic intermediates such as formates are also commonly observed at the surface in the simulations. Formaldehyde, CH2O, is the most frequently observed gaseous by-product. Desorption of this by-product leads to saturation of the redox reaction at the level of two singlet oxygen atoms per CH3 group, where the oxidation state of C is zero, rather than further reaction with oxygen to higher oxidation states. We conclude that the self-limiting chemistry that defines ALD comes about in this case through the desorption by-products with partially-oxidised carbon. The simulations also show that densification occurs when ligands are removed or oxidised to intermediates, indicating that there may be an inverse relationship between Al/O coordination numbers in the final film and the concentration of chemically-bound ligands or intermediate fragments covering the surface during each ALD pulse. Therefore reactions that generate a bare surface Al will produce denser films in metal oxide ALD.

Area-selective atomic layer deposition (AS-ALD) allows nanostructures of arbitrary composition and lateral shape to be built with atomic precision on pre-selected substrate locations. Most current approaches for AS-ALD are based on local inhibition (e.g. with self-assembled monolayers) or activation. However, for some applications of AS-ALD (e.g. in self-aligned fabrication) it is relevant to be able to exploit differences in chemical behavior of a pre-patterned substrate. For this reason, investigating inherent differences in nucleation on diverse substrates is of crucial importance for developing future AS-ALD processes. In this paper we are focussing on substrates of silicon and silicon-based dielectric materials (SiC, Si0 2 and SiNx) used in electronics.