Authors: Dejan Prelević, Michael W. Förster, Stephan Buhre, Fatma Gülmez, Tobias Grützner, Yu Wang, Stephen F. Foley

Abstract:
Mantle-derived magmas are traditionally assumed to originate by melting of an upper mantle consisting of uniform spinel- or garnet peridotite dominated by olivine. However, extensive studies of mantle-derived basalts suggest that the mantle is more mineralogically heterogeneous, so that the genesis of even the most common magmas requires consideration of mixed source regions within the mantle involving pyroxenites and hydrous minerals. We refer to these with the group term metasomes. However, most experimental studies on mantle melting have assumed a homogeneous source composition, presenting a challenge in quantifying the impact of these heterogeneities. This paper provides a comprehensive review of recent advances in reaction experiments that depart from traditional approaches assuming a homogeneous mantle. We begin by assembling evidence for the existence of metasomes, discussing their formation and integration into basaltic melts. Further, we introduce the reaction experiments combining peridotite with hydrous assemblages, such as phlogopite, amphiboles, and apatite, leading to more accurate simulations of natural magmatic processes. These experiments reveal that the melting of hydrous metasomes and subsequent melt-peridotite interactions are key to producing the high alkali contents observed in natural lavas. The melting of hydrous metasomes occurs at lower temperatures than peridotite, resulting in diverse melt compositions. The interaction between metasome-derived melts and peridotite further modifies these melts, influenced by the pressure-dependent melting behaviors of minerals like orthopyroxene and olivine. This dynamic process leads to the generation of ultrapotassic and Na-alkaline melts with varying silica and alkali contents, reflecting the complex interplay of melting and reaction mechanisms in the mantle. Formation of hydrous metasomes have also been studied by reaction experiments. Experimental studies have predominantly focused on potassium-rich systems due to the geochemical signatures of potassic igneous rocks suggesting sedimentary rock contributions to their sources. These studies simulate interactions between melts and mantle peridotite, particularly in sub-arc regions, leading to potassium-rich metasomes. More experimental studies are needed on sodium-rich alkaline systems to understand the formation of amphibole-rich metasomes and bridge knowledge gaps. Future studies should emphasize the detailed compositional variability of melts from metasomes, their reactions with peridotites, and comparisons with surface lavas. Understanding the kinetics of these reactions and the melting mechanisms of metasome-derived melts is essential. However, the considerable mineralogical diversity of hydrous metasomes poses a primary challenge facing experimental studies. It underscores the need for more experiments on additional melt source rocks and their reaction with peridotites, as the story about the reaction of melts from hydrous metasomes with mantle peridotites has only just begun.