Atomic layer deposition (ALD) is known to be an excellent technique for conformal coating. In this work, two models, a kinetic and a Monte Carlo model, are developed to predict the deposited film thickness as a function of depth inside a hole. Earlier work by Gordon et al. assumed a sticking probability of 0/100% for molecules hitting a covered/uncovered section of the wall of the hole, thus resulting in a stepwise coverage profile after a single ALD cycle [1]. However, experimental studies indicate a gradual decrease of film thickness instead of a stepwise drop (figure) [2, and references therein]. It has been argued that the gradual slope may be related to (i) the increasing aspect ratio (AR) during deposition and (ii) the sticking probability, which is less than 100%. The first explanation seems reasonable for the case of microscopic trenches, because during each cycle the deposition of new material results in a decrease of the diameter of the hole, and consequently the effective aspect ratio increases during deposition. However, our experiments using macroscopic structures (~0.1x5x20mm) also show a gradual decrease (figure), suggesting that the sticking probability is an important parameter for predicting the conformality. Therefore, in both models, we related the sticking probability s to the surface coverage θ by Langmuir's equation s(θ)=s 0 (1-θ), whereby the initial sticking probability s 0 is now an adjustable model parameter. For s 0 ≈100%, the models predict a steplike profile, in agreement with Gordon et al., while for smaller values of s 0 , a gradual decreasing coverage profile is predicted. The two models show a good correspondence and follow the same trends as the experimental data (figure).

No abstract

Atomic layer deposition ͑ALD͒ of TiO 2 films from tetrakis͑dimethylamido͒ titanium ͑TDMAT͒ or titanium tetraisopropoxide ͑TTIP͒ precursors was investigated. The growth kinetics, chemical composition, and crystallization behavior of the TiO 2 films were compared for combinations of the two precursors with three different sources of oxygen ͓thermal ALD using H 2 O and plasma-enhanced ALD ͑PEALD͒ using H 2 O or O 2 plasma͔. For TDMAT, the growth rate per cycle ͑GPC͒ decreased with increasing temperature; while for TTIP with either water plasma or O 2 plasma, a relatively constant growth rate per cycle was observed as a function of substrate temperature. It was found that the crystallization temperature of the TiO 2 films depends both on film thickness and on the deposition conditions. A correlation was observed between the TiO 2 crystallization temperature and the C impurity concentration in the film. The TiO 2 films grown using a H 2 O plasma exhibit the lowest crystallization temperature and have no detectable C impurities. In situ X-ray diffraction measurements were used to test the diffusion barrier properties of the TiO 2 layers and proved that all TiO 2 films grown using either H 2 O or O 2 plasma are dense and continuous.

A Metal Organic Framework, containing coordinatively saturated V +IV sites linked together by terephthalic linkers (V-MIL-47) is evaluated as a catalyst in the epoxidation of cyclohexene. Different solvents and conditions are tested and compared. If the oxidant TBHP is dissolved in water, a significant leaching of V-species into the solution is observed and also radical pathways are 2 prominently operative leading to the formation of an adduct between the peroxide and cyclohexene. If however the oxidant is dissolved in decane, leaching is negligible and the structural integrity of the V-MIL-47 is maintained during successive runs. The selectivity towards the epoxide is very high in these circumstances. Extensive computational modelling is performed to show that several reaction cycles are possible. EPR and NMR measurements confirm that at least two parallel catalytic cycles are co-existing: one with V +IV sites and one with pre-oxidized V +V sites, and this in complete agreement with the theoretical predictions.

Vanadium pentoxide was deposited by atomic layer deposition (ALD) from vanadyl-tri-isopropoxide (VTIP). Water or oxygen was used as a reactive gas in thermal and plasma-enhanced (PE) processes. For PE ALD, there was a wide ALD temperature window from 50 to 200°C . Above 200°C , VTIP decomposed thermally, resulting in the chemical vapor deposition (CVD) of vanadium pentoxide. The PE ALD reactions saturated much faster than during thermal ALD, leading to a growth rate of approximately 0.7 Å/cycle during PE ALD using normalH2O or normalO2 . Optical emission spectroscopy showed combustion-like reactions during the plasma step. X-ray diffraction was performed to determine the crystallinity of the films after deposition and after postannealing under He or normalO2 atmosphere. Films grown with CVD at 300°C and PE normalO2 ALD at 150°C were (001)-oriented normalV2normalO5 as deposited, while thermal and PE normalH2O ALD films grown at 150°C were amorphous as deposited. The crystallinity of the PE normalO2 ALD could be correlated to its high purity, while the other films had significant carbon contamination, as shown by X-ray photoelectron spectroscopy. Annealing under He led to oxygen-deficient films, while all samples eventually crystallized into normalV2normalO5 under normalO2 .

This paper focuses on the conformality of the plasma-enhanced atomic layer deposition (PE-ALD) of Al2normalO3 using trimethylaluminum [ Al(CH3)3 ; (TMA)] as a precursor and normalO2 plasma as an oxidant source. The conformality was quantified by measuring the deposited film thickness as a function of depth into macroscopic test structures with aspect ratios of ∼5 , 10, and 22. A comparison with the thermal TMA/normalH2O process indicates that the conformality of the plasma based process is more limited due to the surface recombination of radicals during the plasma step. The conformality can slightly be improved by raising the gas pressure or the radio-frequency power. Prolonging the plasma exposure time results in further improvement of the conformality. Furthermore, there are indications that the normalH2O produced during the plasma step in the PE-ALD process for Al2normalO3 contributes to the observed conformality through a secondary thermal ALD reaction. The conformality of Al2normalO3 is also compared to the conformality of AlN deposited by PE-ALD from TMA and NH3 plasma. For the same exposure, normalO2 plasma results in better conformality compared to NH3 plasma, suggesting a faster recombination of the radicals in the NH3 plasma.

Single-crystal, submicrometer-sized CaS:Eu luminescent particles were synthesized via a solvothermal route, and these moisturesensitive particles were coated with aluminum oxide using atomic layer deposition ͑ALD͒. Photoluminescence ͑PL͒ spectra of coated and uncoated particles were compared. They both showed a broad-band PL emission with a maximum of 650 nm. Microencapsulation by aluminum oxide layers did not have a pronounced effect on the intensity of the emission. In situ luminescence measurements during the accelerated aging ͑80°C, 80% relative humidity͒ of coated and uncoated CaS:Eu particles were performed. While the uncoated phosphor was largely degraded within 30 h of aging, it was observed that a 20 nm thick aluminum oxide coating dramatically increased the resistance of the luminescent material against moisture, showing the conformity of the Al 2 O 3 coating by the ALD process. Upon degradation, CaCO 3 was formed, leading to Eu 3+ emission as observed in cathodoluminescence. Finally, the use of these coated particles as a wavelength conversion material in light-emitting diodes was evaluated.

No abstract