The research, published in Frontiers in Geochemistry, centers on a remarkable specimen designated COMSK3. This octahedral diamond, measuring approximately 1.3 by 1.6 millimeters with a thickness of about 0.6 millimeters, contains at least ten visible mineral inclusions ranging from 0.02 to 0.2 millimeters in size. These inclusions—colorless and pale-green crystals trapped as the diamond grew—are the key to unlocking the gem’s origin story. Because diamonds are chemically inert and mechanically robust, they act as pristine time capsules, preserving samples of the mantle environment in which they formed, sealed away from billions of years of subsequent geological change.

To interrogate this tiny prisoner of the deep, the team deployed an impressive arsenal of analytical techniques. One inclusion, labeled Inc1, was polished to the diamond’s surface and analyzed with an electron probe microanalyzer for major element chemistry and with laser ablation inductively coupled plasma mass spectrometry for trace elements. The remaining entrapped inclusions were probed non-destructively using Raman spectroscopy, synchrotron Mössbauer spectroscopy at the European Synchrotron Radiation Facility in Grenoble, France, and single-crystal X-ray diffraction at Argonne National Laboratory in Chicago. X-ray computed microtomography allowed the researchers to map the exact positions of the buried inclusions before targeting them with the X-ray beam.

The mineralogical results were striking. The inclusions turned out to be a suite of sodium- and chromium-bearing diopside, high-magnesium olivine, and enstatite—a classic peridotitic assemblage, the kind of mineral association found in lithospheric mantle rocks. The polished clinopyroxene inclusion showed a remarkably homogeneous composition, with about 54.82 weight percent silicon dioxide, 6.33 weight percent chromium oxide, and 3.73 weight percent sodium oxide. Based on its crystal chemistry, the researchers identified it as a diopside containing roughly 20 percent of the kosmochlor component, with a chromium number of 0.63 and a magnesium number of 0.93. This combination of high sodium and high chromium is unusual and points to something more exotic than simple peridotite chemistry.

Thermobarometric calculations pinned down the conditions of diamond formation. Using elastic geobarometry—a technique that exploits the difference in compressibility between the diamond host and its trapped inclusions to reconstruct the pressure at which the inclusion was sealed—the team derived entrapment pressures between 4 and 5 gigapascals from the olivine inclusions. Clinopyroxene thermometry on Inc1 yielded a temperature of about 1,060 degrees Celsius, while projecting the mid-range pressure onto the geotherm of the Amazonian craton gave an independent temperature estimate of roughly 1,155 degrees Celsius. Together, these values place diamond formation at temperatures of 1,050 to 1,150 degrees Celsius and pressures corresponding to depths within the lithospheric mantle, comfortably below the graphite-diamond transition boundary.

The redox story proved equally compelling. Using synchrotron Mössbauer spectroscopy, which measures the oxidation state of iron with a spatial resolution of about 15 micrometers, the team determined the ferric iron fractions of three inclusions. The two clinopyroxenes showed Fe3+/ΣFe ratios of 0.16 and 0.14, while the olivine showed a much lower value of 0.02. The clinopyroxene values are notably higher than would be expected for pyroxenes that never encountered an oxidizing agent, and they fall within the range measured for clinopyroxenes from wehrlitic rocks—mantle peridotites that have been chemically refertilized by infiltrating melts. At the pressures and temperatures determined for the diamond, the coexistence of carbonate and diamond buffers oxygen fugacity at values between about 1.3 and 1.6 log units below the fayalite-magnetite-quartz reference, consistent with a metasomatic melt ranging from sodium-bearing dolomitic to carbonate-silicate in composition.

Trace element chemistry added further evidence of metasomatism. The rare earth element pattern of Inc1 shows the steep light rare earth enrichment typical of lherzolitic clinopyroxene inclusions that coexisted with garnet. But the vanadium content—543 parts per million—is far higher than that of typical peridotitic inclusions, which average around 542 parts per million in the reference dataset but are generally much lower. Such enrichment would require crystallization from or equilibration with a parental melt containing 100 to 300 parts per million vanadium, a level consistent with natural kimberlitic rocks. The absence of chemical zonation between the core and rim of the inclusion suggests that the clinopyroxene formed by direct transformation of pre-existing orthopyroxene rather than by contamination of an original lherzolitic pyroxene—exactly the kind of reaction expected when carbonatitic melts percolate through mantle peridotite.

This process, known as wehrlitization, occurs when a sodium-rich, carbon dioxide-bearing melt reacts with garnet harzburgite in the mantle, consuming orthopyroxene and producing clinopyroxene-rich rocks. Experimental studies have shown that the resulting clinopyroxene chemistry differs depending on whether the interacting melt is carbonatitic or silicic, and the simultaneous increase in calcium-to-aluminum ratio with magnesium number, coupled with decreasing aluminum oxide and titanium oxide, matches the signature of carbonatite interaction. The high kosmochlor component in the inclusion has long been recognized as a marker of sodium-rich carbonatite metasomatism of chromium-rich peridotites, and dark halos rimming some inclusions—interpreted in earlier work as residues of carbonate-bearing fluids—provide additional circumstantial evidence that the diamond grew from or alongside carbon-rich, oxidized melt.

The implications extend beyond a single gemstone. Diamonds from Rio Sorriso had previously been classified as sublithospheric, meaning they originated at depths greater than the base of the continental lithosphere, potentially in the lower mantle. Recent dating of sublithospheric Juína diamonds suggests they formed 610 to 450 million years ago from a peri-Gondwanan subduction system before being carried upward into the lithospheric keel of the Gondwana supercontinent. The new finding demonstrates that the same mantle column beneath the Amazonian craton also hosted lithospheric diamond formation, and that the lithospheric mantle there—accreted relatively recently in geological terms—was not chemically inert but actively infiltrated by carbonated melts capable of precipitating diamond.

For the broader Earth science community, the study underscores how much information a single diamond can yield when subjected to a comprehensive multi-technique investigation. The combination of elastic geobarometry, synchrotron Mössbauer spectroscopy, Raman and infrared spectroscopy, and single-crystal X-ray diffraction on one specimen provides a level of detail that no single method could achieve alone. While the authors caution that a single diamond cannot answer every question about lithospheric diamond formation at Juína, the unique characteristics of COMSK3—its peridotitic lherzolitic inclusions, its evidence of redox interaction with sodium-rich carbonatitic melt, and its ferric iron contents exceeding those of unmodified mantle pyroxenes—offer a vivid snapshot of metasomatic fluid-rock interaction in a young lithospheric mantle. The diamond, it seems, is not just a jewel but a messenger from a mantle that was far more chemically dynamic than its serene crystalline beauty suggests.

Subject of Research: Lithospheric diamond formation and mantle metasomatism beneath the Amazonian craton

Article Title: Lithospheric origin of a diamond from the Rio Sorriso area, Mato Grosso State, Brazil

Article References: Angellotti, A., Marras, G., Morana, M., Chariton, S., Stopponi, V., Medeghini, L., Romano, C., Correale, A., Bindi, L., Kaminsky, F. V., & Stagno, V. (2025). Lithospheric origin of a diamond from the Rio Sorriso area, Mato Grosso State, Brazil. Frontiers in Geochemistry, 3, Article 1607472. https://doi.org/10.3389/fgeoc.2025.1607472

Image Credits: AI Generated

DOI: 10.3389/fgeoc.2025.1607472

Keywords: diamonds, Rio Sorriso, Juína, Amazonian craton, mantle inclusions, thermobarometry, metasomatism, carbonated melt, Mössbauer spectroscopy, elastic geobarometry, lithospheric mantle, redox geochemistry