Elizabeth F. Moore

Numerical modeling of particle dynamics in a rotating disk chemical vapor deposition reactor

Abstract Particle contamination is considered to be one of the major problem areas in the processing of semiconductors via chemical vapor deposition (CVD). Thus it is very important to acquire an understanding of particle transport processes in CVD reactors. This paper addresses this issue by presenting the results of a numerical simulation of particle dynamics in a rotating disk CVD reactor. The background flowfield calculation employs the full axisymmetric Navier-Stokes equations, while individual particle trajectories are computed by accounting for inertial, thermophoretic and gravitational effects. The results of this simulation are analyzed to determine under what conditions particles greater than 1 μm in diameter impact and thus contaminate the deposition substrate. It is shown that particle size and injection location as well as flow direction (with or against gravity) and disk characteristics (temperature and rotation rate) all play important roles here. The results for various parameter combinations are presented and discussed, as is the concept of a global type of particle contamination parameter.

An Investigation of Particle Dynamics in a Rotating Disk Chemical Vapor Deposition Reactor

This paper describes a numerical model for the nucleation, growth, and transport of gas-phase particles formed during the chemical vapor deposition (CVD) of epitaxial silicon from silane. These particles can lower the deposition rate by consuming precursor, and contaminate the growing film via diffusion to the surface. This model has been constructed for use with the Sandia SPIN code, which contains a solver for the reacting flow and heat transfer in a vertical, rotating disk CVD reactor. A detailed gas-phase chemical kinetic mechanism for the thermal decomposition of silane was developed to simulate formation of small silicon clusters and the depletion of reactive intermediates through condensation. The particle model uses a moment transport formulation to examine the effects of total reactor pressure, temperature, rotation rate, inlet gas composition, and rate of particle growth via condensation on the characteristics of the particle population. Numerical results are presented in terms of the integral moments of the particle distribution which correspond physically to the particle number concentration, average particle diameter, and particle light scattering intensity. In situ validation experiments have been performed in an optically accessible reactor under conditions typical of silicon CVD. The rate of particle growth via condensation, controlled numerically by a global condensation parameter (GCP), was found to control the characteristics of the particle population. The numerical results were found to compare favorably with experiment if this GCP was properly chosen.

In Situ Gas Phase Diagnostics for Hafnium Oxide Atomic Layer Deposition

Atomic layer deposition (ALD) is an important method for depositing the nanometer-scale, conformal high κ dielectric layers required for many nanoelectronics applications. In situ monitoring of ALD processes has the potential to yield insights that will enable efficiencies in film growth, in the development of deposition recipes, and in the design and qualification of reactors. This report will describe the status of a project to develop in situ diagnostics for hafnium oxide ALD processes. The focus is on an examination of the utility of Fourier transform infrared spectroscopy and diode laser spectroscopy for optimizing deposition conditions, rather than simply monitoring precursor delivery. Measurements were performed in a single-wafer, warmwall, horizontal-flow reactor during hafnium oxide ALD involving tetrakis(ethylmethylamino) hafnium and water. Measurements were performed near the wafer surface under a range of deposition conditions in an effort to correlate gas phase measurements with surface processes.

A Numerical Investigation of the Effects of Gas-Phase Particle Formation on Silicon Film Deposition from Silane

This paper presents a systematic numerical investigation of the effects of particle formation on silicon film deposition from silane in a vertical rotating disk chemical vapor deposition reactor. The numerical model uses the Sandia SPIN code to simulate the reacting flow and heat transfer. A moment transport aerosol model simulates the nucleation, growth and transport of silicon particles. The effects of total reactor pressure, temperature, rotation rate, inlet gas composition, and rate of particle growth via condensation on the deposition rate of the silicon film are investigated. Results are presented detailing the scavenging of film growth precursor molecules by particles through a series of simulations both with and without an aerosol component. Conditions under which the particles affect the deposition rate have been identified. Additionally, the use of a chemically reactive carrier (H2) is shown to effectively suppress the formation of particles in the gas phase. The effect of this particle suppression on the deposition rate is discussed.

An uncertainty analysis of mean flow velocity measurements used to quantify emissions from stationary sources

Point velocity measurements conducted by traversing a Pitot tube across the cross section of a flow conduit continue to be the standard practice for evaluating the accuracy of continuous flow-monitoring devices. Such velocity traverses were conducted in the exhaust duct of a reduced-scale analog of a stationary source, and mean flow velocity was computed using several common integration techniques. Sources of random and systematic measurement uncertainty were identified and applied in the uncertainty analysis. When applicable, the minimum requirements of the standard test methods were used to estimate measurement uncertainty due to random sources. Estimates of the systematic measurement uncertainty due to discretized measurements of the asymmetric flow field were determined by simulating point velocity traverse measurements in a flow distribution generated using computational fluid dynamics. For the evaluated flow system, estimates of relative expanded uncertainty for the mean flow velocity ranged from ±1.4% to ±9.3% and depended on the number of measurement locations and the method of integration. Implications: Accurate flow measurements in smokestacks are critical for quantifying the levels of greenhouse gas emissions from fossil-fuel-burning power plants, the largest emitters of carbon dioxide. A systematic uncertainty analysis is necessary to evaluate the accuracy of these measurements. This study demonstrates such an analysis and its application to identify specific measurement components and procedures needing focused attention to improve the accuracy of mean flow velocity measurements in smokestacks.

Dynamical systems approach to particle transport modeling in dilute gas-particle flows with application to a chemical vapor deposition reactor

A general method for treating the transport of particles in dilute gas-particle flows is presented. This method formulates the particle transport equations as a dynamical system and utilizes the corresponding theory to analyze the resulting dynamics. As an illustration of this technique, the transport of contaminant particles in a barrel-type chemical vapor deposition reactor is analyzed. The background gas flow field is determined via a numerical simulation of the Navier-Stokes equations. The analysis reveals the presence of particle attractors in the flow which are particle size specific. The features of these attractors are discussed, as is their behavior with varying particle size.

Atomic Layer Deposition — Process Models and Metrologies

We report on the status of a combined experimental and modeling study for atomic layer deposition (ALD) of HfO2 and Al2O3. Hafnium oxide films were deposited from tetrakis(dimethylamino)hafnium and water. Aluminum oxide films from trimethyl aluminum and water are being studied through simulations. In this work, both in situ metrologies and process models are being developed. Optically‐accessible ALD reactors have been constructed for in situ, high‐sensitivity Raman and infrared absorption spectroscopic measurements to monitor gas phase and surface species. A numerical model using computational fluid dynamics codes has been developed to simulate the gas flow and temperature profiles in the experimental reactor. Detailed chemical kinetic models are being developed with assistance from quantum chemical calculations to explore reaction pathways and energetics. This chemistry is then incorporated into the overall reactor models.

A numerical/experimental investigation of microcontamination in a rotating disk chemical vapor deposition reactor

Driven by a relentless decrease in feature size, the allowable particle contaminant size during semiconductor fabrication is now less than 100 nm. Particles in this size range (microcontaminants) in chemical vapor deposition (CVD) reactors are primarily gas-phase generated and are poorly understood. The purpose of the investigation described here is to enhance the understanding of the formation, transport and growth of microcontaminants in thermal CVD reactors. The approach being employed is to carry out a combined numerical/experimental study in which the particle dynamics are both modeled and optically probed in a rotating disk CVD reactor. The rotating disk configuration is utilized because of its simple and well-defined flow in which a particle layer forms in a highly accessible region of the reactor just above the substrate. Numerical/experimental comparisons of layer location and shape as a function of disk rotation rate are shown to be excellent if two empirically-determined parameters in the model...

A Microcontamination Model for Rotating Disk Chemical Vapor Deposition Reactors

This paper presents preliminary results from a model currently under development for gas-phase generated submicron-size contaminant particles (i.e., microcontaminants) in rotating disk chemical vapor deposition reactors. These particles present a problem during semiconductor processing, and this model is intended as a useful tool for gaining a better understanding of this problem. A one-dimensional formulation is employed to model the central section of the reactor, a technique which allows the use of detailed chemical reaction sets. The existing Sandia SPIN code, which contains a solver for the reacting flow, is modified by the addition of an aerosol model for the particles. This model utilizes a moment transport formulation which accounts for convection, diffusion, gravity, thermophoresis, chemical production, coagulation and condensation. Results are presented primarily in terms of reactor performance maps which indicate film growth and contamination rates as functions of substrate temperature. The eff...

In Situ Monitoring of Hafnium Oxide Atomic Layer Deposition

Atomic layer deposition (ALD) is increasingly being utilized as a method of depositing the thin (nanometer‐scale), conformal layers required for microelectronics applications such as high κ gate dielectric layers and diffusion barriers. However, significant process development issues remain for implementation of this technology in many applications. One potential solution to some process development issues is in situ monitoring. In situ monitoring of atomic layer deposition processes has the potential to yield insights that will enable efficiencies in film growth, in the development of deposition recipes, and in the design and qualification of reactors. However, demonstrations of in situ monitoring of actual atomic layer deposition processes are limited. In this work, the species present in the gas phase during atomic layer deposition of hafnium oxide were investigated in an attempt to gain insight into the chemistry of this system and evaluate potential in situ gas phase optical monitors. Hafnium oxide was deposited on a silicon substrate using tetrakis(ethylmethylamino) hafnium (TEMAH) and water as the hafnium and oxygen sources, respectively. In situ infrared absorption spectroscopic measurements were performed near the growth surface in a research‐grade, horizontal‐flow reactor under a range of deposition conditions. Density functional theory quantum calculations of vibrational frequencies of expected species were used to facilitate identification of observed spectral features.