Silvaco Group, Inc. · NASDAQ:SVCO

Silvaco — Technical Presentation (December 2007)

39 slides from Silvaco's technical presentation, December 2007

  1. Slide 1: VICTORY Process: Full Physical 3D Semiconductor Simulator

    VICTORY Process Full Physical 3D Semiconductor Simulator

  2. Slide 2: VICTORY Process - 3D Process Simulator

    VICTORY Process - 3D Process Simulator  VICTORY Process integrates four core modules  Etching and Deposition modules  Oxidation and Stress module  Implantation module  Diffusion module  Self explanatory process flow description  Based on ATHENA syntax  Open interface for modelling  Model pa

  3. Slide 3: VICTORY Process - Numerics

    VICTORY Process - Numerics  Level set based framework  for structure representation and  for interface propagation  Very stable surface propagation algorithms  Important for Etching, Deposition, Oxidation  Based on the multi-layer structure concept  Automatic void detection  Automatic loop c

  4. Slide 4: VICTORY Process - Meshes

    VICTORY Process - Meshes  VICTORY Process uses two types of meshes  Hierarchical Cartesian mesh for implicit geometry representation  Irregular Cartesian mesh for volume data discretization  Doping, Stress, etc.  Manual and automatic mesh refinement  Manual refinement of the geometry mesh  Au

  5. Slide 5: Etching / Deposition Module

    VICTORY Process - Etching / Deposition Module  Comprehensive mask support  GDSII – format masks  Lay – format masks (MaskViews)  Definition of mask polygons inside the processing deck  Geometrical Etching  Idealized directional mask pattern transfer  Pattern transfer with tilted sidewalls  I

  6. Slide 6: Etching / Deposition Module (con't)

    VICTORY Process - Etching / Deposition Module (con’t)  Geometrical Deposition  Idealized vertical resists or material regions defined by a mask  Idealized conformal deposition  Deposition of features with tilted sidewalls VICTORY Process - Page 6

  7. Slide 7: Etching / Deposition Module (con't)

    VICTORY Process - Etching / Deposition Module (con’t)  Geometrical CMP  Idealized planarisation VICTORY Process - Page 7

  8. Slide 8: Etching / Deposition Module (con't)

    VICTORY Process - Etching / Deposition Module (con’t)  Physical Etching  Selective etching of several materials  Comprehensive set of default models  Selective Isotropic Etching  Selective Anisotropic Etching  Selective Directional Etching  Plasma Etching  Ion-Milling  Material dependent yi

  9. Slide 9: Etching / Deposition Module (con't)

    VICTORY Process - Etching / Deposition Module (con’t) Ion milling with re-deposition VICTORY Process - Page 9

  10. Slide 10: Etching / Deposition Module (con't)

    VICTORY Process - Etching / Deposition Module (con’t)  Physical Deposition  Comprehensive set of default models  Conformal Deposition  Non-conformal Deposition  Directional Deposition  Selective Deposition  No deposition on selected materials  Ion beam deposition  User accessible yield func

  11. Slide 11: Etching / Deposition Module (con't)

    VICTORY Process - Etching / Deposition Module (con’t) Ion beam deposition with tilted beam VICTORY Process - Page 11

  12. Slide 12: Etching / Deposition Module (con't)

    VICTORY Process - Etching / Deposition Module (con’t)  Physical Etching and Deposition - Open modelling system  C-Interpreter based  User definable surface reactions (surface reaction models)  What contributes to the local etch rate ?  User definable transport characteristics (flux models)  Ho

  13. Slide 13: Modes for Flux Calculation

    Modes for Flux Calculation Primary_Ballistic Mode:  Reactants are approaching from the reactor only  The velocity distribution function is a modelling function definable by the user (C-Interpreter) VICTORY Process - Page 13

  14. Slide 14: Modes for Flux Calculation (con't)

    Modes for Flux Calculation (con’t) Reflective_Ballistic Mode:  Reactants are approaching from the reactor and are also emitted from the surface (either due to reflection or reaction)  The primary distribution and the emission distribution are modelling functions definable by the user (C-Interprete

  15. Slide 15: Modes for Flux Calculation (con't)

    Modes for Flux Calculation (con’t) Rate Coupled - Reflective_Ballistic Mode:  Reactants are approaching from the reactor and are also emitted from the surface  The amount of particles emitted from the surface is a function of the rate (e.g. re-deposition)  The primary distribution and the emissio

  16. Slide 16: Implantation Module

    VICTORY Process – Implantation Module  Analytical and Monte-Carlo method for ion implantation  Monte Carlo method is more appropriate in 3D because many effects cannot be accounted for by even the most elaborate analytical procedure  Monte Carlo method takes into account all important implantatio

  17. Slide 17: MC Implantation Module

    VICTORY Process – MC Implantation Module  Completely physically based  Uses best available electronic and nuclear stopping models  Acceleration techniques  Trajectory splitting VICTORY Process - Page 17

  18. Slide 18: MC Implantation Module (con't)

    VICTORY Process – MC Implantation Module (con‘t) Implant at tilt 50 degrees and twist along structure’s diagonal VICTORY Process - Page 18

  19. Slide 19: MC Implantation Module (con't)

    VICTORY Process – MC Implantation Module (con‘t) Ion distribution in silicon Primary, i.e. direct impact implantation Shadowed, i.e. secondary impact implantation Implant at tilt 50 degrees and twist along structure’s diagonal VICTORY Process - Page 19

  20. Slide 20: Annealing/Diffusion Module

    VICTORY Process – Annealing/Diffusion Module  Compatible with Fermi diffusion model of ATHENA  Solid solubility and impurity segregation at material interfaces are taken into account  Concurrent simulation of several impurities  Finite difference scheme is applied  Highly efficient for numerica

  21. Slide 21: Annealing/Diffusion Module (con't)

    VICTORY Process – Annealing/Diffusion Module (con‘t) Mesh for annealing simulation Material boundary with interface nodes Structured nodes VICTORY Process - Page 21

  22. Slide 22: Annealing/Diffusion Module (con't)

    VICTORY Process – Annealing/Diffusion Module (con‘t)  Efficient linear/non-linear solvers  Space matrix structure is taken into account according to number of dopants/equations  Time integration  One-step Euler scheme  Step size control algorithm  Finite difference method  Conservative monoto

  23. Slide 23: Implant/Diffusion – After Monte Carlo Implant

    Implant/Diffusion – After Monte Carlo Implant r 30 keV Arsenic Implant, Tilt = 7 degrees 5 million trajectories Took 80 minutes of Simulation time VICTORY Process - Page 23

  24. Slide 24: Implant/Diffusion: After Anneal

    Implant/Diffusion: After Anneal 10 minutes 950 C Anneal Fermi Model Simulation time: < 2 minutes VICTORY Process - Page 24

  25. Slide 25: Oxidation/Stress Module

    VICTORY Process – Oxidation/Stress Module  Demanding numerically process:  Involves different processes  Oxygen diffusion  Chemical reaction: Si + O2 = SiO2  Material deformation  Processes are interlinked  Diffusion and reaction depends on oxide flow (via stresses)  Oxide flow depends on di

  26. Slide 26: Oxidation/Stress Module (con't)

    VICTORY Process – Oxidation/Stress Module (con‘t)  Oxidation step is decomposed as a succession of the basic processes :  Diffusion of oxygen through the oxide  Solving diffusion equation  Reaction at SiO2/Si interface  Volume expansion  Interface propagation  Solving moving boundary problem

  27. Slide 27: Oxidation/Stress Module (con't)

    VICTORY Process – Oxidation/Stress Module (con‘t)  Linear viscous model : µ ∂σ ij  ∂Vi ∂V j  ∇ ⋅ D∇C = 0 σ ij + = µ ⋅ +  G ∂t    ∂x j ∂xi  ∂C ∂C Fsource = − D = h(C0 − C )O / SiO Fsink = − D ⋅ = k ⋅ CSiO2 / Si ∂n 2 2 ∂n  ED  D(T ) = D0 ⋅ exp −   kB ⋅T   Eµ  µ (T ) = µ 0 ⋅ exp −

  28. Slide 28: Oxidation/Stress Module (con't)

    VICTORY Process – Oxidation/Stress Module (con‘t)  Oxygen transport, material flow and moving boundary problem are solved on the same Cartesian mesh  High stability  Special boundary nodes are added to implement proper boundary conditions Introduce cross-points with the interface Formulate standa

  29. Slide 29: Oxidation/Stress Module (con't)

    VICTORY Process – Oxidation/Stress Module (con‘t) LOCOS: Long simulations are stable without meshing issues VICTORY Process - Page 29

  30. Slide 30: Oxidation/Stress Module (con't)

    VICTORY Process – Oxidation/Stress Module (con‘t) Poly-silicon oxidation : Simulation is stable for complex structures (Multi-layer concept) VICTORY Process - Page 30

  31. Slide 31: Oxidation/Stress Module (con't)

    VICTORY Process – Oxidation/Stress Module (con‘t) iso-surfaces of pressure Trench oxidation: Can resolve complex tensile and compressive stress near corners VICTORY Process - Page 31

  32. Slide 32: Applications: LOCOS

    Applications: LOCOS VICTORY Process - Page 32

  33. Slide 33: Applications: Re-Oxidation - STRESS

    Applications: Re-Oxidation - STRESS VICTORY Process - Page 33

  34. Slide 34: Applications: Poly Reoxidation - Regions

    Applications: Poly Reoxidation - Regions VICTORY Process - Page 34

  35. Slide 35: Applications: Poly Reoxidation – Stresses in Oxide

    Applications: Poly Reoxidation – Stresses in Oxide VICTORY Process - Page 35

  36. Slide 36: Applications: FinFET Gatestack Doping

    Applications: FinFET Gatestack Doping VICTORY Process - Page 36

  37. Slide 37: Applications: MEMS Cantilever Structure

    Applications: MEMS Cantilever Structure VICTORY Process - Page 37

  38. Slide 38: Current Development

    Current Development  3D modeling of stress evolution  Stress-dependent viscous oxidation  Improvements of adaptive meshing for effective handling of variable scale structures  Extended library with user-defined etch and deposition models  S Selective epitaxy growth simulation  Directional wet

  39. Slide 39: Conclusion

    Conclusion  VICTORY Process is the next generation tool for the investigation of 3D process and device effects  Suitable for many semiconductor technologies such as  FinFET,  Sensor and LED’s  MEMS  Complex quantum structures  Suitable for non-semiconductor applications such as  Hard coating

Silvaco Group — Technical Presentation (December 2007) on 6ix