Hands-on workshop: The Amsterdam Modeling Suite on SURF Research Cloud, 13 May 2026

This hands-on workshop introduces the Amsterdam Modeling Suite (AMS) on SURF Research Cloud through three guided computational materials science exercises. Participants will learn how to launch and manage AMS calculations in the cloud, set up practical workflows in the AMS graphical interface, inspect results, and connect simulation output to physical properties. The examples cover battery materials, semiconductor processing, and heterogeneous catalysis.

Schedule

  • 13:00 – 13:50 Introduction to SURF Research Cloud and AMS
    Overview of the workshop environment, launching AMS, managing jobs, using AMSinput/AMSjobs/AMSmovie, and organizing calculations during the session.
  • 13:50 – 14:00 Break
  • 14:00 – 14:50 Exercise 1: Battery cathodes
    Compute structural and energetic properties for a lithium-ion battery cathode material.
  • 14:50 – 15:00 Break
  • 15:00 – 15:50 Exercise 2: Atomic layer etching
    Set up and inspect a reactive molecular dynamics workflow for HF interaction with a silica surface.
  • 15:50 – 16:00 Break
  • 16:00 – 17:00 Exercise 3: Catalytic reaction on Pt(111)
    Explore a few adsorption structures and set up nudged elastic band calculation for a CO oxidation step.

Exercise 1: Battery Cathodes – Intercalation Potential and Volume Change

In this exercise, participants will study LiFePO4 as a model battery cathode material. The goal is to compare the lithiated and delithiated crystal structures and use their optimized energies and volumes to estimate two practical descriptors: the average lithium intercalation potential and the volume change upon delithiation.

Participants will:

  1. Set up a lithium metal reference calculation.
  2. Optimize the LiFePO4 crystal structure [10 min].
  3. Remove lithium atoms to create the delithiated FePO4 structure.
  4. Optimize the delithiated structure using the same computational settings.
  5. Record total energies and optimized cell volumes.
  6. Use these values to estimate the intercalation potential and relative volume change.
  7. [Advanced] Design a better materials than LiFePO4.

The emphasis will be on understanding the workflow: keeping computational settings consistent, extracting the relevant output quantities, and organizing results. If time permits, we will briefly discuss how the same approach can be adapted to doped cathodes or related sodium materials.

[Solution]

Exercise 2: Atomic Layer Etching of SiO2

In this exercise, participants will model the first part of an atomic layer etching process for SiO2, where HF molecules interact with a hydroxylated silica slab. The calculation uses a reactive force field molecular dynamics setup with a molecule gun, so participants can see how reactive surface simulations are prepared and inspected in AMS.

Participants will:

  1. Load a prepared hydroxylated SiO2 slab.
  2. Define regions for the frozen support layer and the thermostatted active slab region.
  3. Relax the slab with ReaxFF.
  4. Add an HF projectile molecule and assign it to a separate region.
  5. Configure a molecule-gun molecular dynamics calculation.
  6. Add molecule removal regions to keep the simulation cell manageable.
  7. Run [17 min] or inspect the trajectory in AMSmovie.
  8. Look for impact events, surface reactions, bond breaking, bond formation, and desorbing fragments.
  9. [Advanced] Compute the etching rate.

The focus will be on the setup logic rather than production-scale etching statistics. We will discuss why regions, thermostats, frozen atoms, projectile insertion, and sink boxes are useful in this type of simulation.

[Solution]

Exercise 3: CO Oxidation on Pt(111) [advanced]

In this exercise, participants will study elementary steps relevant to CO oxidation on a Pt(111) catalyst surface. To keep the exercise suitable for the workshop time, we will calculate only a small number of adsorption structures and then set up one nudged elastic band (NEB) calculation for a possible CO* + O2* to CO2 + O* reaction step.

Participants will:

  1. Build or load a Pt(111) slab model.
  2. Identify common surface adsorption sites such as top, bridge, fcc hollow, and hcp hollow.
  3. Prepare a few selected adsorbate structures, for example O*, CO*, O2*, and one CO* + O2* coadsorbed structure.
  4. Freeze the bottom Pt layers and optimize the selected adsorption structures [2 min / structure].
  5. Compare relaxed geometries and total energies at a qualitative level.
  6. Choose one coadsorbed initial state for a reaction-path calculation.
  7. Build a simple final-state guess for CO2 formation while keeping atom ordering consistent.
  8. Set up and inspect one NEB calculation [27 min].
  9. [Advanced] Compute the second NEB path and identify most probable reaction.

The emphasis will be on the practical workflow for surface chemistry: building a slab, placing adsorbates at meaningful sites, applying slab constraints, comparing adsorption geometries, and preparing consistent initial and final structures for NEB. The NEB result will be treated as a first-pass reaction-path estimate that can later be refined with more images, improved endpoints, or more extensive sampling.

[Solution]