thermo is open-source software for engineers, scientists, technicians and anyone trying to understand the universe in more detail. It facilitates the retrieval of constants of chemicals, the calculation of temperature and pressure dependent chemical properties (both thermodynamic and transport), the calculation of the same for chemical mixtures (including phase equilibria), and assorted information of a regulatory or legal nature about chemicals.
The thermo library depends on the SciPy library to povide numerical constants, interpolation, integration, differentiation, and numerical solving functionality. thermo all operating systems which support Python, is quick to install, and is free of charge. thermo is designed to be easy to use while still providing powerful functionality. If you need to know something about a chemical, give thermo a try.
Get the latest version of thermo from https://pypi.python.org/pypi/thermo/
If you have an installation of Python with pip, simple install it with:
$ pip install thermo
$ conda install -c conda-forge thermo
To get the git version, run:
$ git clone git://github.com/CalebBell/thermo.git
thermo's documentation is available on the web:
The library is designed around base SI units only for development convenience. All chemicals default to 298.15 K and 101325 Pa on creation, unless specified. All constant-properties are loaded on the creation of a Chemical instance.
>>> from thermo.chemical import Chemical >>> tol = Chemical('toluene') >>> tol.Tm, tol.Tb, tol.Tc (179.2, 383.75, 591.75) >>> tol.rho, tol.Cp, tol.k, tol.mu (862.2380125827527, 1706.0746129119084, 0.13034801424538045, 0.0005521951637285534)
For pure species, the phase is easily identified, allowing for properties to be obtained without needing to specify the phase. However, the properties are also available in the hypothetical gas phase (when under the boiling point) and in the hypothetical liquid phase (when above the boiling point) as these properties are needed to evaluate mixture properties. Specify the phase of a property to be retrieved by appending 'l' or 'g' or 's' to the property.
>>> tol.rhog, tol.Cpg, tol.kg, tol.mug (4.032009635018902, 1126.5533755283168, 0.010736843919054837, 6.973325939594919e-06)
Creating a chemical object involves identifying the appropriate chemical by name through a database, and retrieving all constant and temperature and pressure dependent coefficients from Pandas DataFrames - a ~1 ms process. To obtain properties at different conditions quickly, the method calculate has been implemented.
>>> tol.calculate(T=310, P=101325) >>> tol.rho, tol.Cp, tol.k, tol.mu (851.1582219886011, 1743.280497511088, 0.12705495902514785, 0.00048161578053599225) >>> tol.calculate(310, 2E6) >>> tol.rho, tol.Cp, tol.k, tol.mu (852.7643604407997, 1743.280497511088, 0.12773606382684732, 0.0004894942399156052)
Each property is implemented through an independent object-oriented method, based on the classes TDependentProperty and TPDependentProperty to allow for shared methods of plotting, integrating, differentiating, solving, interpolating, sanity checking, and error handling. For example, to solve for the temperature at which the vapor pressure of toluene is 2 bar. For each property, as many methods of calculating or estimating it are included as possible. All methods can be visualized independently:
>>> Chemical('toluene').VaporPressure.solve_prop(2E5) 409.5909115602903 >>> Chemical('toluene').SurfaceTension.plot_T_dependent_property()
Mixtures are supported and many mixing rules have been implemented. However, there is no error handling. Inputs as mole fractions (zs), mass fractions (ws), or volume fractions (Vfls or Vfgs) are supported. Some shortcuts are supported to predefined mixtures.
>>> from thermo.chemical import Mixture >>> vodka = Mixture(['water', 'ethanol'], Vfls=[.6, .4], T=300, P=1E5) >>> vodka.Prl,vodka.Prg (35.130757024029364, 0.9490586345579207) >>> air = Mixture('air', T=400, P=1e5) >>> air.Cp 1013.7956176577834
This library includes a huge database of (70000+) chemicals taken from the PubChem database (selected by the availability of CAS numbers, which all data included here is indexed by). Regretably, only ~20000 of those have even one chemical property apart from metadata (molecular weight, etc.). Some niche aspects (ions, ionic liquids) have been poorly served by the PubChem, and so extra databases manually curated for these are in development.
The Chemical and Mixture classes may be subject to considerably change in the interests of performance in the future. Because of this, they have been poorly documented and tested. However, each individual property method is mature and not expected to change. Documentation and testing are huge strengths, and it is intended to keep up the current quality of both.
A number of features have been worked on but are not yet included in this library, not ordered by any priority.
Phase equilibria according to activity coefficient methods (NRTL, UNIQUAC, Wilson, Van Laar, Margules): Functionality has been tentatively created, but is not included due to the lack of coefficient databases. Suggestions would be very welcome. UNIFAC has been tested, but is also not included due to the lack of automatic group contribution assignment.
Rigorous equations of state for excess properties, and phase equilibria: Tested EOSs are PR, LK, VdW, SRK, BWRS, and a few others. The holdup here is the determination of analytical expressions for their partial derivatives of mixtures. SymPy is immensely helpful, and has been used to successfully obtain specific values of those derivatives at specific points. Unfortunately, most listed forms as in Walas (1985) are incorrect. If expressions are not eventually found, this will be implemented with numerical derivatives only.
Fundamental Equations of State: The IAPWS-95 model, and that of 20 fluids in "Short Fundamental Equations of State for 20 Industrial Fluids" have been implemented. However, they are quite slow in Python - taking 2-10 ms to solve. This can be reduced to ~1-2 ms if Cython is used, however, this means that distribution through PyPi because harder. Suggestions about this are welcome. Currently, the phenomenal library CoolProp is used instead; which has already been packaged for PyPi. Even if custom code is released for these EOS, CoolProp will remain prioritized; developed in C++, it is simply much faster than code can be in pure Python.
Electrolyte models: The Pitzer, Bromley, and LIQUAC models have been in development along with parameter databases for them. The ion database currently has ~300 species, few of them with much data available. Phase equilibria with these models is also in progress.
Safety information, regulatory information, and economic data for chemicals: This functionality has been included, but is not yet very mature. This is a low priority.
Development follows pep8 and uses pytest for testing. Both Python 2 and 3 are supported.
The latest development version of thermo's sources can be obtained at
To report bugs, please use the thermo's Bug Tracker at:
LICENSE.txt for information on the terms & conditions for usage
of this software, and a DISCLAIMER OF ALL WARRANTIES.
Although not required by the thermo license, if it is convenient for you, please cite thermo if used in your work. Please also consider contributing any changes you make back, and benefit the community.
To cite thermo in publications use:
Caleb Bell (2016). thermo: Chemical properties component of Chemical Engineering Design Library (ChEDL) https://github.com/CalebBell/thermo.