EDS Simulation Scripts for NIST DTSA-II

This repository contains Jython script templates for NIST DTSA-II (introduction video) for simulating energy-dispersive x-ray spectroscopy (EDS) spectra via Monte Carlo method.

DTSA-II is maintained by Nicholas Ritchie. Please cite relevant papers if the simulations help your research (e.g., https://doi.org/10.1017/S1431927609990407).

The scripts can help in getting a better understanding of EDS physics (generation and detection) and aid in planning experiments:

• Check for critical EDS peak overlaps
• Check feasibility of trace-element detection (for a given experimental conditions)
• Check effect of different primary beam energies and doses
• Check effect of a detector window on low-energy x-ray absorption

The source codes for DTSA-II and the EPQ Library are available through GitHub.

Example

Simulation of a 70 nm thick TEM sample of GdBa2Cu3O7 at 200 keV with and without added 12 mol% BaHfO3. Hf-L lines overlap with Cu-K lines and create shoulders on the strong Cu peak. A Hf-M line peak is visible at around 2 keV. No detector window was used here.

This example is related to this paper (see Fig. S5 for an experimental example of the Hf-L peak overlap with Cu-K).

How It Works

NIST DTSA-II uses Monte Carlo method to simulate electron scattering and x-ray generation and transport through the sample (and detector window). DTSA-II models electron scattering and (secondary) x-ray generation/absorption quite well, but descrepancies are to expected especially for "lesser used" low-energy x-ray transitions (e.g., M series). Have a look at the simulation video.

Usage

NIST DTSA-II comes installed with a Jython console. The scripts are loaded and executed within DTSA-II where the simulated spectra are conventiently displayed.

The steps to set up DTSA-II:

1. Install NIST DTSA-II and open it (see here).
2. Define an instrument with an EDS detector in DTSA-II (PDF).
   Simulations for different energy dispersion (e.g., 5/10/20 eV/channel), number of channels, detector window, etc. require the definition of multiple detectors. Call listDetectors() in the DTSA-II scripting window to get the detector names and IDs (d...). For the FEI/TFS TEM SuperX detector, have a look here.
3. Optional: Pre-define your sample material in DTSA-II (Video).
   Alternatively, this can also be done directly in the scripts (materialclass).

The steps to use the scripts:

1. Download/clone this repository or the desired script template (.py file).
2. Recommended: For each simulation, make a copy of the script in a separate folder. By default, the simulation results are stored in this folder.
3. Open the script in a text editor and adjust the user parameters.
4. Run the simulation by opening the script in DTSA.
   
5. Different output files are generated and saved.
   msa files contain the spectrum data and can be typically imported in any EDS software.

Available Scripts

TEM (homogeneous sample)

Free-standing transmission electron microscopy (TEM) sample/film of a given thickness and homogeneous composition. The Series scripts loop over different experimental parameters and create multiple spectra.

• TEM_Homogeneous
  Basic TEM sample simulation.
• TEM_Homogeneous_Series_Thickness
  Simulate different TEM sample thicknesses.
• TEM_Homogeneous_Series_Dose
  Simulate different total electron doses.

Scripts related to doping/trace concentrations (Please double-check the generated sample compositions/stoichiometries given in the DTSA console!):

• TEM_Homogeneous_Series_ElementDopingStoichiometry
  Simulate different doping levels of an element into the sample. Doping level given as stoichiometry ("number of dopant atoms added to the sample"; e.g. a doping value of "1.0" atoms of S in ZnO will lead to "ZnOS" or 33.3 at% S).
• TEM_Homogeneous_Series_ElementDopingWeightPerc
  Simulate different doping levels of an element into the sample. Doping level given as weight%.
• TEM_Homogeneous_Series_ElementDopingAtomicPerc
  Simulate different doping levels of an element into the sample. Doping level given as atomic%.
• TEM_Homogeneous_Series_CompositionDopingWeightPerc
  Simulate different doping levels of another material into the sample. Doping level given as weight%.
• TEM_Homogeneous_Series_CompositionDopingMolePerc
  Simulate different doping levels of another material into the sample. Doping level given as mol% (e.g., 1 mol% TiO2 in SrTiO3 will be (SrTiO3)0.99(TiO2)0.01).

Similar simulations as above can be performed using the Tools -> Simulation Alien... -> Monte Carlo model of a film on a bulk, homogeneous substrate in the GUI and specifying None as the substrate material.

Additional Resources

• Many PDFs with guides are provided on the NIST DTSA-II page
• Repository of John Minter (mostly SEM).
• ProbeSoftware forum