Kimberlite and carbonatite magmatism occurred on continental shields worldwide (Donnelly et al., 2012). Carbonatites are enriched in rare elements, such as niobium and rare earth elements, and may host deposits of these elements. Kimberlite, however, is a unique ultramafic rock derived from the deep mantle and a host of diamonds (Mitchell, 1986; Mitchell, 1995; Woolley et al., 1996). Carbonatites and kimberlites are all volumetrically minor components of continental magmatism, but they are fundamentally significant for our understanding of the carbon cycle, mantle evolution, and deep melting events. Studies have demonstrated that both kimberlite and carbonatite are among the deepest probes in the geochemical cycle (Bizimis et al., 2003; Nowell et al., 2004; Tappe et al., 2007; Ernst and Bell, 2010; Tappe et al., 2017), and it is imminent that recycling volatile-rich components from the Earth’s surface into the mantle is required to create the source regions for these two magma types (Bell and Simonetti, 2010; Tappe et al., 2013; Tappe et al., 2017). Kimberlite and carbonatite are commonly observed at the same place on ancient cratons, which raises the questions regarding their temporal distribution and origins. Regarding the view that kimberlite, carbonatite, and lamproite are genetically related, there are strong opinions both in support (Gaspar and Wyllie, 1984; Haggerty, 1989; Haggerty and Fung, 2006; Tappe et al., 2008; Smith et al., 2013; Tappe et al., 2013) and against it (Mitchell, 1979). One school of thought argues that primary carbonatite melts transform into hybrid carbonated silicate magmas akin to kimberlites by assimilation of cratonic mantle material (Tappe et al., 2020), whereas others suggest that kimberlites are not associated with major carbonatite complexes (Smith et al., 2013). The primary task in understanding the relationship between carbonatite and kimberlite is to identify their emplacement times. Additionally, the ages of carbonatite and kimberlite also play a significant role in discovering the linkage between mantle-sourced volatile-rich ultramafic magmas and tectonic processes (Tappe et al., 2020).

Oman is one of the key regions where the carbonatite and kimberlite association occurs and offers a good opportunity to study the kimberlite–carbonatite association. The key to solving these issues is high-precision emplacement ages. However, unlike granulite, carbonatite and kimberlite are all SiO₂-undersaturated rocks containing very rare zircon. Although some studies have reported the U–Pb ages of zircon in carbonatite from Oman, the origin of these zircon grains is still unknown. Although U–Pb studies of baddeleyite, perovskite, and apatite from carbonatites and kimberlites have been reported in other places (e.g., Sun et al., 2014; Sun et al., 2022 and references therein), such accessory minerals have not been found and studied in Oman kimberlite and carbonatite so far. In this study, we use apatite from the Oman carbonatite, spessartite [one type of calc-alkaline lamprophyre based on the classification of Rock (1986); Tappe et al. (2005)], and kimberlite to date their emplacement times, which would be important in understanding the relationship between kimberlite and carbonatite in Oman in further studies, and the associated tectonic processes would also be discussed.

Samples in this study were collected from Batain Nappes in the eastern Oman Mountains (Figure 1). The Batain Basin was situated at the junction of Arabia, India, Eurasia/Iran, and northeastern Oman until the opening of the Neotethys in the Early Permian (Figure 1A). The geology of the Batain Basin (part of the early Indian Ocean) in eastern Oman is dominated by the Batain Nappes, which comprise an allochthonous sequence of Permian to uppermost Cretaceous marine sedimentary and volcanic rocks (Nasir et al., 2011). The creation of the Batain Basin was led by the separation of Gondwana in the latest Carboniferous/Early Permian (Schreuer and Immenhauser, 1999; Hauser et al., 2001; Nasir et al., 2011), and it is dominated by the Batain Nappes (Nasir et al., 2011), which show two thrust sheets composed of the latest Jurassic (∼150 Ma) ophiolite and a mid-Cretaceous alkali basaltic sequence (115 Ma–125 Ma) (Meyer et al., 1996; Peters and Mercolli, 1998). The Batain Nappes are built by the lower Permian Qarari Formation, the upper Permian Mathbat Formation, the Middle Jurassic to Lowermost Cretaceous Ruwayda Formation, the Oxfordian to Santonian Wahra Formation, and the Santonian to uppermost Maastrichtian Fayah Formation (Gnos et al., 1997). Alkaline volcanic rocks abound within the Wahra and Ruwayda formations, which reflect an active alkaline sea-floor magmatism that produced seamount structures and many dispersed extrusive and intrusive rocks in the Wahra Formation. The UML and carbonatite are found only within the Wahra Formation, with outcrops spanning across the entire Batain Plain.

Samples in this study, including carbonatite, kimberlite, and spessartite, are from Batain Nappes, eastern Oman (Figure 1B). Carbonatite is associated with alkaline rocks and basalt, and their wall rock is composed of Triassic limestone. Spessartite is the dyke cutting through aillikite. Carbonatite (19OMS-83, Figure 2A) is composed of approximately 70% calcite and dolomite. Accessory minerals include ilmenite, phlogopite, barite, and apatite. Spessartite (19OMS-94, Figure 2B) has a porphyroclastic texture with amphibole and minor phlogopite in the phenocryst. The groundmass in this sample has calcite flow texture, which has carbonated plagioclase and minor amphibole, phlogopite, ilmenite, and barite. Based on the classification of Rock (1986) and Tappe et al. (2005), spessartite belongs to the calc-alkaline lamprophyre. Kimberlite (19OMS-97, Figure 2C) contains megacrysts of serpentinized olivine, garnet, diopside, carbonate minerals (Fe-rich dolomite and calcite), chromite, ilmenite, and apatite.

Separated mineral grains of apatite were handpicked, mounted in epoxy resin, and polished until the centers of the grains were exposed. Before isotopic analysis, back-scattered electron (BSE) images were obtained using a JEOL JXA8100 electron microprobe in order to assess internal compositional variation and textures and identify potential target sites for U–Pb analyses.

U–Pb isotopic analyses of apatite from three samples (19OMS-83, 19OMS-94, and 19OMS-97) were performed on an Agilent 7900 ICP-MS instrument (Agilent Technology, Tokyo, Japan) combined with a 193-nm ArF excimer laser (Geolas HD, MicroLas Göttingen, Germany) at the State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Wuhan. The laser spot size is 44 μm with a fluence of 8 J/cm² in this study. Then, the ablated material was transported to ICP using high-purity argon (makeup gas) and helium (carrier gas) (Luo et al., 2018a). To enhance accuracy and precision, a trace amount of water vapor was added before the ablation cell (Luo et al., 2018b; Luo et al., 2020), and a data processing device, serving as a fully interactive mercury signal reductant (Hu et al., 2014) and facilitating signal smoothing and mercury removal, was used in this laser ablation system to obtain ICP data. The counting times were 20 s for background signal acquisition and 50 s after ablation for each single-spot analysis. MAD apatite was used as the external standard to correct the Pb/U fractionation and instrumental mass discrimination. All initial common Pb were corrected using VizualAge in Iolite software. Uncertainties on individual analyses and pooled ages are reported at the 2s level. During the analytical sessions, the Otter Lake apatite has been repeatedly analyzed as unknown, and it yielded a lower intercept age of 916 ± 19 Ma (MSWD=23, n=30), which is consistent with its reference age (913 ± 7, Barfod et al., 2005).

The U–Pb isotope data on apatite grains from Oman kimberlite, carbonatite, and spessartite are listed in Table 1 and shown in Figure 3. As shown by LA-ICPMS data, the apatite grains have high U (3.1–47.8 ppm) and Th (16.3–366 ppm) contents that are favorable for U–Pb age determinations. Apatite grains from all three samples show a large Pb isotopic variation and less Pb loss. On the Tera–Wasserburg diagram, data points define a discordia line with a lower intercept age of 141.6 ± 6.0 Ma (MSWD = 0.71, n = 20) for carbonatite, 137.4 ± 5.2 Ma (MSWD = 1.5, n = 20) for spessartite, and 141.2 ± 6.2 Ma (MSWD=1.15, n = 20) for kimberlite. Therefore, these data indicate that three samples of apatite from the Oman carbonatite, spessartite, and kimberlite yield almost the same U-Pb ages of ∼140 Ma.

Previous studies used zircon and phlogopite grains to date carbonatite (Peters et al., 2001), ultramafic lamprophyric sills, and glimmerite xenoliths within the aillikite (Nasir et al., 2008) within the Batain Nappes, northeastern Oman. Ar-Ar ages of phlogopite in ultramafic lamprophyric sills and glimmerite xenoliths within the aillikite are 154 Ma–162 Ma (Nasir et al., 2008) and 150 ± 2 Ma, respectively (Peters et al., 2001). Zircon from carbonatites produced a weighted average ²⁰⁶Pb/²³⁸U age of 137 ± 1 Ma (Nasir et al., 2011). However, the analyzed mineral, phlogopite, is not a suitable mineral for Ar-Ar geochronology due to the existence of excess Ar. As for the zircons, studies suggested that some zircons in mantle-sourced rocks, such as kimberlite, may be of crustal origin extracted by host magma, which would yield a crustal age rather than an emplaced age (Sun, 2022). Moreover, the age of the kimberlite from Oman has not been studied so far.

Apatite crystallizes directly from the magmas and contains enough U for precise Pb isotopic analyses, which provides an excellent opportunity to remove the uncertainties in the determination of the age of the Oman kimberlite and carbonatite, thus making the U–Pb age determination reliable (Wu et al., 2013). The apatite U–Pb ages indicate that carbonatite and kimberlite were contemporaneous with a weighted average of 141 Ma, which is relatively earlier than the calc-alkaline lamprophyre (spessartite; 137 Ma). We consider the ages in this study to be the best estimate of the emplacement time of kimberlite, carbonatite, and calc-alkaline lamprophyre in Batain Nappes from Oman. Hence, the kimberlite and carbonatite from Oman emplaced at 141 Ma, followed by calc-alkaline carbonatite (spessartite) emplaced at 137 Ma, suggesting a complex magmatic plumbing system existed beneath the Batain Basin of the Oman field. The results of this study are significant in understanding the relationship between kimberlite and carbonatite associations in Oman, which can extend to a global scale in further studies.

The global kimberlite record suggests that approximately 80% of known occurrences are linked to breakup stages of supercontinents, and the others are collision-induced (Jelsma et al., 2009; Tappe et al., 2018; Zhang et al., 2019). The Indian Ocean was a stepwise breakup of east and west Gondwana at 157 Ma, and a breakup of east Gondwana at 130 Ma (Gnos and Perrin, 1996), which is registered only in the Batain Basin of Oman (Hauser et al., 2001). Paleomagnetic data from a previous study also showed that the eastern Oman oceanic lithosphere was formed at latitudes 38° ± 12°S around 150 Ma during the active breakup of Gondwana (Gnos and Perrin, 1996). So it appears that the emplacement of the Early Cretaceous kimberlite and carbonatite magmatism (∼140 Ma) in Oman is related to the breakup of Gondwana, and kimberlite and carbonatite occurred during the opening of the Indian Ocean (Peters and Mercolli, 1998). The most critical petrological variables enabling the formation of kimberlite and CO₂-rich ultramafic magmatism are the availability of oxidized CHO volatile species such as CO₂ and H₂O (Yaxley et al., 2017), as well as the lower temperatures of Earth’s upper mantle (Green and Falloon, 1998; Tappe et al., 2018). Thus, the breaking up of Gondwana, which releases pressure, provided a relatively cool and volatile-fluxed circumstance for the formation of Oman kimberlite and carbonatite melts.

Mesozoic–Cenozoic kimberlite between 250 Ma and 50 Ma is the most remarkable kimberlite bloom globally, and more than 60% of the world’s known kimberlite clusters on every continent were emplaced during this bloom. It was observed that 140 Ma–130 Ma is one of the strongest global kimberlite abundance peaks of this bloom (Figure 8 in Tappe et al., 2018), which corresponds to the period of the Pangea supercontinent breakup (Jelsma et al., 2009). By that time, the Indian Ocean had opened widely and the Gondwana portion of Pangea was separated into West and East Gondwanaland. The Early Cretaceous (∼140 Ma) Oman kimberlite, carbonatite, and spessartite were just formed under the tectonic background of the Pangea breakup.

Apatite U–Pb age dating implies that the Oman carbonatite, kimberlite, and spessartite are broadly coeval in the Early Cretaceous (137 Ma–140 Ma). The occurrence of carbonatite, kimberlite, and spessartite magmatism in Oman was contemporaneous with the time of the Gondwana breakup during the opening of the Indian Ocean. It was observed that 140 Ma–130 Ma is one of the strongest global kimberlite abundance peaks of the 250 Ma–50 Ma kimberlite bloom, which corresponds to the period of the Pangea supercontinent breakup. The breaking up of the Gondwana portion of the Pangea supercontinent provided a cool, volatile-fluxed decompression-related circumstance for the formation of Oman kimberlite and carbonatite melts.