About ExoPlex

A self-consistent mineral physics solver for exoplanets. We model rocky exoplanets with a core, lower and upper mantle, and water/ice layer. We use the Gibbs free energy minimization package, Perple_x (Connolly 09) along with a solution method from Stixrude 2011 to model mantles with natural phase transformations at depth (Pressure and Temperature dependence).

Current authors:

• Cayman Unterborn
• Alejandro Lorenzo

Science contributors:

• Steven Desch
• S.-H. Dan Shim: mineral physics
• Byeongknwan Ko: mineral physics

Getting Started

ExoPlex is written in Python 3.7 and is currently stable to its full capability in the Mac OS and Linux platforms. The mineral physics solver for mantle phase transformations relies on a Fortran program, perplex which is freely available from their website. We have included some precompiled phase solutions from perpleX so exoplex can be run without installing perple_x however, the user is limited in choice of planet composition to what is already included. Below we detail the full dependencies for exoplex.

Requirements

• Mac OS or Linux/Unix

• Perplex_x version 6.8.1

• Python 3.7+

• Python libraries (use latest versions):

  • numpy: handle arrays
  • os: search directories for files
  • math: math package
  • matplotlib: plotting results from ExoPlex
  • pexpect: used to call Perple_X from python
  • multiprocessing: not absolutely necessary but if you can get this working, it cuts code runtime in half
  • scipy: interpolation and integration
  • burnman: solves equations of state BME II,III, IV for the core and ice layers

Installing on Linux

1. Clone the master branch of this repo

2. Setup your workspace so that the scripts you write which call exoplex sit next to a directory called Solutions/ which itself has another directory called Grids/. For your workspace called workspace/ your directory tree should look like:

myworkspace/ │ └───Solutions/ │ │ │ └───Grids/ | | | |__examples/ | |____example_1.py | |____example_2.py | |____params.py

1. From the ExoPlex GitHub page, download the .tar files located in Solutions/. Put this in your local Solutions/ directory and unpack. This will provide many precompiled mineral physics solutions for ExoPlex to use.

2. The first time you run ExoPlex, it will download the necessary programs from the perple_X website. These are located at http://www.perplex.ethz.ch/ExoPlex/

Using ExoPlex

ExoPlex is meant to be run as a library complimentary to your python scripts. We are currently working on a version which is meant to be added to PyPI and used as an external library. Rrigh now, we reccomend cloning this repository and writing your scripts in a workspace near your local copy of exoplexparty library from PyPI. Despite this, ExoPlex is not entirely a standalone program. Users are required to setup their workspace so ExoPlex may read and store local files. The following sections describe these dependencies.

Workspace directory tree

Your workspace should look like the following:

myworkspace/ │ └───Solutions/ │ │ │ └───Grids/ | | | │ └───exoplex/ |___examples/ | |____example_1.py | |____example_2.py | |____params.py

Assuming you are working in an arbitrary directory called myworkspace/. You should have at least one directory at the root level where yoy can store your working files. In the example above, examples/ serves this purpose. There are several example scripts which are stored here to help you get started.

The Soulutions/ directory stores the outputs from perpleX which give mineralogy and thermodynamic parameters depending on composition. Each .dat file represents a different composition. We leave this at the top level so users can directly access these files to diagnose errors that may arise. In some cases, you will need to change the parameters for perpleX (located in params.py. Within solutions, you will need to initialize a Grids/ directory. Grid outputs will be stored here by default.

It is important to remember that (currently) ExoPlex is designed to work only in this environment. That is, the user should have a directory tree that exactly matches the one above. For example, naming your solution folder solutions/ instead of Solutions/ will cause ExoPlex to break.

params.py

This file will hold parameters which determine how ExoPlex makes models. Each parameter is described within this file. params.py also includes the parameters for running perple_x. Be cautious when altering these parameters.

The only parameter users may need to change is the temperature and pressure ranges for the lower and upper mantle. These control the domain perplex will use to find mineral assemblages. For creating very massive planets above 4 Earth masses (not recommended), for example, the user will need to have perplex rerun the solution file for the desired composition with a wider range of pressure and temperature. This is because the conditions reached in the mantles of more massive planets can exceed the default temperature and pressure ranges, causing ExoPlex to break.

calling functions

Once the above dependencies have been resolved and you are in the working directory with the proper directory tree, you may start using ExoPlex! In your script, import exoplex:

python import exoplex.run as run

This will set you up to run the functions detailed in the following section.

Functions

run.exoplex(inputs_file)

Calls ExoPlex functions and executes desired models based on input python script.

args

• inputsfile: String, name of a python script with input parameters. Example format is given in ```examples/input1.py```

returns

• numpy array of dictionaries which contain model information with keys: ['volume', 'phases', 'phasenames', 'temperature', 'density', 'dmass', 'K', 'mantleratios', 'gravity', 'Vp', 'pressure', 'Vs', 'radius', 'Vphi', 'bulk_ratios', 'alpha', 'cp', 'mass']

example

python planets = run.exoplex('inputs_1')

run.pltprem(**kwargs)

Radial density profile of each modeled planet plotted with the PREM (A. M. Dziewonski & D. L. Anderson 1981).

kwargs

• planet: results from run.exoplex()
  
• label: string array (or list), label for each model to be plotted. This will be displayed in a legend on the plot.

returns

• None

example

python planets = run.exoplex('inputs_1') run.pltrho(planet = planets, label = ['Mg/Si=1', 'Mg/Si=2'])

returns

run.pltrho(**kwargs)

Radial density profile of each modeled planet.

kwargs

• planet: results from run.exoplex()
  
• label: string array (or list), label for each model to be plotted. This will be displayed in a legend on theplot.

returns

• plot

example

python planets = run.exoplex('inputs_1') run.pltrho(planet = planets, label = ['Mg/Si=1', 'Mg/Si=2'])

returns

run.grid(**kwargs)

Creates a data file for a planet of some mass for a range of bulk compositions. Core mass fraction is calculated from the inputs.

kwargs

• mass: mass in units of Earth mass
  
• femg: python list of bulk (global) elemental Fe to Mg ratio range in the format [start, stop, step]
• simg: python list of bulk (global) elemental Si to Mg ratio range in the format [start, stop, step]
• h2o: (optional) range of water mass fraction in the format [start,stop,step]
• filename: (optional) name of file, otherwise default filename will be created

returns

• file of solutions for radius as a function of mass for the desired composition range

example

```python FeMg = [0.9, 1.5, 0.1] SiMg = [1.1, 1.5, 0.1] wt_h2o = [0, 0.2, 0.05]

run.singlegrid(femg = FeMg, simg = SiMg, h2o = wth2o, filename = 'composition_1.dat') ```

run.grid_cmf(**kwargs)

Creates a data file for a planet of some mass for a range of compositions. Similar to single_grid however the user specifies the mantle composition directly along with the core mass fraction.

kwargs

• mass : mass in units of Earth mass
  
• femg : python list of bulk (global) elemental Fe to Mg ratio range in the format [start, stop, step]
• simg : python list of bulk (global) elemental Si to Mg ratio range in the format [start, stop, step]
• cmf : python list of core mass fractions in the format [start, stop, step]
• h2o : (optional) range of water mass fraction in the format [start,stop,step]
• filename: (optional) name of file, otherwise default filename will be created

returns

• file of solutions for radius as a function of mass for the desired composition range

example

```python FeMg = [0.9, 1.5, 0.1] SiMg = [1.1, 1.5, 0.1] CMF = [0, 0.6, 0.05] wt_h2o = [0, 0.2, 0.05]

run.singlegrid(femg = FeMg, simg = SiMg, cmf = CMF, h2o = wth2o, filename = 'composition_1.dat') ```

run.mvr_grid(**kwargs)

Solves for the mass and radius relationship for a range of compositions and masses. Solutions are output as data files and a mass vs. radius plot. Each combination of composition is given a data file. The core mass fraction is calculated from the bulk elemental abundances of the planet.

kwargs

• mass : python list of mass range in the format [start, stop, step]
  
• femg : python list of bulk (global) elemental Fe to Mg ratio range in the format [start, stop, step]
• simg : python list of bulk (global) elemental Si to Mg ratio range in the format [start, stop, step]
• plot : True or False. Plots mass vs radius for composition range if True
• h2o : (optional) range of water mass fraction in the format [start,stop,step]
• filenames: (optional) list of filenames for each composition. Otherwise default filenames will be created

returns

• multiple files with solutions for a range of masses. Each composition creates a file.
• a plot of mass vs. radius

example

```python M = [0.3, 2.5, 0.1] FeMg = [0.9, 1.5, 0.1] SiMg = [1.1, 1.5, 0.1] wt_h2o = [0, 0.2, 0.05]

run.mvrgrid(femg = FeMg, simg = SiMg, h2o = wth2o, plot =False) ```

run.mvrgridcmf(**kwargs)

Solves for the mass and radius relationship for a range of compositions and masses. Solutions are output as data files and a mass vs. radius plot. Each combination of composition is given a data file. the cmf designation signifies that this function takes in mantle composition and core mass fraction as opposed to mvr_grid, which calculates core mass fraction from bulk elemental abundances.

kwargs

• mass : python list of mass range in the format [start, stop, step]
  
• femg : python list of mantle elemental Fe to Mg ratio range in the format [start, stop, step]
• simg : python list of mantle elemental Si to Mg ratio range in the format [start, stop, step]
• cmf : python list of core mass fractions in the format [start, stop, step]
• plot : True or False. Plots mass vs radius for composition range if True
• h2o : (optional) range of water mass fraction in the format [start,stop,step]
• filenames: (optional) list of filenames for each composition. Otherwise default filenames will be created

returns

• multiple files with solutions for a range of masses. Each composition creates a file.
• a plot of mass vs. radius

example

```python M = [0.3, 2.5, 0.1] FeMg = [0.2, 0.3, 0.1] SiMg = [0.3, 0.5, 0.1] CMF = [0.3, 0.5, 0.1] wt_h2o = [0, 0.1, 0.1]

run.mvrgridcmf(femg = FeMg, simg = SiMg, cmf = CMF, h2o = wt_h2o, plot = True) ```

returns

run.write(**kwargs)

Radial density profile of each modeled planet.

kwargs

• planet: results from run.exoplex()
  
• filenames: string array (or list). Filename for each planet modeled.
  

returns

• Prints to file a grid of mass, radius, density, pressure, and temperature, at each shell. Mass and radius are accumulative values from the planet center outwards.

example

python planets = run.exoplex('inputs_1') run.write(planet = planets, filenames = ['planet_1.dat', 'planet_2.dat'])

run.writeall(**kwargs)

Radial density profile of each modeled planet.

kwargs

• planet: results from run.exoplex()
  
• filenames: string array (or list). Filename for each planet modeled.
  

returns

• Prints to file a grid of mass, radius, density, pressure, temperature, heat capacity, thermal emissivity, and mineral profile of the compound at each shell. Mass and radius are accumulative values from the planet center outwards.

example

python planets = run.exoplex('inputs_1') run.writeall(planet = planets, filenames = ['planet_1.dat', 'planet_2.dat'])

Examples

Learning how to use ExoPlex is best done by working through the examples provided in the examples/ directory. A new user should be able to gain enough insight from the current examples to use ExoPlex to its fullest. If, after going through this README and you are still having some trouble, please email Alejandro at amloren1@asu.edu.

Contributing

ExoPlex is open source and we encourage users to contribute. In this case, fork the project and make a pull request.

License

Copyright (C) 2017 - by the ExoPlex team, released under the GNU GPL v2 or later.

Citation: see CITATION file

Acknowledgments

• James Connolly (PerPleX) for providing tips and a spot on the perpleX website. Please visit http://www.perplex.ethz.ch/ for more information.
• This project was done as part of the NASA NExSS grant

References

Connolly JAD (2009) The geodynamic equation of state: what and how. Geochemistry, Geophysics, Geosystems 10:Q10014 DOI:10.1029/2009GC002540.