SPRINTCAD QUARTERLY SUMMARY
October 1 - December 31, 1994
ORGANIZATION:
Stanford University
SUBCONTRACTORS:
none
PRINCIPAL INVESTIGATORS:
Robert W. Dutton, dutton@gloworm.Stanford.EDU, (650) 723-4138
Kincho H. Law, law@cive.Stanford.EDU, (650) 725-3154
Krishna Saraswat, saraswat@ee.Stanford.EDU, (650) 725-3610
Peter Pinsky, pinsky@ce.Stanford.EDU (650) 723-9327
PROJECT LEADER
Edwin C. Kan,
kan@gloworm.Stanford.EDU, (415) 723-9796
TITLE OF EFFORT:
"SPRINT-CAD"---Industry-Networked TCAD using Shared Parallel Computers
RELATED INFORMATION:
The URL for Stanford TCAD projects is:
http://www-tcad.stanford.edu
OBJECTIVE:
First-time capabilities to bridge solid modeling, FEM-based
parallel computation of fabrication processes and electrical analysis
of the resulting IC structures will be developed. Models needed to
represent diffusion, etching, deposition, oxidation and stress analysis
resulting from a sequence of process steps necessary in the creation
of electrical devices will be developed. This effort will provide a
radically new HPC framework for technology-based 3D process/device
modeling as well as realistic benchmarks to test HPC architectures
and software.
APPROACH:
We will build, integrate and test TCAD modules based on an
object-oriented approach that both develops and uses information
models in support of CFI-based standards. The modules and software
engineering methodology will be designed specifically to exploit
parallel computers and library components. The 3D process simulation
modules will utilize HPC platforms and provide new functional
capabilities for "computational prototyping" of the following key
technology fabrication steps:
- deposition/etching module---of special interest are CVD and plasma
assisted processes that result in high aspect ratio structures such as
trenches and filling/planarization of structures for metal
interconnects. Algorithmic work focuses on geometric manipulations and
surface evolution.
- thermal/stress analysis module---that can solve nonlinear
constitutive models for key process steps involving growth of
dielectric layers and impurity redistribution as well as the resulting
stress fields. Advanced formulations for finite elements are being
developed that support: parallel computation, adaptive gridding and
domain decomposition.
PROGRESS:
Solid geometry modeling can provide essential
communication between volume-mesh-based and surface-mesh-based
simulation tools by the consistent geometry representation. Also,
by various geometry manipulations, it can fast prototype 3D
structures when only 1D and 2D simulation or measurement results
are available for certain process. The geometry modeling utility
VIP 3D has been implemented using the ACIS
geometry engine and can be accessed through a new
minimal SWR specification. The 3D geometry
utility has been connected with FOREST 2D field server, and the
diffusion and oxidation (SUPREM) and the etching and deposition
(SPEEDIE)
tools have been connected. The connection to Stanford's
3D field server CAMINO is under way. The 3D meshing capabilities
in CAMINO have been improved to robustly include multi-regions.
For efficient device simulation, the anisotropic weight in
oct-tree based meshing has been investigated.
A new constitutive model for oxidation that considers finite
deformation as well as visco-elasticity has been derived. The
finite element formulation is thermodynamically consistent
and numerically stable. This constitutive model has been tested
in the FEAP (finite element analysis program) environment and the
elastic and viscous behaviors in the asymptotic limit have been
verified for correctness and stability.
An object-oriented information model
has been implemented to support FEM formulation and solution of
general systems of PDEs. Abstract classes of field element integration
and interpolation, operators for field-dependent constitutive models,
recombination models, and diffusion, direct and iterative sparse
linear solvers and flexible time-stepping schemes have been implemented
and validated with a Poisson solver and several 2D nonlinear
diffusion examples including segregation effects.
Hooks to Suprem IV structure files have been
provided for both validation and benchmarking purposes.
Physical modeling of etching/deposition based on Hamilton-Jacobi
equations has been demonstrated in 2D SPEEDIE. The surface evolution
based on the string model has been connected to the server-based
SUPREM OO7 architecture. A minimal SWR specification for the surface
mesh server has been defined from our experience with an integrated
example containing all process modeling steps. 2D Modeling of anisotropic
etch for deep trench technology has been successfully
demonstrated. Calibration with SEM measurements has
been performed to determine various parameters (such as sticking
coefficients). 3D SPEEDIE implementation and demonstration
examples have been examined.
RECENT ACCOMPLISHMENTS:
- Quasi-3D modeling for SRAM devices and
oxidation has been prototyped.
- The SUPREM OO7 heterogeneous tool integration environment has
been demonstrated with the minimal SWR specification with FOREST
as the geometry and field servers.
- The 3D meshing program CAMINO has demonstrated anisotropic gridding
capabilities. The algorithm will be presented in SISDEP 95.
- The new oxidation constitutive model based on finite-deformation
has been successfully demonstrated on asymptotic elastic and viscous
cases.
- The dial-an-operator finite-element program has been successfully
implemented for 2D static and transient problems. This module has
been added to the OO7 environment.
- The etch/depo tool SPEEDIE has been used for modeling deep trench
technology and add to the OO7 environment.
FY-`95 PLANS:
- Solid modeling functionality will be improved to support
integration across tools for both 2D and 3D---manipulation of SUPREM
(2D) data in the 3D information model will be demonstrated.
- The 2D implementation of the Dial-an-Operator code will be completed,
interfaced to FOREST and a 3D field server, and benchmarked against
existing impurity redistribution models in Suprem IV and PEPPER.
The combined 2D/3D version will be parallelized based on static
domain decomposition.
- The new constitutive oxidation model will use the finite element
representation in dial-an-operator and hence the integration and
parallelization efforts can be minimized.
- A 3D extension for SPEEDIE will be implemented with a selected
3D surface mesh server. Efficient methods for parallelization
will be investigated and implemented.
TECHNOLOGY TRANSITION:
At a SEMATECH sponsored workshop held in Santa
Clara during November, 1994 the member companies did not support
proposed efforts in further development of the SWR 2.0. As a result
the Stanford team in this project is now working directly with
partners such as IBM, ATT and Intel in the development and testing of
a "minimal" SWR that will provide the functionality needed to complete
the SPRINT-CAD project and to transition it into applications that
leverage directly from the industrial partnerships that will
ultimately be the end users. The further development and use of the
2D FOREST server will be carried out in collaboration with SRC and
several partners such as National Semiconductor and Motorola.
The object-oriented PDE solver has direct relevance in creating a new
module for process simulation that can be linked into the SUPREM OO7
architecture being developed under partial support from SRC. This
architecture and the incorporation of other existing 2D tools such as
SUPREM and SPEEDIE has been demonstrated at the International Electron
Devices Meeting (IEDM) as discussed below. This demonstration and
through industrial interactions in preparation for an SRC industrial
review to be held in February, 1995 the path for technology transition
for the Dial-an-Operator module will be discussed.
SIGNIFICANT EVENTS:
- STANFORD DEMONSTRATES NEW SERVER-BASED ARCHITECTURE
FOR SUPREM.
At the International Electron Devices Meeting (IEDM) held
in San Francisco during December, 1994 the Stanford SPRINT-CAD group
demonstrated the first prototype version of the Dial-an-Operator
module working in the context of a new architecture for the well-known
SUPREM process simulation tool. The agent-based process simulation
environment, now called OO7 to reflect the object-oriented approach,
supports a heterogeneous integration of tools. Shown at the
demonstration was inter-operability between the established SUPREM 4GS
and SPEEDIE tools along with the Dial-an-Operator module being
developed under the SPRINT-CAD project. In addition to the use of
tool agents for control, the demonstration included the use of FOREST
as an integrated server to support both geometry and gridding
operations in support of OO7.
Edwin C. Kan
kan@gloworm.stanford.edu
201 AEL, Stanford University, Stanford, CA 94305
Office: (415)723-9796
Fax: (415)725-7731
Date prepared: 1/27/95