Florida Object Oriented Process Simulator Research Articles

Effects of fluorine incorporation into β-Ga2O3

β-Ga2O3 rectifiers fabricated on lightly doped epitaxial layers on bulk substrates were exposed to CF4 plasmas. This produced a significant decrease in Schottky barrier height relative to unexposed control diodes (0.68 eV compared to 1.22 eV) and degradation in ideality factor (2.95 versus 1.01 for the control diodes). High levels of F (>1022 cm−3) were detected in the near-surface region by Secondary Ion Mass Spectrometry. The diffusion of fluorine into the Ga2O3 was thermally activated with an activation energy of 1.24 eV. Subsequent annealing in the range 350–400 °C brought recovery of the diode characteristics and an increase in barrier height to a value larger than in the unexposed control diodes (1.36 eV). Approximately 70% of the initial F was removed from the Ga2O3 by 400 °C, with the surface outgas rate also being thermally activated with an activation energy of 1.23 eV. Very good fits to the experimental data were obtained by integrating physics of the outdiffusion mechanisms into the Florida Object Oriented Process Simulator code and assuming that the outgas rate from the surface was mediated through fluorine molecule formation. The fluorine molecule forward reaction rate had an activation energy of 1.24 eV, while the reversal rate of this reaction had an activation energy of 0.34 eV. The net carrier density in the drift region of the rectifiers decreased after CF4 exposure and annealing at 400 °C. The data are consistent with a model in which near-surface plasma-induced damage creates degraded Schottky barrier characteristics, but as the samples are annealed, this damage is removed, leaving the compensation effect of Si donors by F− ions. The barrier lowering and then enhancement are due to the interplay between surface defects and the chemical effects of the fluorine.

Extraction of Migration Energies and Role of Implant Damage on Thermal Stability of Deuterium in Ga2O3

Deuterium was incorporated in bulk single crystal β-Ga2O3 samples by either ion implantation (100 keV, 1015 cm−2) or plasma exposure (up to 270°C, 30 mins) of single-crystal β-Ga2O3 and then its stability was examined as a function of annealing temperature. The mechanisms in both cases were extracted by fitting the experimental depth profiles using models within the Florida Object Oriented Process Simulator (FLOOPS) code. In the case of incorporation by implantation, annealing causes the deuterium to migrate toward the surface with simultaneous trapping at the residual implant damage. In the case of plasma incorporation where there is no residual damage, annealing causes outgassing of the deuterium from the surface, mediated through molecule formation. Very good fits to the experimental data are achieved by integrating physics of the outdiffusion mechanisms into the FLOOPS code. The rate constants and diffusivities used for the simulation were found to follow an Arrhenius dependence, with an activation energy for outdiffusion of deuterium of 1.22 eV and 2.58 eV for the release rate of deuterium from trap sites in the implanted case.

Deuterium incorporation and diffusivity in plasma-exposed bulk Ga2O3

Deuterium incorporation depths of ∼0.13–0.65 μm were obtained in bulk, single-crystal Ga2O3 during exposure to 2H plasmas for 0.5 h at 100–270 °C. The data were fit using the Florida Object Oriented Process Simulator simulation code. The estimated diffusivities were in the range of 2.3 × 10−14–6 × 10−13 cm2/V·s over this temperature range. The activation energy for diffusion was 0.30 ± 0.08 eV, suggesting that the diffusing deuterium migrates as an interstitial. The solubility of the deuterium exhibited an activation energy of 0.35 eV. Subsequent annealing at 500 °C removed ∼85% of the deuterium out of the Ga2O3, with an activation energy of 1.35 eV. This indicates that the outdiffusion is mediated through defects or molecule formation. The thermal stability of deuterium retention is lower (∼150 °C shift to lower temperatures) than when the deuterium is incorporated in the Ga2O3 by direct ion implantation, due to trapping at residual damage in the latter case. The incorporation depths of deuterium in the Ga2O3 are significantly less than in bulk ZnO under the same conditions, where we observed incorporation to ∼25 μm.

Implantation and Diffusion of Silicon Marker Layers in In0.53Ga0.47As

Continued effort has been placed on maximizing activation while controlling the diffusion of silicon doping in InGaAs for present and future complementary metal-oxide semiconductor devices. In order to explore the diffusion and activation behavior, Si marker layers were grown in InGaAs on InP by molecular beam epitaxy. The nature of Si diffusion was explored using a series of isoelectronic implants to introduce excess point defects near the layer. It was observed that excess interstitials reduce the Si diffusion consistent with a vacancy-driven diffusion mechanism. A diffusion and activation model implemented in the Florida object oriented process simulator has been developed to predict silicon diffusion behavior over a variety of temperatures and times. Using current and previous experimental data and complimentary density functional theory results, the diffusion model employs the SiIII–VIII pair as the primary mechanism for silicon diffusion in InGaAs.

Continuum Modelling of Silicon Diffusion and Activation in in0.53Ga0.47As

A continuum based-model of Silicon diffusion in ion-implanted InGaAs is presented. Several 5×1014 10keV Si+ implants into InGaAs were Rapid Thermal Annealed (RTA) at 750°C ranging from 10-40s. Box-like profiles characteristic of concentration-dependent diffusion were observed, similar to furnace anneals reported previously. Diffusivity extractions were carried out using FLOOPS (Florida Object Oriented Process Simulator), using point defect pairs, and Fermi level effects. Overall consideration of model fits and possible options for further development are suggested.

Modeling Two Dimensional Solid Phase Epitaxial Growth for Patterned Ge Substrates

Modeling the two-dimensional (2D)c solid phase epitaxial growth (SPEG) of amorphized Ge has become important due to the renewed interest in Ge as an alternative material in complementary metal-oxide-semiconductor (CMOS) devices. No one has modeled the two-dimensional 2D SPEG of amorphized Ge. In this work, a 2D SPEG model that uses level set techniques as implemented in the Florida object oriented process simulator (FLOOPS) to propagate regrowth fronts with variable crystallographic orientation patterned material is presented. Apart from the inherent orientation dependence of the SPEG velocity, it is established that nitride induced stress can affect mask edge defect formation for patterned samples. Data acquired from TEM experiments matches well with simulations, thus providing a stable model for simulating 2D regrowth and mask edge defect formation in Ge.

Modeling Solid Phase Epitaxial Growth for Patterned Ge Substrates

Modeling the two-dimensional (2D) solid phase epitaxial growth (SPEG) of amorphized Ge has become important due to the renewed interest in Ge as an alternative material to Si in complementary metal-oxide-semiconductor (CMOS) devices. In this work, a 2D SPEG model that uses level set techniques as implemented in the Florida object oriented process simulator (FLOOPS) to propagate regrowth fronts with variable crystallographic orientation patterned material is presented. Apart from the inherent orientation dependence of the SPEG velocity, it is established that nitride-induced stress can affect mask edge defect formation for patterned samples. Data acquired from transmission electron microscopy (TEM) experiments matches well with simulations, thus providing a stable model for simulating 2D regrowth and mask edge defect formation in Ge. In comparison to Si, mask edge defect formation is less apparent in Ge due to a more isotropic orientation dependence for Ge SPEG.

Level Set Modeling of Nickel Silicide Growth

ABSTRACTLevel set methods have been used for Solid phase epitaxial regrowth, etching and deposition.This study is to model the growth of nickel silicide accurately using the level set method. NiSi growth has been observed to follow a linear-parabolic law which takes into account both diffusion and interfacial reaction. This linear-parabolic system is very similar to the Deal and Grove model of SiO2 growth. This model uses similar diffusion transport and reaction rate equations. This simulation models the growth of silicide coupling diffusion solutions to level-set techniques. Dual level sets have been used for top and bottom interface propagation of silicide; velocities were estimated based on nickel concentrations at both interfaces as well as diffusivity and reaction rate of nickel. This is important to predict precise shape of silicide that will allow current crowding and field focusing effects to be modeled in transport out of the intrinsic device into the contacting layers. These simulation models can be used for latest technology nodes at 45, 32, 22nm and special devices such as FinFET’s etc. The level set method is successfully implemented and verified in Florida Object Oriented Process Simulator and growth shapes matches well with the literature Transmission Electron Microscopy data.

Modeling two-dimensional solid-phase epitaxial regrowth using level set methods

Modeling the two-dimensional (2D) solid-phase epitaxial regrowth (SPER) of amorphized Si (variously referred to as solid-phase epitaxial growth, solid-phase epitaxy, solid-phase epitaxial crystallization, and solid-phase epitaxial recrystallization) has become important in light of recent studies which have indicated that relative differences in the velocities of regrowth fronts with different crystallographic orientations can lead to the formation of device degrading mask edge defects. Here, a 2D SPER model that uses level set techniques as implemented in the Florida object oriented process simulator to propagate regrowth fronts with variable crystallographic orientation (patterned material) is presented. Apart from the inherent orientation dependence of the SPER velocity, it is established that regrowth interface curvature significantly affects the regrowth velocity. Specifically, by modeling the local SPER velocity as being linearly dependent on the local regrowth interface curvature, data acquired from transmission electron microscopy experiments matches reasonably well with simulations, thus providing a stable model for simulating 2D regrowth and mask edge defect formation in Si.

3-D analysis of semiconductor dopant distributions in a patterned structure using LEAP

This work presents the first 3-D analysis of lateral dopant diffusion in a patterned structure using a pulsed laser-assisted local electrode atom probe (LEAP). A structure similar to a device channel was created for this work by performing a 3 keV, 1×10 15 cm −2 As + implant on a poly-Si line patterned wafer with 70 nm line width and 200 nm line pitch. The wafer was subsequently annealed at 950 °C for 1 s. LEAP samples were made using a site-selective in-situ focused ion beam (FIB) process. The results from LEAP analysis were then compared with high-resolution transmission electron microscopy (HRTEM) and Florida object-oriented process simulator (FLOOPS) results. Good structural agreement was found between the LEAP and HRTEM results. Several 1-D As concentration profiles extracted from the LEAP data were also found to be in good agreement with FLOOPS process simulation results. These profiles also represent for the first time that results from a 3-D process simulator have been able to be confirmed experimentally using a single sample.

The Role of Ion Mass on End-of-Range Damage in Shallow Preamorphizing Silicon

ABSTRACTPreamorphization is commonly used to form shallow junction in silicon CMOS devices. The purposeof this experiment was to study the effect of the preamorphizing species' mass on the interstitial concentration at the end-of-range (EOR). Isovalent species of Si, Ge, Sn and Pb were compared. Silicon wafers with a buried boron marker layer (4700 Å deep) were amorphized using implants of 22 keV 28Si+, 32 keV73Ge+, 40 keV 119Sn+ or 45 keV 207Pb+, which resulted in similar amorphous layer depths. All species were implanted at a dose of 5×1014 /cm2. Cross-sectional transmission electron microscopy (XTEM) was used tomeasure amorphous layer depths (approximately 400 Å). Post-implantation anneals were performed at 750 °C for 15 minutes. Plan-view transmission electron microscopy (PTEM) was used to observe and quantify the EOR defect population upon annealing. Secondary ion mass spectrometry (SIMS) was used to monitor the transient enhanced diffusion (TED) of the buried boron marker layer resulting from the EOR damage introduced by the amorphizing implants. Based upon the SIMS results Florida Object Oriented Process Simulator (FLOOPS) calculated the resulting time average diffusivity enhancements. Results showed that increasing the ion mass over a significant range (28 to 207 AMU) not only affects the quantity and type of damage that occurs at the EOR, but results in a reduced diffusivity enhancement.

Continuum based modeling of silicon integrated circuit processing: An object oriented approach

Continuum based integrated circuit process modeling is the dominant tool used to investigate and understand integrated circuit (IC) development. This paper describes the commonly used models for implantation, diffusion, and material growth. In addition, the supporting numerical techniques are described. This paper focuses on the implementation in object oriented code, Florida Object Oriented Process Simulator (FLOOPS). The software architecture is described for implementing models and numerics. A number of process examples are introduced and discussed.

2-D mesh adaption and flux discretizations for dopant diffusion modeling

Mesh generation and adaption for solving dopant diffusion in process simulation is a difficult task, complicated by both the moving boundaries of oxide growth and the time dependence of the solutions. For both computational and ease of use reasons, automatic mesh generation and discretization error control is desirable. This paper describes an approach based on local error estimates to refine the mesh. The results presented in this paper extend our previous work to two-dimensional problems. The implementation of this approach is done in a process simulator, the Florida Object-Oriented Process Simulator. Several test cases are described to demonstrate the effectiveness of the algorithms.