Origin and Role of Kaolinization in Roll-Front Uranium Deposits and Its Response to Ore-Forming Fluids in the Yili Basin , China

Kaolinite is a common mineral found in most Chinese sandstone-hosted uranium deposits. It occurs particularly in coal-bearing clastic rocks in northwest China, such as the uranium deposits in the Yili Basin, which is well known for hosting several largescale roll-front uranium deposits. Previous studies have provided limited information on the origin of kaolinization and its role in the uranium mineralization. This study uses gas hydrocarbon, fluid inclusions, O and H isotope analysis, and scanning electron microscopy observations to investigate the formation of kaolinite in ore-hosting rocks from the Mengqiguer uranium deposit in the southern margin of the Yili Basin and to determine its role in the uranium mineralization. Results suggest that kaolinization is intense in the coaland ore-bearing clastic rocks and that it is related to leaching of feldspar by acidic fluids. Vermicular kaolinitewas formed by hydrocarbon-bearing fluid generated from coal and carbonaceous mudstone during a shallow-burial diagenetic stage at low homogenization temperatures ranging from 69 to 78C and at relatively high salinities of 7.6−11.0 wt% NaCleq. Consequently, silicate minerals (such as feldspar) were leached and created secondary pores that hosted the subsequently formed uranium minerals. In contrast, micritic kaolinite was formed by infiltration of meteoric fluid enriched in U and O2 at low homogenization temperatures of 51−63Cand low salinities of 1.2−3.7 wt% NaCleq. U 6+ was sorbed by themicritic kaolinite through cation exchange, forming aU-bearing kaolinite complex; it was also reduced by pyrite and carbondetrital, thereby precipitating at the acidic oxidation front.The results of this study confirm that intense kaolinization is closely related to uraniummineralization in coal-bearing clastic rocks.


Introduction
Kaolinite is one of the most common and significant diagenetic minerals, and its formation is controlled by the composition of the material source and detritus, pH, temperature, pore water, and sealing or opening of the diagenetic system [1,2].The relation between pore water and diagenetic kaolinite morphology in sandstones from different depositional environments was studied by Irwin and Hurst [3].According to their model, meteoric water flux results in the rapid generation of "nutrient" ions (A1 3+ , Si 4+ ) for kaolinization through the dissolution of unstable detrital grains, such as feldspar and mica; the consequent rapid kaolinite growth then affects the texture of these minerals.However, in the absence of meteoric water flux, such as in shale-enclosed marine sandstone, kaolinite growth proceeds slowly and produces fine-grained euhedral aggregate [4][5][6][7][8].It has also been found that the formation of kaolinite in sandstones is related to migration of acidic fluids after, or during, diagenesis; for example, hydrocarbon-rich fluids in sandstones dissolve feldspar grains and produce kaolinite [9][10][11][12][13].The variety of kaolinite in sandstone is nearly always wellcrystallized, and it often occurs in vermicules and as discrete aggregates.Morphological and particle size variations in such kaolinite, and their textural relationships with other sandstone components, have been used as indicators to identify whether they are authigenic [14].In addition, the various morphologies of kaolinite and its distribution in sandstones suggest an interaction between pore water composition and detrital minerals.
According to the typical model of alteration and distribution of minerals in roll-front uranium deposits established by Harshman and Adams [15] and DeVoto [16], the kaolinite-rich zone, known as the "white bleaching zone" [17,18], is formed at the limonitic front and is closely associated with uranium mineralization; therefore, it is an effective prospecting indicator for sandstone-hosted uranium deposits [15,19,20].It has been verified that kaolinization in sandstone-hosted uranium deposits resulted from either local organic matter (OM) or foreign organic matter [21], and it is particularly related to liquid or gaseous OM, such as humic acid and hydrocarbon-aqueous species.
One of the most important roles of kaolinite in the uranium mineralization is its sorption of uranyl aqueousspecies.This effect has been experimentally examined in great detail with respect to the types of sorption complexes, changes in pH and the ionic strength of the electrolyte solution, and concentrations of U [22][23][24][25][26][27][28][29][30].Dong et al. [28] studied the sorption of uranyl species onto kaolinite and goethite, two of the most dominant minerals in sandstonehosted uranium deposits, and showed that kaolinite was a stronger sorbent than goethite at pH < 4.This strong sorption capacity can be interpreted as being attributed in part to the ion exchange sites of kaolinite, which enable strong sorption of certain metal ions (e.g., UO 2 2+ , Cu 2+ , and Pb 2+ ) at low pH conditions.
Although kaolinization is widely found in sandstonehosted uranium deposits, and it plays an important role in uranium sorption, few studies have focused on the origin of kaolinization and its role in the uranium mineralization.Kaolinite is a common mineral in most Chinese sandstonehosted uranium deposits, particularly in the coal-bearing clastic rocks of northwest China.The deeply buried (>500 m) high-grade Mengqiguer uranium deposit is a typical rollfront uranium deposit hosed in the coal-bearing clastic rocks in the southern margin of the Yili Basin in northwest China.It contains an estimated resource of more than 10000 tU within an area of approximately 30 km 2 , and an average uranium grade of 0.6%, which is much higher than the medium grade (0.05-0.4%) of this type [31].
The main focus of this study is to determine the association between kaolinization and uranium mineralization in the Mengqiguer uranium deposit.In addition, the origin of kaolinite and its response to ore-forming fluids are investigated using X-ray diffraction (XRD), SEM, gas hydrocarbon analysis, fluid inclusions, and O and H isotopes.

Geological Setting
The Yili Basin is a composite intermontane basin that was formed on a micromassif during the Tianshan orogen.The Yili block is often considered to be a part of the Central Tianshan microcontinent, which has a Proterozoic basement extending westward into Kazakhstan and Kyrgyzstan [33,34].The block is triangular and bordered by two late Paleozoic suture zones: the North Tianshan Suture to the north and the Central Tianshan Suture to the south.Both suture zones were reworked by Permian postorogenic wrench faults, namely, the North Tianshan Fault (NTF) and the Nalati Fault (NF) (Figure 1(a), Liu et al. [35]).Overall, the western part of the southern margin of the Yili Basin shows weak tectonic activity and consists of a series of gentle anticlines and synclines, forming a gently dipping north monoclinic structure (basin slope).In contrast, the eastern part shows evidence of stronger tectonic activity and consists of complex asymmetrical synclines and anticlines.The middle part, consisting of the Zakisitan syncline, is a tectonic transition zone where tectonic activity is strong enough to invert or upturn sedimentary strata in the basin margin.
A series of roll-front uranium deposits have been discovered in the southern margin of the Yili Basin (Figure 1(b)).The Mengqiguer uranium deposit is located in the transition zone of Zakisitan syncline, which is an incomplete box fold with east-west-south side-flanks tilting to the northeast, and a synclinal axis dipping to the northeast.One of the probable uranium sources for the deposit are the acidic volcanic rocks, pyroclastic rocks, and granite of the Carboniferous-Permian age, which are exposed in the ranges of southern provenance.The average uranium content of these rocks is 6.2-12.9ppm [36], and it is considered that uranium was leached and transported to the basin by oxygen-rich groundwater during weathering and erosion.Other probable contributors to the uranium source are the primary gray sandbodies and gray mudstones in the ore-hosting strata, which contain 6-10 ppm and 7-16 ppm uranium, respectively.Furthermore, the spring water in the southern source region and sedimentary terrains contains anomalous uranium concentrations of up to 9 ppb and 100 ppb, respectively, which also supports hypotheses that the uranium for ore formation could be derived from either one of these sources.
The sedimentary strata within the mining area belong to the Lower Jurassic Badaowan Formation (J 1 b), Lower Jurassic Sangonghe Formation (J 1 s), Middle Jurassic Xishanyao Formation (J 2 x), Middle Jurassic Toutunhe Formation (J 2 t), Cretaceous, and Neogene strata.The target ore-bearing strata for exploration are Middle-Lower Jurassic in age and include the lower member of the Sangonghe Formation (J 1 s 1 ), upper member of the Sangonghe Formation (J 1 s 2 ), lower member of the Xishanyao Formation (J 2 x 1 ), and the upper member of the Xishanyao Formation (J 2 x 3 ).The target strata were deposited during multiple sedimentary cycles with a fluvial positive rhythm, and they represent braided river delta sediments that formed coal-sand-coal stacks (Figure 2).The sandbodies between two coal seams are rich in carbon debris and pyrite, with an organic carbon content ranging from 0.11% to 1.93%, and carbon debris and numerous pyrite assemblage are dispersed in the sandstone cement.The host massive rocks have an argillaceous matrix and consist of conglomerate, sandy conglomerate, sandstone, and 10-18% gravel.
The uranium orebodies are tabular, lenticular, and rollshaped (Figure 2).Tabular orebodies generally occur in the flanks of the interlayer oxidation zone, which are located have thicknesses between 1.5 m and 8.6 m, and are located in the downsizing sites of host sandbodies.However, orebody shapes do not present in isolation; a cross section shows that each shape is reworked into another.An analysis of host rock colors, Fe-minerals, alteration, total organic carbon (TOC), and Fe 3+ /Fe 2+ shows that the interlayer oxidation zone of the Mengqiguer uranium deposit can be divided into an oxidation zone, transition zone (mineralized zone), and a primary zone.The oxidation zone can be further subdivided into a strong oxidation and weak oxidation zone (Table 1), where each zone tends to transition into the other.The strong oxidation zone is characterized by the highest Fe 3+ /Fe 2+ value (>0.9), the lowest TOC (0.05-0.07%), and U content (4.0-6.0 ppm), has strong hematitization and limonitization, and is red, or light-red, in color (Figure 3(a)).The weak oxidation zone contains rare carbon debris (TOC of 0.08-0.10%),limonite, and rare pyrite (Fe 3+ /Fe 2+ of 0.5-0.8)and presents as a paleyellow or yellowish-white color (Figure 3(b)).The transition zone (mineralized zone) has the highest TOC (>0.50%) and U content (>100 ppm), with weak silicification and strong kaolinization.The primary zone contains abundant pyrite, magnetite, and siderite and has moderate Fe 3+ /Fe 2+ values (0.2-0.4) and TOC (>0.20%).Both the mineralized and primary zones are rich in carbon debris and pyrite, and their rock colors are gray or grayish-white (Figures 3(c) and 3(d)).

Samples and Methods
Different to the methods used in previous studies, this study uses a representative sample collection of drill cores obtained from the entire interlayer oxidation zone.Samples consist of conglomerate, sandy conglomerate, and coarseand medium-grained sandstone; 4 samples are obtained from the strong oxidation zone, 14 samples from the weak oxidation zone, 4 samples from the mineralized zone, and 9 samples from the primary zone.In the mineralized zone, an HD-2000 handheld gamma radiometer (manufactured by the CNNC Beijing Research Institute of Uranium Geology) was used to screen samples from the drill cores.The measurement time used to screen each sample was 10 s, and ore samples with measured values of more than 100 Ur (ca. 100 ppm) were collected.
XRD analysis was conducted on all collected bulk core samples.In this respect, samples were firstly cleaned in an ultrasonic bath and then analyzed using a PANalytical X'Pert PRO X-ray diffractometer (Beijing Research Institute of Uranium Geology).The incident X-ray radiation was obtained using a line-focused PW3373/10 Cu X-ray tube operating at 40 kV and 40 mA, with an angle ranging from 5 ∘ to 90 ∘ .Mineral identification followed the procedures set out in Moore and Reynolds [37], and the Rietveld method of quantitative analysis was applied to obtain the relative abundance of the minerals.
In addition to XRD, an SEM-EDS (TESCAN-VEGAII LMU) (SEM) and LINK-ISISX alpha-ray spectrometer were used to differentiate the morphologies of kaolinite and investigate the association between kaolinite and uranium minerals.In this respect, samples were cut into 1 cm 3 cubes and then gold-plated.Experiments were conducted at the Experimental Research Centre of the Research Institute of Petroleum Exploration & Development.
To determine the origin of kaolinite, a number of sandstones were obtained at various distances from the underlying coal seam.These sandstones were crushed and passed through a 80-mesh sieve; no less than 50 g of sample was then dissolved with hydrochloric acid in a beaker.Gas generated from the reaction between carbonate and hydrochloric acid was collected and passed through an alkaline solution to remove CO 2 , prior to examination using a gas chromatograph (GC7890F) to determine the content of methane (C 1 ), ethane (C 2 ), propane (C 3 ), n-butane/isobutene (nC 4 /iC 4 ), and n-pentane/isopentane (nC 5 /iC 5 ), with an analytical precision of 0.05 l/kg.In addition, Linkham THMSG-600 heating-freezing stage and Renishaw Invia Reflex-type confocal laser Raman microspectrometry were conducted at the Beijing Research Institute of Uranium Geology to analyze fluid inclusions in the secondary overgrowth zones, in microfractures of quartz grains, and in the carbonate.To test physicochemical properties of the fluids, temperature was tested using the homogenization method, and salinity was tested using the freezing method [38][39][40][41].
From the results of XRD analysis, samples with a high kaolinite content were selected for further tests.After cleaning, the selected bulk samples were crushed and passed through a 60-mesh sieve into a beaker filled with 500 ml distilled water, stirred for 3 min, and kept for 10 min.The suspension in the beaker was then moved to another beaker and dried to obtain the separation product (the main component was kaolinite).Based on XRD data, products with a kaolinite content of more than 80% were selected for H-O isotopic analysis, using the methods of Clayton and Mayeda [42] and a Finnigan MAT253-type mass spectrometer.Oxygen gas was produced by treating the samples with BrF 5 in an externally heated nickel reaction vessel, and hydrogen was determined by treating H 2 O with zinc at 550 ∘ C. SMOW was adopted as the standard, with a precision of ±2‰ for D and ±0.2‰ for  18 O.All the above processes and analyses were conducted at the Key Laboratory of Uranium Resources Exploration and Evaluation Technology, CNNC.

Content and Morphology of Kaolinite in Host
Rocks.XRD results of host rocks from different geochemical zones are presented in Table 2.The average percentage clay mineral content in all geochemical zones varies between 10.9% and 17.5% (percentage contents of clay minerals in the strong and weak oxidation zones are 13.7% and 15.1%, resp.; 17.5% in the mineralized zone; and 10.9% in the primary zone).Kaolinite is the dominant clay mineral (accounting for 50.7-72.5% of clay minerals), followed by mixed illite/smectite and illite.The highest average content of kaolinite reaches 72.5% in the grayish-white host rocks within the mineralized zone; the content increases from the strong oxidation zone to the mineralized zone and decreases sharply in the primary zone.
Based on the SEM observations, kaolinite in the host rocks is found to be mainly vermicular and micritic.Micritic kaolinite is found in both intergranular pores and pores within dissolved feldspar grains (Figure 3(e)) or dispersed on the surface of quartz and other grains (Figure 3(f)).It has an irregular form, dissolved margins, and pseudohexagonal crystal plates, with crystal sizes varying between 2 m and 5 m.Vermicular kaolinite, in contrast, has large pseudohexagonal plate crystals with regular crystal edges and crystal sizes ranging between 10 m and 20 m.
Vermicular kaolinite is rare and is mostly found in the gray or grayish-white sandstones of the primary and mineralized zones and occasionally in the weakly oxidized sandstones.Micritic kaolinite is more common and is found in sandstones within the oxidation and mineralized zones.In the sandy conglomerates of the oxidation zone, which have good porosity, micritic kaolinite is dispersed on the surface of grains and is superposed upon vermicular kaolinite (Figures 3(g) and 3(h)).In addition, gypsum growing on the surface of vermicular kaolinite (Figure 3(i)) is possibly formed under the same conditions as micritic kaolinite.Authigenic quartz grains and quartz overgrowth are widespread in the host rocks, particularly in the mineralized zone.Generally, numerous authigenic quartz grains are formed in intergranular pores and covered by micritic kaolinite (Figure 3(j)).Strong quartz overgrowth occasionally blocks the pores of sandstones and is also covered and replaced with micritic kaolinite (Figures 3(k) and 3(l)).From these results, it is proposed that the authigenic quartz and quartz overgrowth possibly formed during diagenesis, earlier than the formation of micritic kaolinite.
Based on the replacing relationships described above, a paragenetic sequence associated with kaolinite in the Mengqiguer uranium deposit was constructed and is illustrated in Figure 4 (where dashed lines are used to represent  minerals at a relatively weak intensity, or decreasing amounts of minerals, as determined using microscopic observations).Authigenic quartz grains, quartz overgrowth, and vermicular kaolinite are observed in all geochemical zones and belong to the diagenetic stage.Feldspar-leaching, micritic kaolinite, gypsum that pervasively cover or replace quartz overgrowth, and vermicular kaolinite observed in the oxidized and mineralized zones are assigned to the mineralization stage.Quartz overgrowth and vermicular kaolinization are noted to be relatively weak in the oxidized and mineralized zones, and this is partly because they are replaced with micritic kaolinite.

Association between Kaolinization and Uranium. SEM observations combined with measurement results of EDS
indicate that uranium minerals and micritic kaolinite fill in the dissolved pores of K-feldspar and intergranular pores (Figures 5(a) and 5(b)).Uranium minerals (uraninite) generally occur as grape-like crystals imbedded in micritic kaolinite (Figures 5(c) and 5(d)).In addition, uranium is adsorbed by micritic kaolinite, forming a kaolinite-uranium complex with an amorphous state (Figures 5(e), 5(f), 5(g), and 5(h)).Micritic kaolinite has a close association with uranium mineralization.

Gas Hydrocarbon in Host
Rocks.Results of gas hydrocarbon analysis of host rocks are shown in Table 3.Of these gases, the methane (C 1 ) content is dominant (but it decreases sharply with an increase in distance from the underlying coal seam), followed by ethane (C 2 ) and then propane (C 3 ).Other gases (such as nC 4 , iC 4 , nC 5 , and iC 5 ) are present in minor amounts.Consequently, the C 1 /C 2 + and C 1 /∑C values are elevated and range from 11.71 to 21.98 and 0.92 to 0.96, respectively; there is an irregular increase in these values with respect to the increase in distance from the underlying coal seam.
Although relatively rare, LH and HA inclusions are generally present together, have elliptical and subrounded shapes, and coexist with AI inclusions (Figure 6  found in quartz overgrowth zones or at the quartz overgrowth boundary.GI inclusions with a subrounded shape are relatively abundant in samples (Figure 6(d)), and present with a yellowwhite fluorescence color (Figure 6(e)).Raman spectroscopic analysis confirms the presence of methane inclusions (Figure 6(f)).GI inclusions rarely occur alone and frequently coexist with GA inclusions; both occur primarily along healed microfractures in quartz grains and carbonate cement, some of which contain coexisting AI inclusions with dimensions of 2−15 m.
AI inclusions are abundant in all samples; they are typical two phase inclusions, present with rounded and subrounded shapes, and have small vapor bubbles at room temperature.Their sizes vary greatly and are generally between 2 m and 18 m in diameter, with minimum and maximum dimensions of 2 m × 2 m and 15 m × 20 m, respectively.Most AI inclusions occur along healed microfractures (Figure 6(g)) and unhealed microfractures in quartz grains (which do not terminate at the contact between detrital grains) and quartz overgrowth (Figure 6(h)).A small percentage of aqueous inclusions occur in quartz overgrowth zones (Figure 6(i)).AI inclusions coexisting with other types of inclusions occur much less frequently.

Microthermometry. AI inclusions hosted in quartz
healed and unhealed microfractures, and in overgrowth zones (particularly those coeval with LH or GI inclusions in a microfracture), were selected for microthermometric analysis, and the fluid inclusion assemblage (FIA) concept [43] was used to guide microthermometric analysis.Randomly distributed fluid inclusions in carbonate cement were not adopted for microthermometric analysis, because it is difficult to identify that they were entrapped at the same time (i.e., belonging to an FIA) [44].Homogenization temperatures ( ℎ ) and ice-melting temperatures ( -ice ) were determined for AI inclusions with relatively low vapor percentages (<15%) and capability of homogenizing to a single liquid phase.
Microthermometric and salinity data of AI inclusions in ore-bearing sandstones are listed in Table 4. AI inclusions in quartz healed microfractures have  ℎ values ranging from 64 ∘ C to 76 ∘ C and salinities from 0.4 wt% NaCl eq to 6.6 wt% NaCl eq .The  ℎ values of AI inclusions coeval with GI and LH in quartz healed microfractures have similar distribution ranging from 71 ∘ C to 78 ∘ C and 69 ∘ C to 74 ∘ C, respectively, with salinities from 8.4 wt% NaCl eq to 11.0 wt% NaCl eq , and 7.6 wt% NaCl eq to 8.3 wt% NaCl eq , respectively.AI inclusions in quartz overgrowth zones have  ℎ values ranging between 75 ∘ C and 86 ∘ C and salinities from 3.4 wt% NaCl eq to 7.4 wt% NaCl eq .The  ℎ value range of AI inclusions in quartz unhealed microfractures varies between 51 ∘ C and 63 ∘ C, with salinities from 1.2 wt% NaCl eq to 3.7 wt% NaCl eq .
Generally, the  ℎ values of AI and AI coeval with GI and LH inclusions in the quartz healed microfractures have a wide range, mainly varying between 65 ∘ C and 75 ∘ C (Figure 7).In particular, there are relatively higher  ℎ values for AI coeval with GI and LH inclusions (70−75 ∘ C).AI inclusions in the quartz overgrowth zones have the highest  ℎ values, and those in the quartz unhealed microfractures have the lowest  ℎ values.Salinity values increase slightly with  ℎ values: the lowest values are found in AI inclusions of unhealed microfractures, and the highest values are found in AI coeval with GI inclusions of healed microfractures.[45].Although vaporonly inclusions may be produced by necking-down or postentrapment stretching [43,46], their widespread occurrence and abundance in the examined samples contradicts such interpretations.Thus, AI inclusions with the lowest vapor percentages are considered to represent an homogeneously entrapped liquid phase, and the GI inclusions represent an homogeneously entrapped vapor phase.Furthermore, the  ℎ values of AI inclusions with the lowest vapor percentages represent trapping temperatures, and no pressure correction is required.Accordingly, fluid pressures ( ℎ ) at homogenization temperatures represent trapping pressures [45]:  ℎ values of AI coeval with GI and LH inclusions in the quartz healed microfractures calculated with FLUIDS (Bakker 2003) cover a range from 5 bar to 6 bar.

O and H Isotopic Compositions of Kaolinite.
The O and H isotopic compositions of kaolinite-dominated clay minerals with a kaolinite purity of >80% are listed in Table 5.With respect to the low temperature of kaolinite formation in this study,  18 O (H2O)V-SMOW values were calculated based on the fractionation equation between minerals and water following Savin [47].As micritic kaolinite is dominant in the extracted kaolinite-dominated clay minerals, it is considered that the O and H isotope data in Table 5 are related to fluid associated with micritic kaolinite formation.As micritic kaolinite was formed after quartz overgrowth, with respect to the result of the paragenetic sequence analysis presented in Figure 4, the maximum  ℎ values of AI inclusions in the quartz unhealed microfractures were used to calculate  18 O (H2O)V-SMOW values: these range from −9.1‰ to −4.8‰ in the strong oxidation zone, from −6.6‰ to −3.3‰ in the weak oxidation zone, and from −4.4‰ to −1.3‰ in the mineralized zone, with a gradually elevating trend.The same trend can be seen in the D (H2O)V-SMOW values between the strong oxidation zone and the mineralized zone, where the kaolinite-forming fluid underwent sufficient water-rock reaction and O and H isotopic fractionation.Figure 8 shows that all of the D (H2O)V-SMOW and  18 O (H2O)V-SMOW values fall closely under the meteoric water line, which indicates that either no metamorphic water was involved in kaolinite formation, or primary magmatic water was involved.These results could also indicate that surface meteoric fluid was responsible for the formation of micritic kaolinite, and, in combination with the analysis of kaolinite morphology presented in this study, it is suggested that the micritic kaolinite is closely associated with uranium mineralization.

Discussion
5.1.Origin of Ore-Forming Fluids.The vermicular and micritic kaolinite and the five types of fluid inclusions suggest that the host rocks underwent two types (or stages) of fluid alteration.The LH, HA, and GA inclusions present with blue and yellowish-white fluorescence, which indicates the presence of a hydrocarbon-bearing fluid.In addition, the volatile components in the GA inclusions identified by laser Raman suggest that the hydrocarbon-bearing fluid contains CH 4 .Results of gas hydrocarbon analysis of host rocks show that the contents of CH 4 in samples from overlying sandstones adjacent to the coal seam are relatively higher than those in other samples.Additionally, the elevated values of C 1 /C 2 + and C 1 /∑C vary (11.71-21.98 and 0.92-0.96,resp.) and fall in a range representing a coal-derived origin [48].Based on carbon isotope analysis in the southern margin of the Yili Basin and with respect to the decarboxylation of OM that produces organic acids, Wu et al. [49] determined that hydrocarbons in the ore-forming fluid originate from coal beds within the strata.It thus can be concluded that the hydrocarbon-bearing fluid containing CH 4 in the host rocks is an organic acidic fluid that was generated from OM during diagenesis.
In addition, the O and H isotope compositions of micritic kaolinite present a regular increase from the strong oxidation zone to the mineralized zone, which is indicative of the formation of micritic kaolinization and oxidation zones via meteoric water infiltration.It is interpreted from these results, and the close relationship between micritic kaolinite and uranium minerals evidenced by SEM observation that oxygenrich interlayer infiltration fluid (characterized by meteoric water) was involved in uranium mineralization.It is also inferred that the interlayer oxidation fluid played a crucial role in ore formation, owing to the spatially close association between ore bodies and oxidation zones.
Fluid immiscibility, inferred from the coexistence of liquid-dominated and vapor-dominated fluid inclusions within individual FIA in quartz healed microfractures, indicates a shallow-burial environment [45].Hydrocarbon-bearing fluid inclusions confined in the quartz healed microfractures that terminate at the overgrowth boundary suggest that the hydrocarbon-bearing fluid was produced during the shallowburial diagenetic stage, which is characterized by low  ℎ values ranging from 69 ∘ C to 78 ∘ C, relatively high salinity values (7.6-11.0wt% NaCl eq ), and low trapped pressures (5-6 bar).Fluid inclusions in the quartz overgrowth zones are interpreted as primary inclusions occurring during diagenesis [50].The AI-only inclusions in the quartz overgrowth zones suggest that no hydrocarbon was present in the pore fluid during this diagenesis or that a rare hydrocarbon existed.Fluid inclusions in the quartz unhealed microfractures record postdiagenesis pore water migration [50], and AI inclusions in the quartz unhealed microfractures are characterized by low  ℎ values (51-63 ∘ C) and low salinity values (1.2-3.7 wt% NaCl eq ), which represent a supergene meteoric water origin that can be interpreted as an oreforming fluid.This result is also supported by the H-O isotopic plot (Figure 8).

Kaolinite Genesis.
The varying clay mineral content suggests that intensive argillization occurred within the oxidation and mineralized zones.The regular changes in the kaolinite content observed in drill cores from oxidation zone to mineralized zone suggest that intense kaolinization in the mineralized zone was mostly formed in the oxidation zone but migrated to the mineralized zone via pore fluid.Based on SEM observations, vermicular kaolinite is mainly found in intergranular pores in sandstones and is widely interpreted to have been formed during shallow burial (<2000 m) in a stable diagenetic environment [51,52].The later-formed micritic kaolinite is superposed upon vermicular kaolinite and may be related to rapid crystallization under supergene conditions.In addition, the H-O isotopic plot suggests that surface meteoric fluid is responsible for the micritic kaolinite and uranium mineralization.Furthermore, as micritic kaolinite is closely associated with uranium mineralization and is formed by ore-forming fluids, the results of micritic kaolinite O and H isotopes constrain the origin and evolution of the oreforming fluids.
The thermal evolution of OM (coal and carbon debris) in the target strata increased with its burial depth during burial diagenesis.Vitrinite reflectance (Ro) of the coal beds in the mining area is 0.40-0.74%[36], suggesting that OM is semimature to weakly mature, which corresponds to the late-early diagenesis or early-middle diagenesis [53].Generally, when Ro = 0.35%, decarboxylation of OM forms organic acids and enriches hydrothermal fluid with CO, H 2 S, and CH 4 [54,55].There is an increase in the amount of organic acids with an increase in burial depth and geothermal gradient [56].During sedimentation and subsidence of the basin, compaction occurred and fluids flowed from the low-permeability coals, shales, and carbonaceous mudstones to the adjacent highpermeability sandstones, in a direction from bottom to top and from center to edge (Figure 9(a), Cartwright [57]; Xue et al. [58]).When a rich organic acid flow entered sandstones, aluminosilicate and carbonate minerals were dissolved and secondary pores were formed.Simultaneously, authigenic minerals, such as vermicular kaolinite and quartz overgrowths, also formed.The clay mineral content in the mining area and scanning electron microscopy data evidence an obvious and widespread alteration of sandstones by acidic fluid, and the replacement of vermicular kaolinite with micritic kaolinite.
The basin has undergone uplifting and shrinkage since the Miocene [59,60] and basin fluids have changed in action from seeping to permeating.Terrain elevation differences and pressures owing to tectonic compression are the major fluid-driving forces.The basin fluids generally flowed from top to bottom and from edge to center [61].Meteoric water contained O 2 and CO 2 , was strongly oxidizing and weakly acidic, and it leached the feldspars within the sandstones, thereby forming the micritic kaolinite that filled leached pores and superimposed upon the previously formed vermicular kaolinite and other grains.The feldspar grains that were replaced with micritic kaolinite are more frequently identified in the red or yellow shallow sandstones than those in the deep-gray sandstones.Superimposed micritic kaolinite is observed in all sandstone samples collected from the oxidation and mineralized zones, which suggests that orebearing sandstones underwent intense leaching by oxidizing meteoric water.

Uranium Mineralization.
Braided river delta sediments with coal-sand-coal stacks were deposited in the target strata (J 1 s and J 2 x).This rock assemblage offered conducive infiltration for phreatic water in the permeable layer (sandstone) between impermeable layers (coal and mudstone), and an interlayer oxidation zone was thus developed.
During diagenesis, secondary pores formed in the sandstones that was altered by early organic acidic fluid, which then increased permeability and created pore-space for the later long-term infiltration of meteoric water containing U and O 2 .The gas-charging effects of CO, H 2 S, and CH 4 greatly enhanced the reduction capacity of the host sandstones, which favored subsequent uranium reduction and enrichment.In addition, the humic acid and fulvic acid in the organic acidic fluid leached uranium from the surrounding rocks to form uranyl-organic complexes and preconcentrated the uranium [21,[62][63][64][65][66].
Since the Miocene, meteoric water containing U and O 2 infiltrated interlayered sandstones that were rich in pyrite and carbon debris.The abundant pyrite and carbon debris, both of which are highly effective reducers for uranium, favored the rapid reduction of U 6+ in oxidized water to U 4+ which precipitated as solid uraniferous mineral from the solution.Therefore, the sandbodies between two coal seams were ideal for forming an interlayer oxidation zone and uranium mineralization.As the oxidation of pyrite and carbon debris caused by meteoric water, sulfuric and organic acids were generated and lowed the pH of the interlayer fluid and then transformed the alkalescent interlayer fluid to neutral or weakly acidic in the redox front.This process produced micritic kaolinite that sorbed (UO 2 ) 2+ (uranyl) through cation exchange, forming uranium-kaolinite complexes, and also reduced U 6+ in the interlayer fluid to U 4+ by the pyrite and carbon debris at the acid oxidation front (Figure 9(b)).

Conclusions
Intense kaolinization is common in ore-hosting rocks and is related to the leaching of feldspar by acidic fluid.The wellformed vermicular kaolinite is attributed to hydrocarbonbearing fluid formed during shallow diagenesis, with low  ℎ values and high salinity values.In contrast, irregular micritic kaolinite is attributed to meteoric fluid, with low  ℎ values and low salinity values.
During the shallow-burial diagenetic stage of the target strata, thermal evolution of OM in the coal-bearing clastic rocks produced hydrocarbon-bearing fluid, which moved to the adjacent sandstones and enhanced their reduction capacity.Silicate minerals, such as feldspar, were leached and formed vermicular kaolinite with secondary pores, offering a favorable space for uranium mineralization.As the meteoric fluid containing U and O 2 infiltrated the interlayer sandstones, oxidation zone occurred and formed a micritic kaolinite adsorbed uranyl forming U-bearing kaolinite complex.Furthermore, uranium precipitation and enrichment occurred at the acidic oxidation front.
The results of this study confirm that intense kaolinization in sandstones of coal-bearing clastic rocks is closely associated with uranium mineralization.Therefore, to a certain extent, micritic kaolinization can be used as an effective indicator for prospecting sandstone-hosted uranium deposits in coal-bearing clastic rocks.

Figure 6 :
Figure 6: (a) LH inclusions coexisting with HA and AI inclusions occurring along healed microfractures in quartz grain.(b) Blue fluorescence color of LH and HA inclusions shown in Figure 6(a).(c) LH and HA inclusions with elliptical shapes in carbonate cement.(d) Coexistence of GI and GA inclusions in carbonate cement.(e) Yellowish-white fluorescence color of GI and GA inclusions in carbonate cement.(f) Vapor composition of GA inclusion analyzed by Raman spectrum (red cross in Figure 6(d) represents measurement point).(g) Linear group of AI inclusions occurring along healed microfracture in quartz grain.(h) Linear group of AI inclusions occurring along unhealed microfracture cross-cutting quartz overgrowth zone.(i) Random distribution of AI inclusions in quartz overgrowth zone.

Figure 8 :
Figure 8: Plot of D versus 18 O for kaolinite-forming fluid in the Mengqiguer uranium deposit.Fields for SMOW, metamorphic water, primary magmatic water, and meteoric water lines are referred to Taylor Jr.[32].

Figure 9 :
Figure 9: (a) Organic fluid containing CO, CH 4 , and H 2 S occurred from coal beds and carbonaceous mudstones during shallow-burial diagenetic stage and entered into sandstones while flowing from the basin bottom to top.(b) Meteoric water containing U and O 2 infiltrated interlayered sandstones, forming oxidation zones, white micritic kaolinization zones, and uranium orebodies.

Table 1 :
Figure2: Host strata deposited several coal-sand-coal stacks represent braided river delta sediments.Orebodies occur at downsizing sites of host sandbodies and near contact zones with underlying or overlying mudstone.Lenticular orebodies are located in sandbodies of J 2 x 3 ; roll-shaped and tabular orebodies are comparably located in sandbodies of J 1 s 1 and J 2 x 1 .Altered minerals and geochemical indexes of each geochemical zone in the Mengqiguer uranium deposit.

Table 2 :
Average percentage content of clay minerals in host rocks from each geochemical zone.

Table 3 :
Gas hydrocarbon composition in host rocks within the Mengqiguer uranium deposit.

Table 4 :
Microthermometric and salinity data of fluid inclusions in ore-bearing rocks within the Mengqiguer uranium deposit.
4.4.3.Aqueous Inclusion TrappingPressure.The coexistence of AI and LH/GI inclusions within an individual FIA is interpreted to indicate fluid immiscibility

Table 5 :
O and H isotopic compositions of extracted kaolinite from altered zones in the Mengqiguer uranium deposit.SMOW  18 O V-SMOW  18 O (H2O)V-SMOW