MAGEMin, an Efficient Gibbs Energy Minimizer: Application to Igneous Systems. (18th July 2022)

• Record Type: Journal Article
• Authors: Riel, N.; Kaus, B. J. P.; Green, E. C. R.; Berlie, N.
• Is Part Of: Geochemistry, geophysics, geosystems. Volume 23:Number 7(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-07-18
• Subjects: Geochemistry -- Periodicals; Geophysics -- Periodicals; Earth sciences -- Periodicals; 550.5
• DOI: 10.1029/2022GC010427
• Languages: English
• ISSNs: 1525-2027

Abstract

Abstract: Prediction of stable mineral equilibria in the Earth's lithosphere is critical to unravel the tectonomagmatic history of exposed geological sections. While the recent advances in geodynamic modeling allow us to explore the dynamics of magmatic transfer in solid mediums, there is to date no available thermodynamic package that can easily be linked and efficiently be accounted for the computation of phase equilibrium in magmatic systems. Moreover, none of the existing tools fully exploit single point calculation parallelization, which strongly hinders their applicability for direct geodynamic coupling or for thermodynamic database inversions. Here, we present a new Mineral Assemblage Gibbs Energy Minimizer (magem in). The package is written as a parallel C library, provides a direct Julia interface, and is callable from any petrological/geodynamic tool. For a given set of pressure, temperature, and bulk‐rock composition magem in uses a combination of linear programming, extended Partitioning Gibbs Energy and gradient‐based local minimization to compute the stable mineral assemblage. We apply our new minimization package to the igneous thermodynamic data set of Holland et al. (2018), https://doi.org/10.1093/petrology/egy048 and produce several phase diagrams at supra‐solidus conditions. The phase diagrams are then directly benchmarked against thermocalc and exhibit very good agreement. The high scalability of magem in on parallel computing facilities opens new horizons, for example, for modeling reactive magma flow, for thermodynamic data set inversion, and for petrological/geophysical applications. Plain Language Summary: Understanding magmatic systems requires knowing how rocks melt. Because a single melting experiment can easily take weeks, it is impossible to do enough experiments to cover the whole range of pressure, temperature, and composition relevant for magmatic systems. We therefore need a way to interpolate in between conditions that are not directly covered by the experiments. It is long known that the best way to perform such interpolation is by using basic thermodynamic principles. For magmatic systems, this requires a well‐calibrated thermodynamic melting model. It also requires an efficient computational tool to predict the most stable configuration of minerals and melt. Since the 1980s, a number of such computational tools have been developed to perform a so‐called Gibbs energy minimization. These tools work very well for simpler systems but become very slow for recently developed, more realistic, melting models. Here, we describe a new method that combines some ideas of the previous methods with a new algorithm. Our method is faster and takes advantage of modern computer architectures. It can predict rock properties such as densities, seismic velocities, melt content, and chemistry. It can therefore be used to link physical observations with hard rock data of magmatic systems. Key Points: A new, parallel, Gibbs energy minimization approach is presented to compute multiphase multicomponent equilibria It predicts parameters like stable phases, melt content, or seismic velocities as a function of chemistry and temperature/pressure conditions Examples and benchmark cases are presented that apply the approach to magmatic systems