2D Trench Deposition with SingleParticleProcess (CVD)

This tutorial shows a simple 2D Chemical Vapor Deposition (CVD) example in ViennaPS:

  1. Create a 2D simulation domain
  2. Generate an initial trench geometry
  3. Define a single-particle CVD deposition model
  4. Add a new material layer (SiO₂)
  5. Run the deposition process
  6. Save and visualize results

Download Jupyter Notebook: 07_single_particle_cvd.ipynb

1. Import ViennaPS 

import viennaps as ps  # ViennaPS 2D

This imports the ViennaPS Python bindings. (Here we run a 2D example. No explicit setDimension is used.)

2. Create the 2D Simulation Domain 

extent = 30
gridDelta = 0.3

domain = ps.Domain(xExtent=extent, gridDelta=gridDelta)

Parameters

  • xExtent: lateral size of the simulation window (in x-direction)
  • gridDelta: grid spacing (controls resolution and runtime)

This sets up a 2D computational domain with a resolution of 0.3.

3. Generate the Initial Trench Geometry

We create a rectangular trench that acts as the starting topography.

ps.MakeTrench(domain, trenchWidth=10.0, trenchDepth=30.0).apply()

Parameters

  • trenchWidth = 10 → opening width at the top
  • trenchDepth = 30 → depth of the trench

After this step, the domain contains a trench etched into the substrate.

4. Save the Initial Geometry 

domain.saveVolumeMesh("singleParticleCVD_1")

This exports the current geometry to a VTU file. It is useful for visual inspection and for documenting intermediate steps.

5. Define the CVD Deposition Model

Now we define a simple single-particle deposition model.

model = ps.SingleParticleProcess(
    rate=1.0,
    stickingProbability=0.1,
    sourceExponent=1.0,
)

Model parameters

  • rate: base deposition rate on a flat surface

  • stickingProbability: probability that a particle deposits when it hits the surface

    • smaller values usually increase transport into deep features

  • sourceExponent: controls the angular distribution of incoming particles

    • 1.0 is a common “isotropic-like” choice in simple examples

This model represents deposition by particles traveling through the trench and sticking on impact.

6. Add a New Material Layer (SiO₂)

Before deposition, we create a new top-level surface assigned to the deposited material.

domain.duplicateTopLevelSet(ps.Material.SiO2)

This tells ViennaPS that the evolving deposited layer is SiO₂. Deposition will then grow this new material on top of the existing geometry.

7. Run the Deposition Process 

processDuration = 10.0
ps.Process(domain, model, processDuration).apply()
  • processDuration is the simulated time interval
  • During this time, the geometry evolves as SiO₂ deposits on exposed surfaces

  • In a trench, this typically leads to:

    • stronger growth near the opening (more particle access)
    • reduced growth deeper inside (limited transport), depending on sticking

8. Save and Visualize the Result 

domain.saveVolumeMesh("singleParticleCVD_2")
domain.show()
  • saveVolumeMesh() exports the final geometry after deposition
  • show() visualizes the resulting topography

Download Jupyter Notebook: 07_single_particle_cvd.ipynb