Continental growth, buoyancy and stability are key processes in the evolution of the Earth and are governed by temperature. To constrain the thermal history of continental lithosphere over Gyr timescales, we present a novel application of the nitrogen aggregation geothermometer to zoned diamonds. By simultaneously forward modelling nitrogen aggregation in the core and rim of single diamonds, cooling rates and initial temperatures are calculated. Our method is accessible, locally specific and scalable for global coverage. We demonstrate its application on samples from Diavik, Canada, finding most likely cooling rates of 42-62 K/Gyr, depending on the selection of zone ages from literature. Further uncertainty is primarily controlled by the quality of age determinations. Respective core and rim isochrons of 3.52 +/- 0.17 and 1.86 +/- 0.19 Ga produce asymmetric 95% confidence limits of around 30 and 80 K/Gyr when their uncertainties are jointly propagated with Monte Carlo analysis. Initial temperatures remain tightly constrained around 1200 degrees C. Extensive investigations show the method to be resilient to other sources of systematic and random error, and to potential effects of transient thermal events. Confidence in the model is further supported by its agreement with previous inclusion thermometry and tectonic evolution models of the studied area. It also reconciles previously discrepant results yielded by the simple application of the nitrogen aggregation thermometer and inclusion trapping temperatures, emphasising the inaccuracy of the former. Applying our model to other localities will allow a global picture of the thermal evolution of the lithospheric mantle to be built.

The aluminous calcium-ferrite type phase (CF) and new aluminous phase (NAL) are thought to hold the excess alumina produced by the decomposition of garnet in MORB compositions in the lower mantle. The respective stabilities of CF and NAL in the nepheline-spinel binary (NaAlSiO _4 –MgAl _2 O _4 ) are well established. However with the addition of further components the phase relations at lower mantle conditions remain unclear. Here we investigate a range of compositions around the nepheline apex of the nepheline-kalsilite-spinel compositional join (NaAlSiO _4 –KAlSiO _4 –MgAl _2 O _4 ) at 28–78 GPa and 2000 K. Our experiments indicate that even small amounts of a kalsilite (KAlSiO _4 ) component dramatically impact phase relations. We find NAL to be stable up to at least 71 GPa in potassium-bearing compositions. This demonstrates the stabilizing effect of potassium on NAL, because NAL is not observed at pressures above 48 GPa on the nepheline-spinel binary. We also observe a broadening of the CF stability field to incorporate larger amounts of potassium with increasing pressure. For pressures below 50 GPa only minor amounts ( <0.011(1)K/K+Na+Mg ) of potassium are soluble in CF, whereas at 68 GPa, we find a solubility in CF of at least 0.088(3)K/K+Na+Mg . This indicates that CF and NAL are suitable hosts of the alkali content of MORB compositions at lower mantle conditions. For sedimentary compositions at lower mantle pressures, we expect K-Hollandite to be stable in addition to CF and NAL for pressures of 28–48 GPa, based on our simplified compositions.

Larnite (β-Ca2SiO4) has previously been reported as an inclusion in sub-lithospheric diamonds and is generally interpreted as a retrograde reaction product of calcium silicate perovskite. In this study, we review the controls on the stability of the Ca2SiO4 polymorphs and show that phosphorus is likely essential for the preservation of β-Ca2SiO4. We also report a detailed study of the solubility of water and its incorporation mechanisms in γ-Ca2SiO4 and phosphorus-doped β-Ca2SiO4 using FTIR spectroscopy on high-pressure experiments quenched from 4–9.5 GPa and 1000–1200 °C combined with ab initio calculations. The experimentally determined water solubilities are in the range of 107–178 ppm. Our FTIR spectra and ab initio calculations indicate that for phosphorus-free γ-Ca2SiO4 the incorporation mechanism involves protonated Si and Ca1 vacancies. For phosphorus-bearing β-Ca2SiO4, our preferred incorporation mechanism involves one Si4+ ion replaced by one P5+ ion with a single protonated Ca2 vacancy. The low water solubility observed here for larnite implies that if primary calcium silicate perovskite inclusions trap high water concentrations during diamond growth from a volatile-rich fluid, measurements of the concentration of water in larnite will not provide a useful record of the initial volatile concentration. Instead, water would be hosted in other retrograde reaction products, possibly including exsolved fluids.

Subduction related to the ancient supercontinent cycle is poorly constrained by mantle samples. Sublithospheric diamond crystallization records the release of melts from subducting oceanic lithosphere at 300–700 km depths 1 , 2 and is especially suited to tracking the timing and effects of deep mantle processes on supercontinents. Here we show that four isotope systems (Rb–Sr, Sm–Nd, U–Pb and Re–Os) applied to Fe-sulfide and CaSiO 3 inclusions within 13 sublithospheric diamonds from Juína (Brazil) and Kankan (Guinea) give broadly overlapping crystallization ages from around 450 to 650 million years ago. The intracratonic location of the diamond deposits on Gondwana and the ages, initial isotopic ratios, and trace element content of the inclusions indicate formation from a peri-Gondwanan subduction system. Preservation of these Neoproterozoic–Palaeozoic sublithospheric diamonds beneath Gondwana until its Cretaceous breakup, coupled with majorite geobarometry 3 , 4 , suggests that they accreted to and were retained in the lithospheric keel for more than 300 Myr during supercontinent migration. We propose that this process of lithosphere growth—with diamonds attached to the supercontinent keel by the diapiric uprise of depleted buoyant material and pieces of slab crust—could have enhanced supercontinent stability.

Diamond‐hosted majoritic garnet inclusions provide unique insights into the Earth's deep, and otherwise inaccessible, mantle. Compared with other types of mineral inclusions found in sub‐lithospheric diamonds, majoritic garnets can provide the most accurate estimates of diamond formation pressures because laboratory experiments have shown that garnet chemistry varies strongly as a function of pressure. However, evaluation using a compilation of experimental data demonstrates that none of the available empirical barometers are reliable for predicting the formation pressure of many experimental majoritic garnets and cannot be applied with confidence to diamond‐hosted garnet inclusions. On the basis of the full experimental data set, we develop a novel type of majorite barometer using machine learning algorithms. Cross validation demonstrates that Random Forest Regression allows accurate prediction of the formation pressure across the full range of experimental majoritic garnet compositions found in the literature. Applying this new barometer to the global database of diamond‐hosted inclusions reveals that their formation occurs in specific pressure modes. However, exsolved clinopyroxene components that are often observed within garnet inclusions are not included in this analysis. Reconstruction of inclusions, in the 8 cases where this is currently possible, reveals that ignoring small exsolved components can lead to underestimating inclusion pressures by up to 7 GPa (∼210 km). The predicted formation pressures of majoritic garnet inclusions are consistent with crystallization of carbon‐rich slab‐derived melts in Earth's deep upper mantle and transition zone.

QUIDDIT is a free Python software-package designed to process Fourier Transform Infrared (FTIR) spectra of diamonds automatically and efficiently. Core capabilities include baseline correction, determination of nitrogen concentration, nitrogen aggregation state and model temperature and fitting of both the 3107 cm-1 and platelet (B’) peaks. These capabilities have allowed the authors to study platelet defects and their relationship to nitrogen aggregation in previous studies. Data visualisation, vital to interpreting and evaluating results, is another key component of the software. QUIDDIT can be applied to single spectra as well as linescan and 2-dimensional map data. Recently, additional features such as manual platelet peak and nitrogen fitting, custom batch peak fitting and two-stage aggregation modelling were made available. QUIDDIT has been used successfully for natural diamonds containing aggregated forms of nitrogen in the past and has since been adapted for the study of diamonds containing C-centres as well.

Diamonds from the Zimbabwe craton have received the least attention of all the major diamond producing cratons. The potential for an extremely depleted, old continental lithospheric keel to the Zimbabwe craton is suggested by the low Os isotopic compositions of crustal chromites, an average Re-Os model age for peridotites of 3.2 Ga, the dunitic character of peridotite xenoliths and silicate inclusions in diamonds, and high seismic velocities. We have begun a thorough ReOs study of sulfides and chromite inclusions in Murowa diamonds to establish their composition, paragenesis, and age, and to place diamond formation in context of lithospheric mantle depletion. A significant question is the relationship of the Zimbabwe craton keel to that of the adjacent Kaapvaal lithospheric keel to the west and south.

Bulanova* GP1, Marks A.1, Smith CB1, Kohn SC1, Walter MJ1, Gaillou E2, Shirey SB2. Trautman R.3 and Griffin B.J.4 1. School of Earth Sciences, University of Bristol, United Kingdom 2. Department of Terrestrial Magnetism, Carnegie Institution of Washington, Washington D.C., USA 3. Perth, Western Australia, 4. University of Western Australia, Perth *corresponding author (galina_bulanova@hotmail.com)

The Juina-5 pipe is located within a discrete kimberlite cluster 30 km to the south of Collier-4 pipe and the intervening well-described Juina alluvial placer. Juina-5 diamonds are different in appearance and many other properties to those of the alluvials. Most notably they carry a distinctive sublithospheric Naand K-rich mineral assemblage never before encountered in a natural environment. Here we describe a set of inclusions found in such diamonds, characterize them, propose the manner of their origin and review the other superdeep diamond occurrences in the Juina Kimberlite field. New data on diamonds from Collier-4 kimberlite and Cinta Larga river are also presented.

OverviewSulphide inclusions hosted in sublithospheric ("superdeep") diamonds from the Juina-5 and Collier-4 kimberlites (Brazil) are being studied by a variety of techniques.Protected from any reequilibration by their host diamond, these samples offer a unique opportunity to investigate the deep sulphur cycle, potentially adding new constraints on recycling of surface material at depths.

Machado River alluvial diamonds are located in Rondonia, Brazil, downstream from the diamondiferous Pimenta Bueno and Carolina Kimberlites (Fig. 1).The nearby Araras kimberlites are barren.The occurrence lies within the 1.55-1.8Ga Rio Negra-Juruena Mobile Belt on the southern margin of the Amazonian Craton, some 200 km W of the Juina ultradeep diamond field.

The Juina Field is one of the many expressions of Cretaceous alkaline volcanicity developed along the AZ125 o lineament ascribed to the passage of the Brazilian continent over the Trindade Plume. The disputed geological evolution of the Rio Negro – Juruena Mobile Belt (lateral versus vertical accretion?) results in controversy over Juina’s kimberlite geotectonic setting (“on-craton” or “off craton”). Eclogites and garnet lherzolites are the most abundant mantle xenoliths in the kimberlites and have been dated at ca. 1.6 Ga (Costa et al., 2003), suggesting a link to Mesoproterozoic subduction of oceanic crust during formation of the Mobile Belt. Methods and samples 30 diamonds with mineral inclusions were selected for study under optical microscope from a Rio Tinto collection. The diamonds are 2 3.5 mm in size colourless or brown transition forms and more rare octahedrons and dodecahedrons; some macles and aggregates present. Most of the crystals are broken, heavily resorbed, plastically deformed and have internal cracks and deep etch channels (Fig. 2).

Platelets are one of the most common defects occurring in natural diamonds but their behaviour has not previously been well understood. Recent technical advances, and a much improved understanding of the correct interpretation of the main infrared (IR) feature associated with platelets (Speich et al. 2017), facilitated a systematic study of platelets in 40 natural diamonds. Three different types of platelet behaviour were identified here. Regular diamonds show linear correlations between both B-centre concentrations and platelet density and also between platelet size and platelet density. Irregular diamonds display reduced platelet density due to platelet breakdown, anomalously large or small platelets and a larger platelet size distribution. These features are indicative of high mantle storage temperatures. Finally, a previously unreported category of subregular diamonds is defined. These diamonds experienced low mantle residence temperatures and show smaller than expected platelets. Combining the systematic variation in platelet density with temperatures of mantle storage, determined by nitrogen aggregation, we can demonstrate that platelet degradation proceeds at a predictable rate. Thus, in platelet-bearing diamonds where N aggregation is complete, an estimate of annealing temperature can now be made for the first time.