Silvaco Group, Inc. · NASDAQ:SVCO

Silvaco — Technical Presentation (November 2012)

42 slides from Silvaco's technical presentation, November 2012

  1. Slide 1: Victory Process: Etching and Deposition Simulation

    Victory Process Full Physical 3D Semiconductor Simulator Etching and Deposition Simulation

  2. Slide 2: Victory Process – 3D Process Simulator

    Victory Process – 3D Process Simulator • Victory Process provides the capability to simulate comprehensive full process flows • Etching, Deposition • Oxidation, Stress • Implantation • Diffusion • Self explanatory process flow description • Open interface for modeling • Model parameters and function

  3. Slide 3: Level Set Framework

    Victory Process – Level Set Framework • The structure is represented as a set of implicit surfaces • Hierarchical Cartesian meshes are used to improve the accuracy around critical areas • Support for automatic and manual mesh refinement Victory Process – Full Physical 3D Semiconductor Simulator Etch

  4. Slide 4: Level Set Framework

    Victory Process – Level Set Framework • Very stable surface propagation algorithms • Automatic void detection • Avoids the problem of loops creation and correction Victory Process – Full Physical 3D Semiconductor Simulator Etching and Deposition Simulation -4-

  5. Slide 5: Etching / Deposition Modes

    Victory Process – Etching / Deposition Modes • Geometrical mode • Physical mode • Numerical error is limited by the mesh • Simulates real physical processes size only • Accurately handles complex shading • Orders of magnitude faster than physical and visibility effects simulation of corresponding pr

  6. Slide 6: Geometrical Mode

    Victory Process – Geometrical Mode • Comprehensive mask support • GDSII – format masks • lay – format masks (MaskViews) • Definition of mask polygons inside the processing deck • Mask variations via the deck (shrink and expand) • Selection of a simulation window Victory Process – Full Physical 3D Se

  7. Slide 7: Geometrical Mode: Lithography

    Victory Process – Geometrical Mode • Lithography • Calculation of aerial images • Pattern transfer of aerial images Aerial Image Transferred Pattern Mask Layer Victory Process – Full Physical 3D Semiconductor Simulator Etching and Deposition Simulation -7-

  8. Slide 8: Geometrical Mode: Geometrical etching

    Victory Process – Geometrical Mode • Geometrical etching • Idealized directional mask pattern or image transfer • Pattern transfer with tilted sidewalls and rounded corners With Tilted Sidewalls Ideal Pattern Transfer and Rounded Corners With Tilted Sidewalls Victory Process – Full Physical 3D Semic

  9. Slide 9: Geometrical Mode: Wet and dry etching

    Victory Process – Geometrical Mode • Geometrical etching • Idealized wet and dry etching • Selective and non-selective mode Wet Etching Initial Structure Dry Etching Wet Selective Etching Victory Process – Full Physical 3D Semiconductor Simulator Etching and Deposition Simulation -9-

  10. Slide 10: Geometrical Mode: Geometrical CMP

    Victory Process – Geometrical Mode • Geometrical CMP • Idealized planarization • Selective and non-selective mode Non-Selective Selective Victory Process – Full Physical 3D Semiconductor Simulator Etching and Deposition Simulation - 10 -

  11. Slide 11: Geometrical Mode: Geometrical deposition

    Victory Process – Geometrical Mode • Geometrical deposition • Idealized vertical resists or material regions defined by a mask • Idealized conformal deposition • Deposition of features with tilted sidewalls and rounded corners • Planar mode to partially fill holes Conformal Deposition Planar Deposit

  12. Slide 12: Geometrical Mode - Summary

    Victory Process – Geometrical Mode - Summary • Set of models for fast structure manipulation • Based on idealized processing steps • Used for • Fast structure prototyping • Creation of the initial shapes for subsequent physical analysis Victory Process – Full Physical 3D Semiconductor Simulator Etch

  13. Slide 13: Reactor-Scale vs. Feature-Scale

    Victory Process – Physical Mode Reactor-Scale vs. Feature-Scale • The numerical engine of Victory Process only operates on the feature scale level • Ballistic transport within the simulation domain is assumed • Constant particle properties within the simulation domain are assumed • Particle-particle

  14. Slide 14: Numerical engine and open model library

    Victory Process – Physical Mode • Numerical engine : • Calculates the amount of reactants reaching the surface from the reactor domain • Takes into account secondary effects • Re-deposition of removed material • Reflection of reactants • Calculates the surface propagation • Open model library (acces

  15. Slide 15: Boundary conditions

    Victory Process – Physical Mode • Boundary conditions : • The structure is symmetrically and periodically extended in X and Y directions. This is necessary to properly take into account secondary effects. • The number of 'reflections' depends on the desired redeposition accuracy • Shading effects an

  16. Slide 16: Etching models without particle flux

    Victory Process – Physical Mode • Etching models without particle flux • Particle flux is not taken into account • No visibility and shading effects are taken into account • Selective etching capability • Isotropic etching model • Anisotropic etching model Selective Isotropic Etching Initial Structu

  17. Slide 17: Deposition models without particle flux

    Victory Process – Physical Mode • Deposition models without particle flux • Particle flux is not taken into account • No visibility and shading effects are taken into account • Selective deposition capability • Conformal deposition model • Non-conformal deposition model Non-Conformal Selective Depos

  18. Slide 18: Etching and deposition models with a single primary particle

    Victory Process – Physical Mode • Etching and deposition models with a single primary particle • Only the flux of a single particle coming from the reactor is taken into account • Full consideration of visibility and shading effects • The spacial velocity distribution of the particles coming from th

  19. Slide 19: Etching models with single primary particle, no reflection

    Victory Process – Physical Mode • Etching models with a single primary particle where surface reflection is neglected • For these models a high sticking efficiency is implicitly assumed • Hence surface reflection can be neglected • Selective etching capability • The etch rate is a linear function of

  20. Slide 20: Directional etching model

    Victory Process – Physical Mode • Directional etching model • Is a single primary particle etching model • The velocity vector of all particles coming from the reactor is identical and by default perpendicular to the plane wafer surface Initial Structure Selective Directional Etching Victory Process

  21. Slide 21: Primary etching model

    Victory Process – Physical Mode • Primary etching model • Is a single primary particle etching model • The spacial velocity distribution of the particles can vary from an isotropic distribution (default) to a highly focused distribution • Width of the distribution function may be used as a parameter

  22. Slide 22: RIE etching model

    Victory Process – Physical Mode • RIE etching model • Is a single primary particle etching model • The two physical particles (ion and neutral) are superimposed in one flux distribution • This is possible because secondary fluxes are neglected and identical surface interaction (reaction) properties

  23. Slide 23: RIE etching model

    Victory Process – Physical Mode • RIE etching model • The spacial velocity distribution of the ions is highly focused • Von Mises spacial velocity distribution is applied • The standard deviation is used as a parameter • The spacial velocity distribution of the neutral is isotropic • Ratio between t

  24. Slide 24: RIE etching model

    Victory Process – Physical Mode • RIE etching model Initial structure Etching with RIE Model Profile Sensitivity to RIE Model Parameters (ion focus, ion/neutral ratio) Victory Process – Full Physical 3D Semiconductor Simulator Etching and Deposition Simulation - 24 -

  25. Slide 25: Deposition models with single primary particle, no reflection

    Victory Process – Physical Mode • Deposition models with a single primary particle where surface reflection is neglected • For these models a high sticking efficiency is implicitly assumed • Hence surface reflection can be neglected • Selective deposition is possible • Directional deposition model •

  26. Slide 26: Directional deposition model

    Victory Process – Physical Mode • Directional deposition model • Is a single primary particle deposition model • The velocity vector of all particles coming from the reactor is identical and by default perpendicular to the plane wafer surface • Particle direction may be used as a parameter Initial S

  27. Slide 27: Primary deposition model

    Victory Process – Physical Mode • Primary deposition model • Is a single primary particle deposition model • The spacial velocity distribution of the particles can vary from an isotropic distribution (default) to a highly focused distribution • Width of the distribution function may be used as a par

  28. Slide 28: Ion beam deposition models

    Victory Process – Physical Mode • Ion beam deposition models • Single primary particles are considered • Static and rotating beams • Beam can temporarily be switched off during rotation • Ideally focused and Gaussian shape • Beam with divergence • Beam shape accessible via open model library • Mater

  29. Slide 29: Ion beam deposition models

    Victory Process – Physical Mode • Ion beam deposition models Initial Structure Ion Beam Deposition With Single Directional Beam Victory Process – Full Physical 3D Semiconductor Simulator Etching and Deposition Simulation - 29 -

  30. Slide 30: Etching models with surface reflection

    Victory Process – Physical Mode • Etching models with a single primary particle where surface reflection is taken into account • Selective etching capability • Material specific sticking efficiencies • The etch rate is a linear function of the local particle flux • Re-emission etching model Victory

  31. Slide 31: Re-emission etching model

    Victory Process – Physical Mode • Re-emission etching model • Is a single particle etching model • Spacial primary velocity distribution of the particles can vary from an isotropic distribution (default) to a highly focused distribution • Width of the distribution function may be used as a parameter

  32. Slide 32: Re-emission etching model

    Victory Process – Physical Mode • Re-emission etching model Selective Etching with Initial Structure Re-emission Etching Model Effect of Varying Sticking Efficiencies Victory Process – Full Physical 3D Semiconductor Simulator Etching and Deposition Simulation - 32 -

  33. Slide 33: Etching models with emission of etched material

    Victory Process – Physical Mode • Etching models with a single primary particle where emission of etched material is taken into account • Ion Milling Etching Models • Static and rotating beams • Beam can be temporarily switched off during rotation • Ideally focused or Gaussian beam shape • Beam with

  34. Slide 34: Ion milling etching models

    Victory Process – Physical Mode • Ion milling etching models beam direction Initial Structure Static beam Redeposition of alloy Selective deposition efficiency After Ion Milling Material specific mill rate functions Victory Process – Full Physical 3D Semiconductor Simulator Etching and Deposition Si

  35. Slide 35: Deposition models with surface reflection

    Victory Process – Physical Mode • Deposition models with a single primary particle where surface reflection is taken into account • Selective deposition capability • The deposition rate is a linear function of the local particle flux • Reemission deposition model Victory Process – Full Physical 3D S

  36. Slide 36: Reemission deposition model

    Victory Process – Physical Mode • Reemission deposition model • Is a single particle deposition model • Spacial primary velocity distribution of the particles can vary from • an isotropic distribution (default) to • a highly focused distribution • Width of the distribution function may be used as a

  37. Slide 37: Reemission Deposition Model

    Victory Process – Physical Mode • Reemission Deposition Model Initial Structure Vary Sticking Efficiency Deposition with Re-emission Model Victory Process – Full Physical 3D Semiconductor Simulator Etching and Deposition Simulation - 37 -

  38. Slide 38: Advanced etching models with multiple primary particles

    Victory Process – Physical Mode • Advanced etching models with multiple primary particle • Multiple primary particles • Some particles may be reflected • Material specific sticking efficiencies • Material specific surface reaction properties • IECE (Ion Enhanced Chemical Etching) model Victory Proce

  39. Slide 39: Ion enhanced chemical etching model

    Victory Process – Physical Mode • Ion enhanced chemical etching model • Two particle model where ions and neutral coming from the reactor are taken into account • Neutrals – Chemically active, uncharged particles • Ions – Accelerated charged particles • Neutrals are chemically reacting at the surfac

  40. Slide 40: Ion enhanced chemical etching model

    Victory Process – Physical Mode • Ion enhanced chemical etching model • e.g Silicon Etching in SFx Plasma 1. The neutrals (F) chemically attach themselves SiFx removed to Si surface (dangling bonds) Ions by ions 2. Newly formed SiFx molecules cover the F SiFx removed by surface preventing further re

  41. Slide 41: Ion enhanced chemical etching model

    Victory Process – Physical Mode • Ion enhanced chemical etching model • Ions decrease the surface coverage at the bottom of the trench Trench aspect ratio increases with ion energy Victory Process – Full Physical 3D Semiconductor Simulator Etching and Deposition Simulation - 41 -

  42. Slide 42: Summary

    Victory Process – Summary • Victory Process is a powerful tool for simulating 3D structure transformations by etching and deposition processes • Command deck syntax based on technological processes • Very robust numerical algorithms for geometrical transformations • Ability for rapid structure proto

Silvaco Group — Technical Presentation (November 2012) on 6ix