A general analytical model has been developed to calculate particle transport and spatial step coverage evolution within 2-dimensional and 3-dimensional microelectronic device structures during low-pressure chemical vapor deposition. The model can account for spatially dependent nonunity reactive ‘‘sticking probabilities,’’ anisotropic source fluxes, and trench ‘‘shadowing’’ effects. There is no restriction on the initial and evolving shape of the structure. Model results are compared to direct Monte Carlo simulations for step coverage on rectangular trenches, and are found to more accurately describe the observed experimental step coverages during phosphorous-doped silicon dioxide glass film deposition. We present here, for the first time, detailed calculations of step coverage in circular vias for a wide range of reactive sticking probabilities.
Chemical vapor deposition (CVD) TiN is an attractive replacement for PVD TiN as a barrier and glue layer for subhalf-micron contacts and vias. CVD TiN films have been deposited in a commercial reactor via the thermal decomposition of tetrakis-dimethyl-amino-titanium (TDMAT) precursor in an N2 ambient. The deposition can be characterized by a simple Arrhenius rate expression with a half-order dependence on TDMAT concentration and an activation energy of 0.53 eV. Designed experiments show that the deposition transitions from being kinetically limited at high TDMAT flow rates and low temperatures to transport limited at the other extreme. In the kinetically limited regime, the deposition rate increases with increasing TDMAT mole fraction and increasing temperature. Step coverage simulations have been performed by coupling SPEEDIE (a profile evolution program) with the Arrhenius rate expression. Experimental and simulated step coverages show good agreement over a wide range of process conditions. Step coverage improves with increasing deposition rate and decreasing temperature. Adequate deposition rates (≳400 Å/min) with good deposition uniformity and high step coverage (50%–85%) can be achieved. Film stress and roughness are mostly invariant with process conditions. These films are close to stoichiometric (i.e., 1:1 Ti:N ratio), but contain up to 30% carbon. Exposure of these films to air causes rapid oxidation with a steady state concentration of 15%–20% oxygen after 24 h of air exposure. The high carbon concentration results in high film resistivity (5000 μΩ cm), which doubles after 24 h of air exposure due to oxidation of the film. Film resistivity decreases and film stability improves with increasing N2 flow rate. Electrical characterization of two process variants have been performed. The first provides a film with 85% step coverage with a resistivity of 5600 μΩ cm, while the second process achieves a lower resistivity of 3700 μΩ cm and a lower step coverage of 50%. Electrical performance of both films is similar. Contact induced diode leakage is lower for CVD TiN compared to PVD TiN. Contact resistance for 100–200 Å CVD TiN is comparable or lower than 500 Å PVD TiN. A 100 Å CVD TiN barrier is adequate; 300 Å CVD TiN is too thick because of high film resistivity. An Al–Cu/100–300 Å CVD TiN/Si stack shows lower resistance increase after 450–500 °C, 30 min N2 anneal compared to a Al–Cu/200 Å PVD TiN/Si stack indicating that CVD TiN is a superior barrier and is more inert than PVD TiN.
A new, physical based 3-D profile simulator has been developed that includes the dominant effect of re-emission. This simulator is part of the Stanford Profile Emulator for Etching and Deposition in ! C -Engineering (SPEEDIE). Unlike previous simulators which consider only the arrival of deposition precursors by unshadowed direct transport and by surface diffusion, SPEEDIE also considers transport into shadowed areas by adsorption and re-emission. The importance of re-emission was established by using overhang test structures to separate the roles of surface diffusion and re-emission For the depositions investigated (SiO,, poly-Si and W) it was found that re-emission dominates over surface diffusion in controlling surface contours. Using the simulator to fit experimental LPCVD SiO, profiles, it was found that a single constant sticking coefficient (SJ model with a cosine re-emission distribution gave excellent fits independent of geometry for a given deposition condition. Both Monte Carlo and analytic methods are used to calculate the precursor flux along the growing surface.
The low-pressure deposition of SiO2 from tetraethylorthosilicate (TEOS) is studied. Experiments have been done to get the profile evolution in trenches of different aspect ratios and at various time steps until closure. A fast analytical simulator, using an adsorption/reemission model, which can handle multiple species, has been developed to simulate the profile evolution. The deposition profiles were simulated using a single or a two rate limiting precursor model. It has been previously shown that low-pressure chemical vapor deposition (LPCVD) of SiO2 from other sources, such as silane, can be modeled accurately using only one rate limiting precursor. Comparison of simulation results and experimental profiles of LPCVD of SiO2 from TEOS indicates that a deposition model which includes two rate limiting precursors is consistent with experiments. It is suggested that an intermediate species, having a very high reaction sticking coefficient (Sc∼1), is likely to be formed by gas phase reactions. This along with another species of low reaction sticking coefficient, thought to be from the source gas, react separately with the surface to deposit SiO2. Although a one precursor model gives approximate results for early stages of deposition or for thin coverages, it fails to predict the entire profile evolution, from the initial profile until closure. The two precursor model gives more accurate results for all profiles from the initial until closure. The pressure dependence of the step coverage and deposition rate is due to the change in the ratio of the partial pressures of the two rate limiting precursors with deposition pressure.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.