Fluorine Vapor Turned an Ancient Finnish Marble Into Molten Rock, Revealing New Minerals

Deep beneath the quiet town of Lappeenranta in southeastern Finland lies one of the strangest rock stories ever reconstructed by geologists. The Ihalainen marble deposit, mined continuously since 1910 and worked in some form since the 1500s, has long been valued for its calcite and wollastonite, a calcium silicate prized by industry. But a new study of the deposit reveals that, roughly 1.6 billion years ago, the marble was invaded by a fluorine-rich granitic magma that did something remarkable: it melted the rock. Not just the intruding magma, but the marble itself, a carbonate rock that most geologists would assume should simply sit inert at such temperatures, was fluxed, dissolved, and stirred into a molten emulsion of silicate and carbonate liquids that froze in place like a snapshot of an ancient, violent chemical reaction.

The research, led by Robert F. Martin of McGill University together with Dirk Schumann, Markku J. Lehtinen, and Sebastian Fuchs, was published in the European Journal of Mineralogy. The team examined five specimens of wollastonite ore from the Ihalainen open pit, a quarry 1.2 kilometers long and up to 120 meters deep, using large-area image mosaics from transmitted-light microscopy and scanning electron microscopy, combined with electron-microprobe analyses at the Federal Institute for Geosciences and Natural Resources in Hanover, Germany. Two of the specimens, cut through narrow aplitic dikelets that intrude the bluish marble, proved to be the key witnesses of an extraordinary magmatic event that had gone essentially unnoticed in the deposit’s long mining history.

The geological backdrop is essential to understanding what happened. The Ihalainen Formation consists of Paleoproterozoic calcitic and dolomitic marbles, originally limestones deposited about 1.9 billion years ago. During regional metamorphism at temperatures of 650 to 700 degrees Celsius and pressures around 500 megapascals, dated by radiometric analysis of titanite at 1.858 billion years, water infiltrated siliceous horizons in the limestone and produced the main wollastonite-rich zones. Roughly 300 million years later, the area was engulfed by the Wiborg batholith, a vast body of anorogenic rapakivi granite. The marble block, now largely surrounded by granite, is interpreted as a roof pendant, a floating remnant of older crust captured within the batholith. Crucially, the Wiborg granites are known to be strongly enriched in fluorine, and locally in uranium, thorium, beryllium, tin, and niobium.

What the team found in specimen MKL-1021 was unlike anything documented before in this setting. Narrow dikelets of aplitic material cut the marble, but the dikelets contain neither quartz nor feldspar in the studied sections. Instead, they consist of a texturally complex assemblage of forsterite, now altered to a lizardite-like mixed-layer mineral provisionally labeled “aliettite”, diopside, and a striking graphic intergrowth of wollastonite and fluorite, in which rods of the two minerals grew simultaneously and interfered with each other as they crystallized from a melt. The researchers interpret the material inside a ribbon-like envelope of diopside, wollastonite, or fluorite as a near-eutectic fluorosilicate melt that quenched on injection into the calcite host. The graphic intergrowth is the smoking gun: such textures form only when two minerals crystallize together from a common liquid, in this case a melt in the system CaO–MgO–SiO2–H2O–F.

The source of the fluorine, and the heat, was almost certainly the underlying Wiborg granite. Experiments show that the solubility of fluorite, CaF2, in water rises sharply with increasing pressure and temperature, so a fluorine-rich vapor phase emanating from the cooling pluton could have dissolved fluorite deep in the system and deposited it in the heated ore above. Because fluid advection is the most efficient mode of heat transfer available, the same vapor delivered both the flux and the thermal energy needed to melt the ore. Once fluorite and heat were added, the wollastonite assemblage crossed its melting threshold. The team estimates that in the pseudobinary system CaSiO3–CaF2–H2O, the relevant eutectic temperature would fall to roughly 800 degrees Celsius, well within reach of the granite’s thermal footprint.

Specimen ML-3A, collected closer to the granite contact, records an even more dramatic episode: true anatexis of the marble itself. Here the dikelet consists of an emulsion of two immiscible melts, a silicate liquid and a carbonate liquid, that mingled turbulently, exchanged components, and then froze. The incoming granitic melt was desilicated, stripped of its silica by the aggressively fluxing carbonate melt, until it became a feldspathic, effectively syenitic liquid. The carbonate melt, in turn, dissolved silicates and became silicocarbonatitic. The researchers visualize a vigorous, turbulent reaction zone in which globules of disaggregated granitic melt churned through the carbonate liquid, an intimate mingling that likely resembled processes documented in the Early Paleozoic Ol’khon collision system near Lake Baikal in Russia. Skeletal crystals of new wollastonite along the dikelet margins, and even an entrained fragment of what appears to be a wollastonite cumulate, testify to the speed and violence of the mixing.

The feldspathic melt crystallized as a single calcium-poor alkali feldspar, a sanidine solid solution, and its subsequent cooling history is preserved in one of the finest displays of spinodal decomposition ever described in natural feldspar. As the crystal cooled through the coherent solvus, sodium and potassium atoms interdiffused within the aluminosilicate framework, spontaneously organizing into sinusoidal, interpenetrating domains of albite and K-feldspar without any nucleation barrier. The resulting lamellae, less than a micrometer wide and not crystallographically aligned, form stripe and labyrinthine patterns oriented perpendicular to the enclosed blobs of calcite. Because the two feldspar domains remained structurally coherent, both were strained, and the process is considered rapid relative to ordinary strain-free exsolution. A later influx of aqueous fluid recrystallized the feldspars and relieved the strain, yet left the tell-tale spinodal texture intact, a billion-year-old fingerprint of a phase separation that occurred under non-equilibrium conditions at perhaps 550 to 500 degrees Celsius.

Perhaps the most tantalizing discovery lies in three minerals so unusual that they cannot yet be matched to any known species. Labeled UK1, UK2, and UK3, they concentrate high-field-strength elements transported by the fluorine-rich fluids. UK1 is a calcium-tin sulfide with an extraordinary sulfur content, found in pitted zones of the calcite matrix and unlike the only known comparable mineral, genplesite. UK2 is a hydrous calcium silicate of niobium and tantalum, with minor zirconium and titanium, coating cavity rims in calcite. UK3 is a hydroxyl-bearing calcium-niobium silicate found in a cavity within euhedral wollastonite, with proportions that differ from the only IMA-approved minerals containing those elements, komarovite and mongolite. Malayaite, the tin analog of titanite, and tiny grains of scheelite and cassiterite round out the exotic inventory, all consistent with vapor transport of tin, tungsten, niobium, and tantalum from the evolved granite into the marble.

The fluid history of the deposit reads like a chemical drama in acts. It began carbon dioxide-rich, favoring the dry conditions under which spinodal decomposition operated, then shifted to an aqueous regime as the system cooled, unleashing a bewildering cascade of secondary calcic minerals: pectolite replacing wollastonite, paragonite replacing albite, lawsonite pseudomorphs after plagioclase, holtstamite rims around calcite, prehnite, gonnardite, xonotlite, and fluorapophyllite-(K), which systematically replaced wollastonite at submagmatic temperatures. The circulating fluid was dominantly calcic, a consequence of the high solubility of both calcite and fluorite in hot water, which explains the sheer diversity of the alteration assemblage documented across dozens of element-distribution maps.

Beyond its scientific novelty, the study carries practical implications. The Ihalainen deposit is one of the world’s notable wollastonite sources, and the demonstration that fluorine-rich vapor can flux-melt and locally enrich such ores opens a new lens on skarn deposits worldwide, particularly those fringing evolved granites like the Kymi topaz granite cupola of the Wiborg batholith, which is enriched in fluorine, lithium, beryllium, rubidium, tin, and niobium. The authors note that larger crystals of the three unknown minerals will be needed before formal proposals to the International Mineralogical Association can proceed, but given the documented enrichment of the Wiborg batholith in high-field-strength elements, they expect more such discoveries at Ihalainen. A quarry that has been worked for over five centuries, it turns out, still guards secrets that only modern microscopy could unlock.

Subject of Research: Fluorine-induced melting and magma-carbonate interaction in the Ihalainen wollastonite marble deposit, Finland

Article Title: Fluorine-induced melting in the Ihalainen calcite-wollastonite marble deposit, Lappeenranta, Finland

Article References: Martin, R. F., Schumann, D., Lehtinen, M. J., & Fuchs, S. (2026). Fluorine-induced melting in the Ihalainen calcite-wollastonite marble deposit, Lappeenranta, Finland. European Journal of Mineralogy, 38(4), 431-448. https://doi.org/10.5194/ejm-38-431-2026

Image Credits: AI Generated

DOI: 10.5194/ejm-38-431-2026

Keywords: wollastonite, fluorine, marble, rapakivi granite, Wiborg batholith, anatexis, spinodal decomposition, carbonate melt, new minerals, niobium, tantalum, Finland