Edit on GitHub

Configuration Files

The detector, its surroundings and symmetries can be specified in configuration files.

SolidStateDetectors.jl supports YAML and JSON as formats for the configuration files.

Example Configuration Files

Several example configuration files can be found under

<package_directory>/examples/example_config_files/.

They are accessible through a dictionary, SSD_examples, defined in the package:

using SolidStateDetectors
keys(SSD_examples) # dictionary holding the full path to the corresponding configuration files
KeySet for a Dict{Symbol, String} with 18 entries. Keys:
  :CoaxialTorus
  :TwoSpheresCapacitor
  :InfiniteCoaxialCapacitorCartesianCoords
  :CGD_CylGrid
  :Coax
  :IsochroneTest
  :InvertedCoaxTrapping
  :CGD
  :InfiniteCoaxialCapacitor
  :Hexagon
  :Spherical
  :InvertedCoax
  :ConeSym
  :InvertedCoaxInCryostat
  :SigGen
  :Cone2D
  :BEGe
  :InfiniteParallelPlateCapacitor

They can be loaded via

path_to_config_file = SSD_examples[:InvertedCoax]
sim = Simulation(path_to_config_file)

General Structure

The configuration files need a minimum of information in order to define the detector, its surroundings and symmetries.

This is a minimum working example of a simple true coaxial detector with two contacts:

name: Simple True Coax # optional
units:
  length: mm
  angle: deg
grid:
  coordinates: cylindrical
  axes:
    r: 45
    z:
      from: -40
      to: 40
medium: vacuum
detectors:
- semiconductor:
    material: HPGe
    geometry:
      tube:
        r: 
          from: 0.5cm
          to: 4cm
        h: 6cm
  contacts:
  - material: HPGe
    id: 1
    potential: 6000
    geometry:
      tube:
        r:
          from: 5
          to: 5
        h: 60
  - material: HPGe
    id: 2
    potential: 0
    geometry:
      tube:
        r:
          from: 40
          to: 40
        h: 60

It will be used to guide through the different parts of the configuration file.

Units

Internally, SolidStateDetectors.jl performs its calculations in SI units. However, configuration files can be written in custom units.

The field units denotes the standard units with which values will be parsed. Standard units can be defined for length, angle, potential and temperature.

In the example above, 

units:
  length: mm
  angle: deg

will lead to all length values to be parsed in units of mm, while all angle values will be parsed in units of deg (degree).

The configuration files also allow for directly passing units to the values that will be parsed using uparse from the Unitful.jl package, e.g.

units: 
  length: mm
  # ....
tube:
  r: 
    from: 5
    to: 40
  h: 60

is equivalent to

tube:
  r: 
    from: 5mm
    to: 40mm
  h: 60mm

or

tube:
  r: 
    from: 0.5cm
    to: 4cm
  h: 6cm

In the last example, even if the length unit was set to mm, the values will be parsed in units of cm. Please note to not leave a white space between the value and the unit and to use the Unitful.jl notation.

Grid

The calculations are performed on a finite world. To define the world, SolidStateDetectors.jl requires the properties of the grid, which are the coordinate system type and the dimensions. These are defined in the grid section of the configuration file, e.g.

grid:
 coordinates: cartesian
 axes:
   x: 
     from: -40
     to: 40
   y:
     from: -40
     to: 40
   z:
     from: -40
     to: 40

The coordinates of the grid can be:

  • cartesian (with the axes x, y and z)
  • cylindrical (with the axes r, phi and z).

The axes field is used to define the dimensions of each axis and, optionally, the boundary handling. In the example above, the x, y and z axes range from -40 to 40 units.

Grid boundary handling

Symmetries of the world can be used to reduce the calculation only to a fraction of the world. These can be passed as boundaries to the different axes.

For linear axes (x, y, z), the boundaries can be chosen infinite, periodic, reflecting, or fixed.

For radial axes (r), the boundaries can be chosen r0 for the left boundary and infinite, reflecting or fixed on the right boundary. If no boundaries are given, the default is r0 for the left boundary and infinite for the right boundary.

For angular axes (phi), the boundaries can be chosen reflecting or periodic. If no boundaries are given, the default is periodic for both edges.

All $\varphi$-symmetric configurations can be calculated in 2D if phi ranges from 0 to 0 with periodic boundary handling, i.e.

grid:
 coordinates: cylindrical
 axes:
   r: #...
   phi:
     from: 0
     to: 0
     boundaries: periodic
   z: #...

All $\varphi$-periodic configurations can be calculated on the fraction of the full $2\pi$ interval, i.e. for a 120°-periodic system

grid:
 coordinates: cylindrical
 axes:
   r: #...
   phi:
     from: 0°
     to: 120°
     boundaries: periodic
   z: #...

Different boundary handling can be chosen for the left and right end of the interval, i.e. for a 60°-periodic system with mirror symmetry at 30°

grid:
 coordinates: cylindrical
 axes:
   r: #...
   phi:
     from: 0°
     to: 30°
     boundaries:
       left: periodic
       right: reflecting
   z: #...

Detector Constituents

The detectors for the simulation are defined in an array detectors, where each entry corresponds to one detector. Each detector consists of exactly one semiconductor, a minimum of two contacts and, optionally, passives and virtual_drift_volumes.

detectors:
  - name: "Detector 1"
    semiconductor: #...
    contacts: 
      - # Contact 1
      - # Contact 2 
    passives: 
      - # Passive 1 (optional)
    virtual_drift_volumes:
      - # Virtual Drift Volume 1 (optional)
  - name: "Detector 2"
    semiconductor: #...
    contacts: 
      - # Contact 1
      - # Contact 2

Semiconductor

An example definition of the semiconductor looks like this:

semiconductor:
  material: HPGe
  temperature: 78
  impurity_density: # ...
  charge_drift_model: # ...
  geometry: # ...

The different fields of the semiconductor are:

  • material: the material of the semiconductor. This is important to know the electric properties of the semiconductor for the electric potential calculation. Possible choices are HPGe (high-purity germanium) and Si (silicon).
  • temperature (optional): the temperature of the semiconductor. If no temperature is given, the default is 78K for germanium and 293K for all other materials.
  • impurity_density (optional): the distribution of impurities in the semiconductor material. This has a strong impact on the electric potential calculation. If no impurity_density is given, the default is an impurity-free material ($\rho(\vec{r}) = 0$). Find a selection of implemented impurity density models and how to define an own model under Impurity Densities.
  • charge_drift_model (optional): a model to describe the drift of charge carriers in the semiconductor material. If no charge_drift_model is given, the default is ElectricFieldChargeDriftModel. Find a detailed description on how to define an own model under Custom Charge Drift Model.
  • geometry: the geometry of the semiconductor object. Find a detailed description on how to define geometries under Constructive Solid Geometry (CSG).

Contacts

An example definition of contacts looks like this:

contacts:
  - name: "n+ contact"
    id: 1
    potential: 5000V
    material: HPGe # optional
    geometry: # ....
  - name: "p+ contact"
    id: 2
    potential: 0
    material: HPGe #optional
    geometry: # ....

where each entry of the contacts array defines one contact.

The different fields of a contact are:

  • name (optional): the custom name for the contacts, relevant for plotting.
  • id: a unique id of the contact that will unambiguously identify the contact, for example in the signal generation. All contacts should be given an integer id from 1 to N where N is the number of contacts.
  • potential: the electric potential applied to the contact that is fixed throughout the whole contact geometry. This value can be parsed with units (5000V) or without (0 with the units defined in the units section).
  • material (optional): the material of the contact. This is important to know the electric properties of the contact for the electric potential calculation. If no material is given, the default is HPGe (high-purity germanium).
  • geometry: the geometry of the contact. Find a detailed description on how to define geometries under Constructive Solid Geometry (CSG).

Passives and Charged Surfaces

Passive objects and charged surfaces can be defined through entries of the passives array for each detector, e.g.

passives:
  - name: Passivated Surface
    material: HPGe
    charge_density: # ...
    geometry: # ...
  - name: Cryostat
    id: 3
    potential: 0
    temperature: 293K
    material: Al
    geometry: # ...

The different fields of a passive are:

  • name (optional): the name of the passive object. If no name is given, the default name is "external part".
  • id (optional): a unique id of the contact that will unambiguously identify the passive object. If no id is given, the default is -1.
  • potential (optional): the electric potential to which the passive object is fixed. This value can be parsed with units (5000V) or without (0 with the units defined in the units section). If no potential is given, the passive object will be treated as floating.
  • temperature (optional): the temperature of the passive object. This value can be parsed with units (293K) or without (78 with the units defined in the units section). If no temperature is given, the default is 293K.
  • material: the material of the passive object. This is important to know the electric properties of the passive object for the electric potential calculation.
  • charge_density (optional): model to describe charge density distributions within the passive object, e.g. charged surfaces. Find a detailed description on how to define charge densities under Charge Densities.
  • geometry: the geometry of the contact. Find a detailed description on how to define geometries in the section under Constructive Solid Geometry (CSG).

Surroundings

The medium of the world is passed as field medium, i.e.

name: # ... # optional
units: #...
grid: #...
medium: vacuum
detectors: # ...

If no medium is given, the default is vacuum. Implemented media are vacuum and LAr (liquid argon), but other media can be easily added to the material_properties dictionary in MaterialProperties.jl.

Passive objects, especially cryostats or holding structures can be defined in an array surroundings without being assigned to a specific detector. 

name: # ... # optional
units: #...
grid: #...
medium: vacuum
detectors:
- semiconductor: # ...
  contacts: 
    - # ...
    - # ...
  passives:
    - # ...
    - # ...
surroundings:
  - #...
  - #...

The definition of passive objects in the surroundings array is equal to that in the passives array of a detector.

Splitting Configuration Files

Configuration files for complex geometries can get quite long. SolidStateDetectors.jl allows for splitting configuration files into smaller ones and loading them using the include keyword. This feature supports YAML and JSON files.

When including a separate file, the user has to add its file path in the main configuration file at the place it is supposed to be added. To identify the file, set the key of this entry to include. Here, the user can also give an array of file paths. The file paths can be relative to the path of the configuration file or absolute. When including nested files and using relative paths, please always refer to the last parent file.

Including one file:

include : "file_to_be_included.yaml"

Including a list of files:

include: 
  - "first_file_to_be_included.yaml"
  - "second_file_to_be_included.yaml"

Add files to an array in the main configuration file

detectors:
  - include: "first_file_in_array.yaml"
  - include: "second_file_in_array.yaml"
  - include: "thrid_file_in_array.yaml"

A fully working example can be seen in SSD_examples[:InvertedCoaxInCryostat]. Here, the channels, the geometry and other parts are split into separate configuration files.