In recent years, economically significant sandstone-hosted uranium mineralization has been identified in the Louzhuangzi area along the southern margin of the Junggar Basin. However, the controls on uranium enrichment and their links to depositional architecture and post-depositional fluid processes remain insufficiently constrained. This study integrates field geological investigations, drill-core lithofacies logging, and systematic sampling with petrographic and micro-analytical techniques, including optical microscopy, scanning electron microscopy (SEM), electron probe microanalysis (EPMA), and transmission electron microscopy (TEM). The objective is to elucidate the depositional characteristics, alteration processes, and uranium occurrence mechanisms of the ore-bearing sandstones within the Toutunhe Formation (J2t). Results show that the upper member (J2t2) represents meandering-river deposits, whereas economically significant uranium mineralization is hosted in the lower member (J2t1), characterized by braided-river sandstone units with high permeability. The sandstones of the Toutunhe Formation (J2t) exhibit intense oxidation by surficial fluids, overprinted by post-mineralization hydrothermal alteration and sulfide-forming alteration associated with reducing fluids. Uranium is closely associated with pyrite, organic matter, and clay minerals. Uranium minerals are dominated by coffinite and pitchblende (~68%), with UO₂ contents of 52.89-86.07%. Minor Ti-bearing uranium phases (~16%), interpreted as possible brannerite, contain 38.01-41.46% UO₂ and 31.67-36.09% TiO₂, while nanoscale uranium minerals (~16%) show UO₂ contents of 9.15-60.28%. These results indicate that uranium mineralization was controlled by the coupling of braided-channel architecture and multi-stage fluid processes. Uranium was initially precipitated from oxidized fluids and subsequently modified and preserved by later thermal and reducing fluids, highlighting the importance of multi-fluid interactions in sandstone-hosted uranium systems.

In practical dual-energy X-ray (DE-XRT) copper ore sorting industrial settings, process parameters such as belt speed, X-ray source intensity, and ore particle size are typically maintained relatively stable, while the thickness of ore particles naturally fluctuates significantly due to the randomness of crushing, making it a key factor affecting sorting stability. To address the sorting instability stemming from thickness variations, this paper proposes a robust classification method centered on thickness-consistency regularization. The method is based on the X-ray attenuation mechanism, constructing logarithmic-domain thickness perturbations to simulate the input distribution shift caused by thickness variations. Experiments are conducted using a two-channel ResNet18 as the base model on 7,245 DE-XRT images, comparing three methods—Baseline, Brightness, and Consistency—and employing the Swin Transformer backbone to validate the architectural generalizability of the proposed method. Results show that Consistency achieves 92.18% accuracy on the clean test set, which is comparable to the 92.16% of the strong Brightness baseline. Under a large positive thickness shift (δ = +0.25), Consistency yields higher Macro-F1 and Worst-group accuracy, improving over Brightness by 7.76 and 11.09 percentage points, respectively. Consistency also demonstrates superior robustness over Brightness under substantial negative thickness shifts. These findings demonstrate that the proposed method effectively enhances model robustness under extreme thickness variations and is applicable to multiple deep learning architectures.

Roll-front sandstone-type uranium deposits are important uranium resources worldwide, but their orebodies vary significantly in location, geometry and grade despite broadly similar redox-controlled mineralization processes. This variability is attributed to differences in redox conditions, uranium source availability and groundwater flow, yet how these geological factors control orebody variability under hydrodynamic regimes remains poorly constrained. This study is the first to use reactive transport modeling to reproduce the dynamic process of roll-front uranium mineralization and to evaluate the effects of different geological factors on uranium orebody formation. The modeling results demonstrate that the dynamic balance between the reducing capacity of host sandstones and the oxidizing capacity of fluids exerts primary control on the localization, geometry and lateral extent of roll-front orebodies. Both strongly reducing host sandstones and weakly oxidizing fluids can limit interlayer oxidation, favoring narrow, relatively high-grade roll-front orebodies near the recharge area. In contrast, weakly reducing host sandstones and more oxidizing fluids allow significant interlayer oxidation, producing larger but relatively lower-grade mineralized zones farther from the recharge area. The uranium concentration in the fluids and the pre-enriched uranium content of the host sandstones are positively correlated with ore grade and together exert a primary control on the formation of high-grade roll-front uranium orebodies. Groundwater flow rate also exerts a non-linear influence on uranium mineralization, with moderate flow rates most favorable for uranium dissolving in the oxidization zone and reprecipitation at the redox front. Extremely low flow rates limit oxidation-front advance, whereas high flow rates promote extensive oxidation. These findings help explain the geological controls on orebody localization, geometry, and grade variability in roll-front uranium systems and provide a quantitative framework to guide future exploration.

The South China Block (SCB) is recognized as one of the most significant uranium deposit clusters in the world, characterized by its complex genetic types and geodynamic drives. Based on host rocks, uranium deposits in the SCB can be categorized into three primary types, exhibiting a trend from black shalerelated deposits in the west, to granite-related, and ultimately to volcanic-related deposits toward the eastern margin of the SCB. We identify that three types of deposits are primarily distributed within or along margins of ancient crustal domains. Geochronological data reveals large-scale uranium mineralization occurred predominantly during Cretaceous and Paleogene periods. Uranium mineralization was mainly controlled by structures in the extensional setting, developed particularly at subsidiary faults, lithological (unconformity, intrusion contacts) and physicochemical interfaces. Uranium mineralization is dominantly characterized by medium to low ore-forming temperature with pitchblende as the main industrial mineral, and with silicification, carbonatization, hematitization, fluoritization and chloritization as common alteration. Isotopic studies show that sulfur sourced from host rocks, while carbon isotopes distinguish mantle-derived signatures in granite- and volcanic-related deposits from primarily sedimentary organic matter sources in black shale-related deposit. Uranium was mainly contributed by host rocks which are relatively U-fertile geological formations. Magmatic and/or mantle-derived mineralizing agents promote the activation and migration of uranium in host rocks, and accelerate the accumulation of U in ore-forming fluids. Our study suggests that the coupling of shallow and deep-seated energy and conduit system within a crustal extension setting, together with the pre-enrichment of uranium in basement and host rocks, controlled the formation of uranium deposits in the SCB. (c) 2025 China University of Geosciences (Beijing) and Peking University. Published by Elsevier B.V. on behalf of China University of Geosciences (Beijing). Thi s is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Mineral prospectivity mapping for sandstone-hosted uranium deposits commonly relies on isotropic distance decay, despite the fact that ore-forming fluids are typically guided by faults and paleochannel systems with strong directional organization. As a result, structurally controlled geological signals may be diffused into quasicircular anomalies, weakening the geological realism of prospectivity predictions. To overcome this limitation, we propose a Direction-Aware Geographically Weighted Neural Network (DA-GWNN), which embeds expertdefined structural orientation priors into neighborhood weighting to preferentially capture spatial influence along plausible ore-forming directions. In addition, a criteria-based consensus soft-labeling scheme is developed at the grid-cell scale to represent different levels of mineralization evidence under sparse supervision. The method was tested in the Kailu Basin against six benchmark models under nested spatial blocked crossvalidation. DA-GWNN achieved the best predictive performance, with a mean AUC of 0.912 +/- 0.019 and a mean AUPRC of 0.794 +/- 0.026. SHAP analysis indicates that the oxidation-reduction transition zone is the principal control on prospectivity, whereas the role of fault proximity is conditional and varies with the local redox buffering capacity reflected by the gray-sand ratio. Area-constrained analysis shows that the 36.6% P-A envelope captures 78.2% of all positive-labeled cells and represents a basin-scale mineralization-system footprint. Under the stricter economic-positive definition (y = 1.0), DA-GWNN captures 45.0%, 67.5%, and 80.0% of economic positive cells within the Top 5%, Top 10%, and Top 15% cumulative area budgets. These results indicate that DA-GWNN supports a hierarchical exploration workflow from regional system recognition to focused target verification.

The Kuqa fold-thrust belt is a tectonically active intracontinental deformation zone in northwest China. The 2021 MS 5.4 Baicheng earthquake in this region caused significant casualties and property damage, drawing considerable attention to its seismogenic mechanism. However, the mechanisms governing the spatial distribution of surface ruptures and the role of salt structures in deep-shallow fault activity coupling are not yet fully understood. In this study, we integrated InSAR coseismic deformation field with high-resolution seismic profiles, and discrete element numerical simulations to investigate surface rupture characteristics and fault activity mechanisms. Here we show that D-InSAR results reveal a distinct NNW-SSE-trending surface rupture zone with composite thrust-strike-slip motion, consistent with field observations. The deep Kelasu Fault is identified as the primary seismogenic structure, transmitting tectonic stress upward through plastic flow of the salt layer and triggering slip on the shallow Laohutai Fault at the top of the salt dome. Numerical simulations further confirm that stress concentration at the dome apex and the synergistic interaction between deep seismogenic faults and shallow salt-bearing structures jointly controlled the formation and spatial distribution of the surface rupture zone. These findings provide multi-source evidence for the seismogenic processes in salt-influenced fold-thrust belts and offer insights for seismic hazard assessment in similar tectonic settings.

While compressional tectonics has long been recognized as a significant factor in uranium mineralization, the role of extensional tectonics remains comparatively understudied. This investigation analyzes representative sandstone-hosted uranium provinces, including the Colorado Plateau (North America), the Agadez region (West Africa), the North China–Transbaikal region (Northeast Asia), and the Massif Central (Western Europe). Our systematic analysis reveals that extensional tectonics, building upon pre-existing rifting structures, amplify tectono-magmatic activity, which in turn exerts spatiotemporally control on uranium deposition and gives rise to diverse mineralization ages. Key structural elements within tectonic transition zones, such as uplifts, fault systems, and volcanism, play critical roles: uplifts and shallow-level faults govern the recharge and migration of supergene fluid, whereas deep-seated faults facilitate the transport of reductants (e.g., hydrocarbons) and magma. These deep pathways provide the essential thermal energy and additional components for uranium precipitation. Collectively, these processes collectively form high-relief transition zones that act as prime sites for uranium enrichment. In addition, multi-phase extensional events generate permeable conduits and thermal anomalies that are crucial for mobilizing and concentrating uranium from source rocks. This study provides an enhanced understanding of the genesis of sandstone-hosted uranium deposits and offers a theoretical framework for future exploration targeting.

The Guyuan-Hongshanzi uranium metallogenic belt (GHUMB) in North China, particularly the Hongshanzi area within the southern Great Xing'an Range (GXR), hosts uranium mineralization significantly impacted by post-mineralization uplift and exhumation. Field investigations, drill core analysis, electron probe microanalysis (EPMA) and apatite fission track (AFT) dating reveal that uranium orebodies occur within basement uplift zones hosted by the Upper Jurassic Manketou'ebo and Xinmin formations, Lower Permian Dashizhai Formation and Hercynian granodiorite. These orebodies are predominantly shallow and are structurally controlled, associated with hydrothermal alteration including carbonatization, silicification, hematitization, chloritization, fluoritization and sulfidization. Critically, deep-seated potash feldspathization and albitization features are now exposed at shallow depths or the surface. AFT analysis and thermal history modelling indicate rapid cooling during 70-50 Ma, signifying rapid uplift through the partial annealing zone to near-surface levels. This evidence, combined with the surface exposure of deep alteration, demonstrates that primary uranium mineralization initially formed at depths of similar to 2-4 km during the Cretaceous. Subsequent rapid uplift triggered intense erosion, preserving some orebodies shallowly while eroding others and transporting uranium into adjacent basins, potentially contributing to secondary sandstone-type mineralization.

The central Jiangnan orogenic belt in South China harbors a world-class magmatic-hydrothermal mineralization cluster, encompassing supergiant resources of tungsten (W), lithium (Li), copper (Cu), antimony (Sb), and gold (Au). This review synthesizes geological, geochemical, and geochronological data from key deposits to delineate two genetically distinct mineralization series: (1) A Cu-Au-W series associated with I- to S-type porphyry systems, characterized by potassic alteration and silicification; (2) An Nb-Ta-Li-W-Sn-Cu series linked to the evolution of S- to A-type granites, marked by sodic alteration, fluorine-rich muscovitization, and silicification. We propose a unified metallogenic model in which prolonged and episodic magmatism served dual roles: (i) as the source of granite-related rare metal and tungsten mineralization, and (ii) as the thermal engine driving regional hydrothermal circulation. This circulation leached Sb, U, and Au from Precambrian sedimentary sequences, re-precipitating them in structurally controlled sites as intermediate- to low-temperature vein-type deposits. The formation of these diverse yet genetically related deposits is interpreted as the result of extensive crustal magmatism within a unique extensional tectonic regime. This model provides a robust framework for future mineral exploration targeting in this metallogenic belt.

Intra-block ductile shear zones preserve critical records of orogenic deformation driven by plate convergence, encompassing oceanic subduction, continental collision, and post-collisional tectonic reorganization. Newly identified ductile shear zones within the interior of the Central Tianshan Block provide a pivotal record of the convergence processes associated with the North and South Tianshan oceans in the Eastern Tianshan Orogen. Our integrated structural and geochronological analyses of five ductile shear zone segments reveal a polyphase deformation history. Segments 1 and 2 exhibit sinistral transpressional shearing, with the former dominated by top-to-the-north thrusting and the latter by strike-slip motion. Segment 3 is characterized by dextral transpression, Segment 4 by top-to-the-north thrusting, and Segment 5 by dextral strike-slip deformation. Our zircon U-Pb and mica 40Ar/39Ar dating constrain the timing of sinistral transpression to 312-309 Ma, top-to-the-north thrusting to 285 Ma, and dextral strike-slip to 279-277 Ma, while the dextral transpression is constrained to 295 Ma by previous studies. We interpret that the 312-309 Ma sinistral transpression was driven by the northward subduction of the South Tianshan Ocean slab coupled with clockwise rotation of the Tarim Craton. The kinematic switch to dextral shearing between 309 and 295 Ma is interpreted as a consequence of the collision between the Tarim Craton and the Central Tianshan Block, leading to the final closure of the South Tianshan Ocean. The subsequent 295 Ma dextral transpression marks the initial post-collisional tectonic reorganization following this collision. The 285 Ma top-to-the-north thrusting event is a far-field tectonic response to the terminal closure of the North Tianshan Ocean. Finally, the widespread 279-277 Ma dextral strike-slip deformation signifies the first regionally pervasive, orogen-scale intracontinental tectonic reorganization, indicating that the Eastern Tianshan had evolved into a coherent dynamic system. Our study refines the convergence history of the Eastern Tianshan Orogen by deciphering how ductile shear zones within block interiors respond to regional tectonic events, thereby providing significant insights into analogous convergence processes in other accretionary orogens globally.

The Guidong complex in northern Guangdong constitutes a key segment of the Nanling Metallogenic Belt and hosts significant granite-related uranium deposits. However, the petrogenesis of the Yanshanian intrusions in its western part remains poorly constrained because of insufficient high-precision geochronological and mineralogical data. This limitation hinders a comprehensive understanding of regional metallogenesis. To address this, we conducted an integrated study of the Siqian two-mica granite and the Changping biotite granite, incorporating zircon U–Pb geochronology, whole-rock geochemistry, Hf–Nd isotope analysis, and mineral chemistry. Our objectives were to determine their emplacement ages and to investigate their petrogenesis and relationship with uranium mineralization. Zircon U–Pb dating indicates that the Siqian and Changping granites formed at 160 ± 1 Ma and 156 ± 1 Ma, respectively, suggesting both were emplaced during the Early Yanshanian magmatic event. These high-K calc-alkaline rocks are peraluminous (A/CNK = 1.04–1.44), enriched in large-ion lithophile elements (LILEs; e.g., Rb, Th, U), and depleted in high-field-strength elements (HFSEs; e.g., Ba, Nb, Ta). They exhibit negative zircon εHf(t) values ranging from – 15.4 to – 8.4, with corresponding crustal model ages of 1668–2142 Ma. Whole-rock samples show εNd(t) values from –11.2 to –9.1 and Nd model ages of 1693–1860 Ma. These isotopic features indicate that both plutons represent moderately differentiated S-type granites derived from partial melting of Paleo- to Mesoproterozoic metasedimentary sources (e.g., pelitic and psammitic rocks), accompanied by fractional crystallization of minerals such as ilmenite and apatite. Integrated geochemical and mineralogical data reveal that the Siqian and Changping granites are characterized by low Th/U ratios, high F contents, and low magmatic oxygen fugacity—signatures typical of U-rich granites. These attributes provided a favorable lithogeochemical environment for uranium enrichment and acted as primary sources for uranium-bearing, high-temperature hydrothermal fluids, thereby playing a critical role in regional uranium mineralization.

Calcrete-type uranium prospectivity prediction is challenged by the strong heterogeneity of multi-source geoscientific raster datasets, weak anomaly responses, and the lack of explicit heterogeneous information organization in conventional deep learning models. In this study, the Yilgarn Craton of Western Australia was selected as the study area, and a geology-guided fixed-group fusion ResUNet model (GGF-ResUNet) was developed based on 12-channel multi-source geoscientific raster datasets. At the input stage, the evidence layers were divided into four fixed geoscientific proxy groups according to their data modality and geological interpretation, namely gravity, aeromagnetic, radiometric, and geochemical groups, and intra-group channel weighting together with inter-group gating was introduced to enhance the hierarchical representation and adaptive fusion of heterogeneous information. Ablation results showed that GGF-ResUNet achieved better performance than the baseline ResUNet, with AUC increasing from 0.9340 to 0.9740 and F1-score improving from 0.7264 to 0.8356. Further comparative experiments with Attention U-Net, U-Net, SegNet, and FCN showed that GGF-ResUNet achieved comparatively better quantitative performance and more spatially coherent prediction results under the current experimental setting. Without substantially increasing model complexity, the proposed method improves the representation and integration of heterogeneous geoscientific information and provides a feasible technical pathway for calcrete-type uranium prospectivity prediction under weak-anomaly conditions.

Uranium mineral prospectivity mapping is of great significance for uranium exploration and the identification of prospective mineralized areas. Hard labels constructed from buffer zones around known uranium occurrences tend to discretize continuously varying mineralization responses, making it difficult to adequately characterize the continuous variation in mineral prospectivity and the uncertainty associated with mineralization boundaries. To address this issue, this study focuses on sandstone-hosted uranium deposits in the Beverley–Four Mile area, South Australia, and introduces a self-distillation strategy into the ResNet50 model. Soft labels generated by the model are employed to supplement the differences in mineral prospectivity and spatial transitional information that are difficult to represent using hard labels, thereby enhancing the model’s ability to learn mineralization-related features. Experimental results show that, after incorporating the self-distillation strategy, ACC, AUC, F1, and PR-AUC are improved by 3.13

The formation of sandstone-type uranium deposits is closely related to the local tectonic evolution of basins. However, while existing studies mainly focus on qualitative analyses at the regional scale, quantitative research on tectonic ore-controlling mechanisms at the local scale remains relatively limited. Focusing on the Daqing Placanticline and adjacent areas in the northern Songliao Basin, this study applied back-stripping inversion, based on interpretations of seismic and borehole profiles, to quantitatively characterize spatiotemporal variations in local tectonic subsidence since the Late Cretaceous and reveal their controls on sedimentary filling and uranium mineralization. The results show that tectonic subsidence in the study area exhibited distinct staged evolution and regional differentiation. Strong tectonic inversions at the end of the Nenjiang and Mingshui periods caused the greatest uplift and denudation in the southern Daqing Placanticline, forming a large structural window and activating faults that served as channels for the upwelling of deep reducing fluids. Differential tectonic subsidence and sediment supply jointly controlled sedimentary filling patterns. Under the combined influence of low tectonic subsidence rates and sufficient sediment supply, the southern study area exhibited an overcompensated filling pattern and developed meandering river facies with high sandstone percentages, thereby providing favorable channels for fluid migration. In contrast, under the influence of relatively high tectonic subsidence rates or insufficient sediment supply, the central and northern parts of the study area exhibited undercompensated or weakly overcompensated to quasi-balanced filling patterns and were dominated by meandering river, shallow lake, or meandering river–shallow lake facies, with sandbodies generally less developed than those in the south. The spatiotemporal coupling between sedimentary filling during the Sifangtai period and tectonic inversion at the end of the Mingshui period established a uranium metallogenic framework characterized by more favorable conditions in the south than in the central and northern areas. The southern area contains a favorable configuration of three core metallogenic elements—structural window recharge, conductive migration, and fault-related reduction—and therefore represents the most favorable area for sandstone-type uranium mineralization. This study advances quantitative research on local tectonic ore-controlling mechanisms and provides new insights for uranium exploration in the Songliao Basin and similar continental basins.

Data scarcity remains a major challenge for the deployment of dual-energy X-ray transmission (DE-XRT) in intelligent copper ore sorting. Although generative adversarial networks (GANs) have been widely used for data augmentation, most existing methods treat X-ray images as generic visual data and overlook the underlying attenuation mechanism, which may result in physically inconsistent synthetic samples. To address this issue, we propose PhysDualR-GAN, a physics-informed data augmentation framework guided by the Beer-Lambert law. Instead of directly synthesizing low-energy X-ray images, the proposed model predicts attenuation-ratio maps from high-energy images and reconstructs the corresponding low-energy images through a differentiable physics layer, thereby introducing explicit physical constraints into adversarial training. Experimental results on industrial copper ore datasets show that the proposed method generates physically consistent R-value maps that better preserve X-ray attenuation characteristics. The generated samples achieve a radial-spectrum correlation coefficient of 0.9999, an average MSE of 2.73 × 10-2, and an FID of 83.58. More importantly, when used for downstream sorting, the augmented data improve mean Average Precision (mAP) from 90.77% to 98.95%. The proposed framework also maintains competitive sorting performance under reduced training-data conditions. These results demonstrate that PhysDualR-GAN provides an effective physics-informed solution for data-efficient training in X-ray-based mineral sorting.

The Ordos basin is a crucial comprehensive energy base in China, where coal and sandstone-type uranium resources frequently coexist in an associated form within the same clastic aquifer system, characterized by a typical “coal below, uranium above”configuration. This coal-bearing uranium aquifer possesses a dual nature: it serves as the direct water-filling aquifer for underlying coal seam extraction, impacting coal mine safety, and simultaneously acts as the reservoir and migration pathway for in-situ leaching (ISL) of the overlying uranium deposit, influencing the effective displacement and recovery of lixiviant. Its spatial heterogeneity, manifested in mineral composition, pore structure, sandbody geometry, and spatial distribution, constitutes a significant geological challenge for the safe and efficient coordinated mining of these resources. Therefore, developing a methodological framework capable of systematically revealing and three-dimensionally characterizing the multi-scale heterogeneity of this aquifer is of considerable theoretical and practical value.Focusing on the coal-bearing uranium aquifers of the Jurassic Zhiluo formation and Cretaceous Luohe formation in the Wushen and Hangjin banner areas of the north-central Ordos Basin, a multi-scale, multi-attribute coupled research methodology encompassing micro-, meso-, and macro-scales is proposed. At the microscopic scale, techniques including polarizing microscopy, scanning electron microscopy (SEM), mercury intrusion porosimetry, nuclear magnetic resonance (NMR), and core computed tomography (CT) scanning are employed to quantitatively analyze the compositional and pore-structural heterogeneity at the micrometer level. At the mesoscopic scale, outcrop analysis is integrated with core-logging prediction models to characterize sandstone morphology and vertical heterogeneity at the meter scale. At the macroscopic scale, based on sequence stratigraphic division and sedimentary facies distribution analysis, and combined with geostatistical methods such as sequential Gaussian simulation, a multi-attribute three-dimensional geological model controlled by sedimentary facies is established. This model achieves a three-dimensional spatial characterization of aquifer heterogeneity at the kilometer scale.The results demonstrate that this methodological framework can systematically reveal the heterogeneity characteristics of the coal-bearing uranium aquifer from micro- to macro-scales. The findings provide direct geological basis and a modeling foundation for predicting seepage fields, which is essential for preventing roof water hazards in coal mining and for the in-situ leaching of sandstone-type uranium deposits. This research holds significant importance for ensuring the safe and efficient coordinated mining of coal and uranium resources.

The HLJ and DL sandstone-type uranium deposits in the southern Songliao Basin show marked differences in mineralization characteristics, yet their contrasting uranium enrichment mechanisms have not been systematically compared. Based on scanning electron microscopy (SEM), electron probe microanalysis (EPMA), X-ray diffraction (XRD), and rock geochemistry, this study shows that uranium minerals in the HLJ deposit are commonly associated with TiO2 minerals, kaolinite, and pyrite, whereas those in the DL deposit are mainly associated with pyrite and organic matter. This contrast indicates two fundamentally different microscopic uranium enrichment mechanisms: in the HLJ deposit, uranium enrichment was controlled by a multistage mineral-interface process involving TiO2 minerals and kaolinite-related uranium adsorption, remobilization, and reprecipitation, with pyrite contributing to local reduction and final uranium fixation; whereas in the DL deposit, uranium precipitation was more directly controlled by pyrite- and organic matter-dominated reduction barriers. Furthermore, the HLJ deposit records stronger early acidic alteration, as evidenced by more pervasive feldspar kaolinization, higher kaolinite abundance, and the widespread close association of uranium minerals with kaolinite. Such alteration likely promoted uranium adsorption and local enrichment at reactive mineral interfaces, and enhanced pyrite destabilization, thereby contributing to the coupled enrichment of U with Mo, As, and Pb. In addition, the HLJ deposit experienced a stronger and higher-temperature late hydrothermal overprint, as indicated by coarser pitchblende (15–60 μm, locally up to ∼ 100 μm) and higher fluid-inclusion homogenization temperatures, which likely promoted uranium remobilization and short-distance redistribution. These differences were most likely related to contrasts in local structural characteristics and associated fluid-migration pathways. The results provide important insights into the mechanisms of uranium mineralization in the southern Songliao Basin and have significant implications for exploration methodologies and subsequent mining.

Mineralization in the sandstone-type uranium deposits of the Songliao Basin is characterized by hypogene fluid alteration, as recorded at the regional tectonic, deposit, and mineral scales. However, the mechanism of uranium enrichment under the influence of hypogene fluids remains poorly understood. Based on their spatial distribution and cross‑cutting relationships, the carbonate minerals are divided into four stages, and this study focuses on the hydrothermal ore‑forming stage. Petrographic observations show complex textures and compositional zoning in carbonate minerals, indicating multiple precipitation–dissolution stages and synchronous precipitation of U‑minerals and pyrite, which provide a basis for evaluating the influence of hypogene fluids on uranium enrichment. In situ U-Pb dating using laser ablation–inductively coupled plasma mass spectrometry (LA-ICP-MS) provides well-defined ages of 40.1 ± 1.3 Ma for U-rich ankerite (Ank-2), 36.3 ± 0.2 Ma for U-poor high-Mg siderite cement (Sd-3), and 35.6 ± 3.4 Ma for U-poor calcite vein (Cal-5). These ages are approximately synchronous with widely distributed diabase dikes (zircon ages of 39.1 ∼ 42.0 Ma) in the southern Songliao basin. Compared with siderite and dolomite, U-rich ankerite shows higher enrichment in middle and heavy rare earth elements. Hydrothermally altered sandstone, especially the red altered uranium ore, shows higher contents of U, Co, Ni, V, Mo, Zr, Y, W, Pb, and HREE, and U exhibits positive correlations with W, As, Zr, and P. Most uranium minerals (including pitchblende, titanium-bearing uranium minerals, and coffinite) contain high concentrations of elements such as P, Ca, Zr, La, and Ce. Sr isotope ratios decrease progressively from early calcite (Cal-1) through ∼60 Ma microbial dolomite (Dol-1) to hydrothermal carbonates, with U enrichment confined to 87Sr/86Sr of 0.706–0.709. The above evidence indicates that thermal fluids activate and alter the uranium already present in the ore-bearing target layer, leading to further uranium enrichment and the formation of high-grade uranium ores. Because carbonate is a common mineral in U deposits worldwide, and other minerals in such deposits suitable for isotope dating are generally absent, in situ U-Pb dating of carbonates opens a new window for better defining the ore genesis of this globally important U deposit and for tracking hydrothermal fluid flow in sedimentary basins.

The Mianhuakeng deposit, located within the Zhuguangshan batholith in the Nanling area, is currently recognized as the largest granite-related uranium deposit in China. A portion of the uranium ore bodies is spatially associated with NE-trending mafic veins within the granite. In this study, the field investigation, zircon U-Pb dating, S and Pb isotope analysis, and whole-rock geochemical analysis were conducted on these mafic veins to explore their crystallization age, petrogenesis, tectonic setting, and relationships with uranium mineralization. The weighted mean result of zircon U-Pb is 189 +/- 3 Ma, suggesting that the mafic dyke was crystallized during the Early Jurassic. The whole-rock geochemistry and isotopes exhibit characteristics of intraplate basalts, suggesting that the mafic dykes originate from an enriched mantle source consisting of garnet-spinel lherzolite, with an estimated partial melting of 1%-5%. Mafic magmas underwent low-degree contamination from the lower crust during upwelling, induced by the extension of the lithosphere during the Early Jurassic. The analyses of pyrite sulfur isotopes in mafic samples vary between -2.9 parts per thousand and 1.8 parts per thousand, significantly different from that of pyrite (-14.4 parts per thousand to -7.8 parts per thousand) formed during the uranium mineralization. Furthermore, the ages of the pitchblende of 127-54 Ma are much younger than the crystallization ages of mafic dykes, indicating that the mafic magmas did not contribute to the uranium mineralization of Mianhuakeng deposit during magmatism. However, the abundant reducing minerals (e.g., pyrite, hornblende, and Fe2+-bearing minerals) in the mafic dykes can act as a redox barrier, reducing mobile U6+ to immobile U4+ during fluid-rock interaction, thereby facilitating uranium precipitation from the hydrothermal ore-forming fluids. The secondary fractures created by the intrusion of mafic magma probably provided favorable pathways for the movement of hydrothermal fluids.

The middle segment of the northern margin of the North China Block is heavily covered by thick Quaternary strata, and as a result, only limited physical research data have provided restrict understanding of the tectonic evolution of this area and the neighboring eastern section of the Central Asian Orogenic Belt. This article presents detailed petrographic observations and rock geochemistry, zircon U-Pb dating, zircon trace elements, and in-situ Hf isotopic analysis of the Baimashigou pluton by drilling-core samples in the coverage area. The study shows that the Baimashigou pluton formed in the Late Permian (254 +/- 0.7Ma). The analyzed rock samples are characterized by high silicon contents (SiO2=69.13%similar to 72.27%), alkali and potassium-rich (K2O+Na2O=8.20%similar to 9.51%, K2O/Na2O=0.98 similar to 2.52) and (weak) peraluminous (A/CNK=0.98 similar to 1.10), indicating they belong to high-K peraluminous unfractionated S-type granite. Furthermore, the geochemical characteristics of these samples also reveal that they are formed by the partial melting of metasandstone with low clay content (CaO/Na2O=0.27 similar to 0.38) and rich in primary crustal debris (epsilon(Hf)(t)=-3.93 similar to 6.36) on top of the descending slab under a relatively low temperature (612.9 similar to 799.8 degrees C), high pressure ((Gd/Yb)(N)=0.88 similar to 2.41, delta Eu=0.72 similar to 0.83, Sr=405.1x10(-6)similar to 545.6x10(-6), Y=11.13x10(-6)similar to 14.55x10(-6), Sr/Y=27.83 similar to 41.16), and high oxygen fugacity (QFM similar to QFM+10) environment. Combined with previous research results, it is believed that the middle segment of the northern margin of the North China Block had entered the collision orogeny stage before the end of the Permian Period. The Baimashigou pluton provides direct petrological evidence of the plate collision and amalgamation. The study holds significant importance for further analyzing the geological structure and tectonic evolution of the eastern segment of the Central Asian Orogenic Belt, particularly within the coverage area.