Welcome to the documentation for Reaktoro, an open source computational framework developed in C++ and Python to simulate chemically reactive processes.

Reaktoro has been designed from the ground up to be a flexible and extensible computational modeling framework for simulating chemical reactions. Reaktoro’s algorithms for chemical equilibrium and chemical kinetics calculations can be applied in a wide variety of modeling applications, from geochemical modeling of water-gas-rock systems to modeling the combustion of energetic materials.

Examples of specific and broad applications that Reaktoro can be useful for include, but are not limited to:

  • speciation calculations in aqueous electrolyte solutions, seawater, groundwater

  • dissolution of gas in aqueous solutions

  • mineral dissolution and precipitation in aqueous solutions

  • mixing of aqueous and/or gaseous solutions

  • evaporation processes

  • ion exchange processes

  • kinetically controlled reactions (e.g., mineral, aqueous, gaseous reactions)

  • adiabatic flame temperatures at constant pressure or volume

  • thermodynamic modeling of cement hydration and corrosion in concrete

  • ore formation processes

  • hydrometallurgical process

  • fluid-rock chemical reactions in geothermal energy systems

  • scaling in wells due to mineral precipitation

  • carbon storage in geological media via solubility and mineral trapping mechanisms

  • geological disposal of radioactive waste

Reaktoro can also be coupled with other software (e.g. reservoir simulators) to model even more complex processes such as:

  • reactive transport in porous media at pore or reservoir scale

  • reactive fluid flow for combustion modeling

For large-scale modeling applications where millions to billions of chemical equilibrium and/or chemical kinetics calculations are required, Reaktoro offers accelerated on-demand machine learning (ODML) solvers that can speed up chemical reaction calculations by one to three orders of magnitude as demonstrated in Leal et al. [2017], Leal et al. [2020] and Kyas et al. [2022].