Evidence for Biogenic Silica Occurrence in the Lower Silurian Longmaxi Shale in Southeastern Chongqing, China

: The gas shale in the Lower Silurian Longmaxi Formation contains a considerable amount of biogenic silica. Various originated silicas in shale, derived from di ﬀ erent depositional environment, are commonly associated with di ﬀ erent degrees of organic matter enrichment, resulting in di ﬀ erent mechanical and physical properties of shale reservoirs. Thin section identiﬁcation, scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), total organic carbon (TOC) analysis, X-ray di ﬀ raction (XRD) analysis, and X-ray ﬂuorescence (XRF) spectroscopy were used to investigate the Lower Silurian Longmaxi shale from Well Yuye 1 in southeastern Chongqing, China to obtain a better understanding of the origin of silica in the Longmaxi Shale. The results show ubiquitous cryptocrystalline silicas with poorly crystalline morphology, which di ﬀ ers from that of the detrital silica, authigenic silica, and hydrothermal silica, proving that the cryptocrystalline silicas may have a biogenic origin. Major element and mineral composition analysis indicate no correlations between K 2 O / Al 2 O 3 and SiO 2 / Al 2 O 3 and between illite and SiO 2 , and negative correlations between TiO 2 and SiO 2 / Al 2 O 3 , between illite and quartz and excess Si, and between Al 2 O 3 and excess Si, and all samples being located in the area of non-hydrothermal origin in the Al-Fe-Mn diagram, excluding silicas of terrigenous detrital origin, clay mineral transformed origin, and hydrothermal origin. Moreover, the fact that almost all samples plot above the illite Si / Al line in the cross-plot of Si versus Al and the mean values of Al / (Al + Fe + Mn) and Si / (Si + Al + Fe + Ca) are close to the values of biogenic silica prove that the silicas are primarily of biogenic origin. Positive correlations between TOC and quartz and excess Si and numerous siliceous organisms are observed, indicating that the silicas are associated with siliceous organisms. The postmortem siliceous organisms underwent silica diagenesis via a dissolution-precipitation mechanism following the sequence of opal-A → opal-CT → cryptocrystalline biogenic silica as the burial depth and temperature increased.


Introduction
After the Hirnantian glaciation at the end of Ordovician, a global eustatic transgression led to deposit a group of black shales in many regions of the world [1]. In North Africa and Arabia, the Lower Silurian organic-rich shales were deposited during Rhuddanian times (Rhuddanian = early Llandovery), which was a result of transgression, and they are the main origin of Palaeozoic sourced hydrocarbons [2]. In China, the black shales are widely distributed in the Upper Yangtze region and usually exist in the Longmaxi Formation of the lower part of the Silurian Llandovery strata, which were mainly the products of large-scale anoxic events [1,3].
The Lower Silurian Longmaxi shale in southern China is characterized by widespread gas-bearing, large-area enrichment, and high yield [4]. At present, the industrial exploitation of shale gas has been successfully realized in the Longmaxi shale [5]. Therefore, the Longmaxi shale has been regarded as a target layer for shale gas exploration [1]. Dong et al. (2014) suggested that the controlling factors of shale gas enrichment and high yield in Longmaxi Formation included: (1) stable distribution of deep-water shelf facies shale; (2) high organic matter abundance and moderate thermal evolution; (3) development of pores and fractures in siliceous rich shale; (4) stable structure, favorable preservation conditions, and over-pressured formation [6].
The high-quality reservoir of Wufeng-Longmaxi Formations is mainly siliceous shale containing biogenic silica, and the high content of biogenic silica not only makes the shale brittle, but also facilitates the development of pores and fractures, providing sufficient space for shale gas enrichment [4]. Meanwhile, there is a positive correlation between the silica content and organic carbon content of the high-quality shale in the lower part of Wufeng-Longmaxi Formation, showing that the organic carbon content increases with an increase in the silica content [7]. The high-quality shale of sweet spot intervals in the Wufeng-Longmaxi Formations in the Sichuan Basin has an average silica content of about 60% and a radiolarian content of 30%, and up to 90% of the total silica is biogenic silica [5]. Therefore, the study of biogenic silica is of great significance to obtain an in-depth understanding of the shale formation environment, organic matter enrichment, and reservoir characteristics.
Silicas in shale include terrigenous detrital silica, biogenic silica, hydrothermal silica, and authigenic silica formed during clay mineral transformation [8][9][10][11]. Terrigenous detrital silica is a weathering product of terrestrial parent rocks and can be moved to marine environments by mechanical transportation [12]. Biogenic silica is derived from opaline tests of siliceous organisms, such as radiolarians [13]. The diagenesis of siliceous organisms involves a dissolution-redeposition reaction during the conversion of opal-A to opal-CT to cryptocrystalline silica [14]. Moreover, authigenic silica, which forms during clay mineral transformation, is generated from Si released during the smectite to illite (S-I) reaction [15]. The S-I reaction is smectite + Al + K → illite + Si, in which Si decreases in the clay minerals, and the release of Si results in the formation of authigenic silica [16]. In addition, crystalline silica can be precipitated from hydrothermal fluids at temperatures in the range of 50-750 • C [17]. Under hydrothermal conditions (temperature: 100-300 • C; pressure: 3 kb), silica gel crystallizes to chalcedonic spherulites [18]. At temperatures of 300-500 • C and pressures of 0.2-4 kbar, amorphous silicas, such as kieselgel and silica gel, can transform to crystalline silicas via cristobalite and keatite [19].
In previous studies, the identification of the source of silica is mainly based on the methods of total organic carbon (TOC) analysis, X-ray diffraction (XRD), thin section analysis, scanning electron microscopy (SEM), and X-ray fluorescence (XRF) spectroscopy, involving analysis of biological structure, organic matter, mineral, and major element [20][21][22][23][24][25][26][27]. First, the analysis of major element is a necessary method to determine the source of silica. Magnall et al. (2015) utilized the correlation between SiO 2 /Al 2 O 3 and K 2 O/Al 2 O 3 and TiO 2 of the siliceous radiolarian-bearing mudstones to judge whether the silica has a terrigenous detrital origin [28]. Adachi et al. (1986) and Yamamoto (1987) used the Al-Fe-Mn diagram and the Al/(Al + Fe + Mn) ratio to distinguish hydrothermal cherts and non-hydrothermal cherts by examining the effect of hydrothermal activities [9,29]. Rowe et al. (2008) proposed the cross-plot of Si versus Al to estimate the abundance of excess Si, which was the proxy for biogenic silica [30]. Ruiz-Ortiz et al. (1989) utilized the Si/(Si + Al + Fe + Ca) ratio to estimate the biogenic silica content of Subbetic radiolarites in respect of aluminosilicates and calcareous and ferruginous minerals [31]. Second, the analysis of mineral and organic matter plays an important role in identifying the occurrence of biogenic silica. The positive correlation between the TOC content and the quartz content confirms that a substantial proportion of silica in the shale have a biogenic origin [32,33]. Third, the biological structure being recognized in thin section, such as radiolarian and sponge spicule, can provide evidence for the presence of biogenic silica [34]. Although previous studies have proved the existence of biogenic silica in shales by excluding terrigenous detrital silica and hydrothermal silica, few studies have investigated how to exclude authigenic silica formed during clay mineral transformation. Moreover, biogenic silica has unique crystal morphology, but there are few studies on the development mechanism of the unique crystal morphology of biogenic silica.
In this study, crystal morphology, elemental, and mineral analyses of silica are conducted to exclude the possibility that cryptocrystalline silica in the Longmaxi shales is terrigenous detrital silica, hydrothermal silica, and authigenic silica formed during clay mineral transformation. Organic matter analysis and the existence of siliceous organisms confirm that the formation of cryptocrystalline silica in the Longmaxi shales is associated with siliceous organisms, proving the existence of biological silica. In addition, we establish a formation model of cryptocrystalline biogenic silica.

Geological Setting and Sampling
Chongqing, which is located in the margin of the Sichuan Basin, China, belongs to the southeastern part of the Upper Yangtze Plate ( Figure 1A). The structure in this region is characterized by high uplift and strong compression [35]. The exposed strata include Cambrian, Ordovician, Silurian, Devonian, and Permian, and the other strata are thin or missing [36]. The Lower Silurian Longmaxi shale in southeastern Chongqing was formed in a deep-water shelf environment [37]. The black high-over mature organic-rich shales are widely distributed, very thick, strongly deformed, and shallowly buried [38].
All samples were collected from Well Yuye 1 ( Figure 1B). This well is located in Lianhu Town, Pengshui County, Chongqing Province, in the eastern part of Qiyao Mountain and is situated structurally at the axis of the Guochangba anticline [39,40]. The drilling depth of Well Yuye 1 was 325.48 m, and a continuous core of the main black shale section of the Longmaxi formation was obtained [35]. In this study, 24 samples were collected from depths ranging from 170.9 m to 316.0 m ( Figure 2). The samples were analyzed to investigate the origin of silica in the Lower Silurian Longmaxi shale.
calcareous and ferruginous minerals [31]. Second, the analysis of mineral and organic matter plays an important role in identifying the occurrence of biogenic silica. The positive correlation between the TOC content and the quartz content confirms that a substantial proportion of silica in the shale have a biogenic origin [32,33]. Third, the biological structure being recognized in thin section, such as radiolarian and sponge spicule, can provide evidence for the presence of biogenic silica [34]. Although previous studies have proved the existence of biogenic silica in shales by excluding terrigenous detrital silica and hydrothermal silica, few studies have investigated how to exclude authigenic silica formed during clay mineral transformation. Moreover, biogenic silica has unique crystal morphology, but there are few studies on the development mechanism of the unique crystal morphology of biogenic silica.
In this study, crystal morphology, elemental, and mineral analyses of silica are conducted to exclude the possibility that cryptocrystalline silica in the Longmaxi shales is terrigenous detrital silica, hydrothermal silica, and authigenic silica formed during clay mineral transformation. Organic matter analysis and the existence of siliceous organisms confirm that the formation of cryptocrystalline silica in the Longmaxi shales is associated with siliceous organisms, proving the existence of biological silica. In addition, we establish a formation model of cryptocrystalline biogenic silica.

Geological Setting and Sampling
Chongqing, which is located in the margin of the Sichuan Basin, China, belongs to the southeastern part of the Upper Yangtze Plate ( Figure 1A). The structure in this region is characterized by high uplift and strong compression [35]. The exposed strata include Cambrian, Ordovician, Silurian, Devonian, and Permian, and the other strata are thin or missing [36]. The Lower Silurian Longmaxi shale in southeastern Chongqing was formed in a deep-water shelf environment [37]. The black high-over mature organic-rich shales are widely distributed, very thick, strongly deformed, and shallowly buried [38].
All samples were collected from Well Yuye 1 ( Figure 1B). This well is located in Lianhu Town, Pengshui County, Chongqing Province, in the eastern part of Qiyao Mountain and is situated structurally at the axis of the Guochangba anticline [39,40]. The drilling depth of Well Yuye 1 was 325.48 m, and a continuous core of the main black shale section of the Longmaxi formation was obtained [35]. In this study, 24 samples were collected from depths ranging from 170.9 m to 316.0 m ( Figure 2). The samples were analyzed to investigate the origin of silica in the Lower Silurian Longmaxi shale.

Analytical Methods
Thin section analysis, SEM, energy dispersive spectroscopy (EDS), TOC content analysis, XRD analysis of the whole rock and clay minerals, XRF spectroscopy were performed on the 24 shale samples. The TOC, mineralogical composition, and major element analyses were conducted in the Analytical Laboratory of the Beijing Research Institute of Uranium Geology, Beijing, China. SEM and EDS were performed in the State Key Laboratory of Oil and Gas Resources and Exploration at the China University of Petroleum (Beijing), Beijing, China.

TOC
The TOC contents were determined using a CS580A carbon and sulfur analyzer and following the Chinese National Environmental Protection Standard HJ 695 (2014) [42]. First, the samples were ground to a powder, with a particle size of less than 0.097 mm (160 mesh); 0.05 g of the powdered samples were weighed and placed into a quartz cup. Second, a phosphoric acid solution (φ = 5%) was slowly added to the quartz cup until the bubbling stopped to remove the carbonate. Finally, the quartz cup was placed into the TOC analyzer and was heated to above 680 using a carrier gas rich in oxygen. The organic carbon in the samples was oxidized to CO2. Subsequently, the TOC content was detected by an infrared detector. The standard deviation was less than 3%.

Analytical Methods
Thin section analysis, SEM, energy dispersive spectroscopy (EDS), TOC content analysis, XRD analysis of the whole rock and clay minerals, XRF spectroscopy were performed on the 24 shale samples. The TOC, mineralogical composition, and major element analyses were conducted in the Analytical Laboratory of the Beijing Research Institute of Uranium Geology, Beijing, China. SEM and EDS were performed in the State Key Laboratory of Oil and Gas Resources and Exploration at the China University of Petroleum (Beijing), Beijing, China.

TOC
The TOC contents were determined using a CS580A carbon and sulfur analyzer and following the Chinese National Environmental Protection Standard HJ 695 (2014) [42]. First, the samples were ground to a powder, with a particle size of less than 0.097 mm (160 mesh); 0.05 g of the powdered samples were weighed and placed into a quartz cup. Second, a phosphoric acid solution (ϕ = 5%) was slowly added to the quartz cup until the bubbling stopped to remove the carbonate. Finally, the quartz cup was placed into the TOC analyzer and was heated to above 680 using a carrier gas rich in oxygen. The organic carbon in the samples was oxidized to CO 2 . Subsequently, the TOC content was detected by an infrared detector. The standard deviation was less than 3%.

Mineralogical Composition
The XRD analysis of the whole rock and clay minerals was carried out using a Panalytical X'Pert PRO X-ray diffractometer with a Cu X-ray target (40 kV voltage, 40 mA current). The operation and calculation were in accordance with the Chinese Oil and Gas Industry Standard SY/T 5163 (2018) [43].

•
Whole rock XRD analysis.

•
The samples were crushed to a powder with a particle size less than 40 µm, and the samples were measured in an angular range of 5-70 • 2θ to obtain the XRD patterns. The total clay mineral content and non-clay mineral content were determined using quantitative analysis.

•
Clay mineral XRD analysis. • First, the samples were ground to a particle size of less than 1 mm, and powdered samples were dispersed ultrasonically in distilled water. A suspension with a particle size of less than 2 µm was extracted and was centrifugally separated to cause the clay particles to settle. After drying, the clay samples were made into oriented slices, which were analyzed in an angular range of 3-30 • 2θ to obtain the XRD patterns. Quantitative analysis was used to obtain the relative contents of different clay minerals.

Major Elements
The major element contents were detected by an Axios-mAX wavelength dispersive X-ray fluorescence spectrometer, following the Chinese National Standard GB/T 14506.28 (2010) [44]. First, the samples were ground to a powder with a particle size of was less than 74 µm. Then, 0.7 g of the powdered sample was mixed with anhydrous lithium tetraborate (Li 2 B 4 O 7 ), lithium fluoride (LiF), ammonium nitrate (NH 4 NO 3 ) and lithium bromide (LiBr). The mass ratio of the sample to solvent was 1:8. Second, the mixture was melted at 1150-1250 • C to creat a glass disc. Finally, the major elements were determined on the glass disc using an XRF spectrometer. The analytical precision was better than 1%.

SEM and EDS
The experiment was performed using the HITACHI SU8010 cold field-emission scanning electron microscopy (FE-SEM) to observe the morphology of the silica, and an IXRF X-ray spectrometer was used to analyze the chemical composition. One small sample of about 1 cm 3 was taken from a large sample to obtain a natural surface. The surface was sprayed with gold conductive film with an appropriate thickness before the analysis [45]. The composition of the target spot or micro-area was analyzed by EDS.

Organic Matter Characteristics
The TOC contents of the 24 samples of the Longmaxi shales are presented in Table 1. The TOC contents range from 1.05% to 2.57%, with an average of 1.64%. Additionally, the vitrinite reflectance (Ro) value is in the range of 1.62%−2.26%, with an average of 2.04% [35], indicating that the shales of Longmaxi Formation belong to high-over mature shales.
According to the results of XRD analysis, the lithofacies types of shales are divided by ternary diagram of mineral compositions [46]. The classification results show that the Longmaxi shales are mainly including siliceous shale and clayey-siliceous mixed shale (Figure 3).

Major Elements
The contents of the major element oxides of the 24 samples of the Longmaxi shales are presented in Table 2, including the loss on ignition (LOI). The composition of the major element oxides is consistent with that of the whole rock minerals. SiO 2 is the most abundant of all oxides, ranging from 47.29% to 66.84% (average: 62.74%), the Al 2 O 3 content ranges from 10.95% to 16.6% (average: 13.47%), and the CaO content ranges from 1.16% to 8.72% (average: 2.48%). The ternary plot of SiO 2 -Al 2 O 3 -CaO shows that the Longmaxi shales are enriched in SiO 2 , whereas Al 2 O 3 and CaO have relatively low levels ( Figure 4). TFe 2 O 3 represents total iron, including Fe 2 O 3 and FeO, and its content ranges from 3.6% to 6.02% (average: 4.82%). The average MgO content is 2.34% (range: 1.5%-5.0%), the average Na 2 O content is 1.33% (range: 0.73-1.69%), and the average K 2 O content is 3.69% (range: 3.21-4.63%). In addition, the average contents of the other major element oxides are less than 1%, i.e., TiO 2 (0.7%), P 2 O 5 (0.12%), and MnO (0.04%).

Crystal Morphology of Silica
The SEM observations and EDS analysis of the samples of the Longmaxi shale indicate a considerable amount of cryptocrystalline silicas with inferior crystallinity and irregular shape, and some crystals even have a high concentration of C element ( Figure 5). It is obviously different in the morphology from terrigenous detrital silica (with subrounded, subangular, and angular outlines), the authigenic silica formed during smectite-illite transformation (granular silica, grain size is <1 µm; sheet-like silica, thickness is <100 nm), and the hydrothermal silica (trigonal structure) [11,12,47].
Natural hydrous silicas include opal-A and opal-CT [48]. Opal-A is near amorphous and highly disordered, consisting primarily of radiolarian and sponge spicules; Opal-CT is composed of disordered α-cristobalite and α-tridymite and formed from biogenic opal-A via a solution-precipitation mechanism during diagenesis; At high temperatures, opal-CT transforms into cryptocrystalline silica through a second solution-precipitation process [14,[48][49][50]. Therefore, silicas originating from siliceous marine organisms have poorly crystalline morphology, and some have the typical morphology of siliceous tests. Therefore, the crystal morphology is regarded as a suitable method to identify biogenic silica [20].
Owing to the poorly crystalline morphology of the observed silicas, which is different from that of the detrital silica, authigenic silica, and hydrothermal silica, it is speculated that biogenic silicas are present in the Longmaxi shale.

Assessment of Detrital Input
In marine sediments, aluminum and titanium are nearly terrigenous [51]. Al always represents the aluminosilicate mineral in most sediments [52]. Moreover, Ti usually occurs in clay minerals and

Assessment of Detrital Input
In marine sediments, aluminum and titanium are nearly terrigenous [51]. Al always represents the aluminosilicate mineral in most sediments [52]. Moreover, Ti usually occurs in clay minerals and heavy minerals [23,53]. During weathering and transportation, the relative abundances of Si, Al, and Ti typically remain unchanged [8].
Normalization of the elements to Al is common practice in the assessment of detrital input [28,54]. The uniform K 2 O/Al 2 O 3 ratios of the Longmaxi shales are consistent with terrestrial muscovite, whose molar proportion of K 2 O/Al 2 O 3 is 0.29 [28]. However, there is no correlation between K 2 O/Al 2 O 3 and SiO 2 /Al 2 O 3 ( Figure 6A). In terms of TiO 2 , there is a negative correlation with SiO 2 /Al 2 O 3 ( Figure 6B). These results provide evidence that a considerable proportion of silica in the Longmaxi shale is nondetrital input [28].

Assessment of Clay Mineral Transformation
The excess Si represents the amount of SiO 2 that cannot be explained by the aluminosilicate phase [24]. The excess Si content of the samples (i.e., above normal detrital background) is calculated as follows [24,25,32]: Si excess = Si sample − Al sample × (Si/Al) background (1) The (Si/Al) background is the ratio between Si and Al in the detrital material. The value of the (Si/Al) background is used for the average shale; it is typically set at 3.11 [24][25][26]32,55]. Based on Equation (1), a maximum of 11% of the SiO 2 in the Longmaxi shales cannot be explained by the aluminosilicate phase. The excess Si content ranges from 0.41-10.88%, with an average of 7.16% ( Table 2).
The smectite to illite reaction causes the release of Si into solution, and the Si released in the illitization of smectite is partly precipitated to form authigenic silica [15]. Van According to Equation (2), when 787.2 g illite is generated, 197.7 g SiO 2 is released during the smectite to illite reaction. As a result, the content of authigenic silica formed during clay mineral transformation should be positively correlated with the illite content. However, the illite content is uncorrelated with the SiO 2 content ( Figure 7A) and negatively correlated with the quartz content ( Figure 7B) and the excess Si content ( Figure 7C). In addition, there is a negative correlation between the Al 2 O 3 content and excess Si content ( Figure 7D). These results indicate that the Si released during illitization of smectite [15] is not the source of the authigenic silica [22], demonstrating that a substantial proportion of the silica in the Longmaxi shale is not derived from clay mineral transformation.

Assessment of Hydrothermal Alteration
Previous research has proved that Fe and Mn originate from the hydrothermal solutions of submarine volcanism, and Al is nearly terrigenous [51]. The Al concentration of the Longmaxi shale ranges from 5.8% to 8.79% (average: 7.13%), the Fe concentration ranges from 2.52% to 4.21% (average: 3.37%), and the Mn concentration ranges from 0.02% to 0.12% (average: 0.03%). The element concentrations of Al, Fe, and Mn are calculated from the contents of Al 2 O 3 , TFe 2 O 3 , and MnO, respectively. Adachi et al. (1986) proposed the Al-Fe-Mn diagram, which shows the difference in chemical composition between hydrothermal and non-hydrothermal siliceous rock [29]. The influence of hydrothermal activities on the chemical composition of the Longmaxi shales can be identified using this diagram [9]. In the Al-Fe-Mn diagram, all the samples of the Longmaxi shales have extremely low Mn concentrations and are located in the non-hydrothermal origin area, indicating that hydrothermal activities had no effect on the Longmaxi shales during sedimentation and diagenesis and a large proportion of the silicas was not derived from hydrothermal dissolution (Figure 8).

Biogenic Origin of Silica
The calculation results of the major oxides of the Longmaxi shales indicate that all samples have high SiO 2 /Al 2 O 3 ratios (Table 2), ranging from 3.57 to 5.43, with an average of 4.69. It is generally believed that the ratio of SiO 2 /Al 2 O 3 in the average crust is about 3, and it has been suggested that if ratio of SiO 2 /Al 2 O 3 is greater than 3, the portion of SiO 2 exceeding Al 2 O 3 can be regarded as biogenic silica [51]. Moreover, in the cross-plot of Si versus Al, almost all the samples of the Longmaxi shales are located in the area above the illite Si/Al line (Figure 9), showing that they have excess Si, which is indicative of a considerable proportion of biogenic silica [30]. In addition, the Al/(Al + Fe + Mn) ratio is typically used to estimate the influence of hydrothermal activities on marine sediments, and the ratio increases with a decrease in the hydrothermal input [9,29,57]. The Al/(Al + Fe + Mn) ratio of hydrothermal precipitates is 0.01, and the value of biogenic cherts is 0.60 [29]. The Al/(Al + Fe + Mn) ratio of the Longmaxi shales ranges from 0.57 to 0.74 (average: 0.68) (Table 2, Figure 10), which is close to the value of biogenic cherts, indicating that the silicas in the Longmaxi shales are primarily of biogenic origin. Furthermore, the Si/(Si + Al + Fe + Ca) ratio is used to estimate the biogenic silica content of shales [31,53]. The value of Si/(Si + Al + Fe + Ca) for 30 samples of Middle and Late Jurassic radiolarites of the Subbetic in southern Spain ranges from 0.661 to 0.90 (average: 0.784) (Figure 10) [31]. Meanwhile, the Si/(Si + Al + Fe + Ca) ratio of the Longmaxi shales ranges from 0.58 to 0.75 (average: 0.70) (Table 2, Figure 10), which is close to the values of the Subbetic radiolarites, demonstrating that the silicas are mainly of biogenic origin. All of the element concentrations are calculated from the oxide contents of the major elements.

Relationship between Mineral and TOC
The majority of organic carbon in marine shales is derived from phytoplankton [58]. As the basis of the marine food chain, the phytoplankton converts CO 2 into organic matter through photosynthesis [59]. Silicon is an important nutrient in phytoplankton growth, and phytoplankton, such as diatoms, has a high growth rate in an environment with sufficient silicon [60]. The nutrient silicon plays an important role in determining the proportion of inorganic carbon and organic carbon [61]. Primary production controls organic carbon enrichment in the ocean [62].
In addition, the lipid content of modern radiolarians is high and is similar to that of diatoms [63]. Moreover, radiolarians often co-exist with a large number of symbiotic algae, which supply nutrients to their hosts by translocating the photosynthates [64,65]. There is a positive correlation between the radiolarian abundance and the TOC content of Changhsingian radiolarian fauna [66]. Therefore, siliceous zooplankton, especially radiolarians, also contributes to the enrichment of organic carbon in the shale.
Furthermore, many previous studies have focused on the vertical distribution of plankton in the ocean. Phytoplankton occurs at depths ranging from ∼40 to 100 m in the Southern Ocean [67]. Radiolarians are mainly found at depths of 40-300 m in the Japan Sea [68]. The depth range of radiolarians is generally greater than that of phytoplankton. As a result, the radiolarians reach the seafloor faster after death than phytoplankton, shortening the residence time in the water during settlement and reducing the decomposition of organic matter, which is conducive to the enrichment of organic matter [63].
The quartz contents of the Longmaxi shale are 25.2-46.5%, and the TOC contents are 1.05-2.57% (Table 1). Moreover, the excess Si contents are 0.41-10.88% ( Table 2). The relationship between the TOC content and the quartz and excess Si contents shows that the TOC content increases with an increase in the quartz and excess Si content ( Figure 11). The positive correlation between the TOC content and the quartz and excess Si contents in the Longmaxi shales is the result of the accumulation of abundant siliceous organisms on the seabed, and then silica and organic matter are buried and deposited together [69]. This result provides evidence that the silica in the shale may be derived from siliceous organisms in the ocean.

The Link between Ancient Organism and Biogenic Silica
Thin section analysis of the samples of the Longmaxi shales from Well Yuye 1 indicates a considerable amount of ancient organism microfossils, primarily radiolarians and sponge spicules ( Figure 12). The ancient organism microfossils provide evidence for the presence of biogenic silica [34,58,70]. The radiolarians are spherical or ellipsoidal, with a particle size of 40-260 µm ( Figure 12A,B). The chemical composition of the radiolarian test is SiO 2 , which belongs to the siliceous test [71]. Some radiolarian tests have a high degree of crystallization, and the tests are dissolved and have holes ( Figure 12A). In addition, some radiolarians are well preserved and have circular tests ( Figure 12B). Moreover, the sponge spicules are in the shape of monaxons, with a length of 70-700 µm. One end of the sponge spicule is thick, and the other end is sharp ( Figure 12C,D). Under natural conditions, siliceous sponges are catalyzed by silicatein to synthesize spicules (biogenic silica), and the Si concentration of the siliceous sponge spicules is 200 times higher than the total concentration of the other elements [72].
Silicon is nutrient element required for the growth of siliceous marine organisms, such as radiolarians and sponge spicules, and siliceous organisms flourish in silicon-rich water [27]. During growth and development, siliceous marine organisms generate siliceous tests/skeletons by absorbing dissolved silica from the water, then they are eventually buried by sediments in the postmortem stage [58]. Most of the biogenic silicas are dissolved again, and only a small fraction remains in the sediments; the burial/production ratio of biogenic silica is only ∼3% [73].
A large number of siliceous organisms are present in the Longmaxi shales, such as radiolarians and sponge spicules. High productivity and settling velocity of siliceous organisms are beneficial to their preservation and the subsequent diagenetic silica conversion after the dissolution of the siliceous tests/skeletons [74]. Therefore, the cryptocrystalline silicas observed in the samples of the Longmaxi shales are associated with siliceous organisms.

Diagenetic Transformation Mechanism of Biogenic Silica
Cryptocrystalline silicas with inferior crystallinity and irregular shape are ubiquitous in the Longmaxi shales ( Figure 5). Our comprehensive analysis indicates that the cryptocrystalline silicas are primarily biogenic origin.
During the Early Silurian period, siliceous marine organisms, such as radiolarians and sponge spicules, flourished in the southeastern Chongqing area ( Figure 12). Dissolved silica was extracted from the seawater by the siliceous organisms and was used as a nutrient to generate opaline tests/skeletons and accomplish biological fixation [75]. After the death of the siliceous organisms, the silica was recycled by dissolution of the opaline tests/skeletons [74]. The buried biogenic opal-A experienced continuous dissolution, and the dissolved silica diffused into the seawater [76].
After burial, in addition to the dissolution of partial siliceous organisms, some underwent diagenesis. The diagenetic transformation of the biogenic silica in the Longmaxi shales followed a sequence with increasing depth and temperature: unstable amorphous biogenic opal-A → disordered metastable opal-CT → stable cryptocrystalline biogenic silica [14,74,77]. The dissolved silica of the biogenic opal-A supplied a silica source to form opal-CT through precipitation [74]. In the initial stage, individual opal-CT blades with irregular serrated edges crossed each other, and incipient lepisheres (spherical aggregates) were generated when numerous opal-CT blades were combined; Subsequently, groups of opal-CT blades were regularly intergrown to form complete lepispheres; Finally, several complete lepispheres merged to generate composite lepispheres [78]. Moreover, cryptocrystalline biogenic silica formed from opal-CT via a dissolution-reprecipitation mechanism [49,79].
On the basis of summarizing the schematic diagrams of the cycle and diagenetic transformation of silicas established by the predecessors [74,75,77], we have established a formation model of cryptocrystalline biogenic silica in Longmaxi shale ( Figure 13). Although the previous schematic diagrams of the cycle and diagenetic transformation of silicas are relatively perfect, the morphology of products at each stage, such as biogenic opal-A, opal-CT, and cryptocrystalline biogenic silica, are not shown in the schematic diagrams. In addition, the opal-CT has different structures at different growth stages, including individual opal-CT blades, incipient lepisheres, complete lepispheres, and composite lepispheres [78], which are not shown in the previous schematic diagrams. Moreover, there are few studies on the morphology of cryptocrystalline biogenic silica, and people have no intuitive understanding of its morphology. Based on the deficiency of previous schematic diagrams, we have systematically established a formation model of cryptocrystalline biogenic silica on the basis of the comprehensive previous research results, including biogenic silica cycle and diagenetic transformation. The model diagram presents the morphology of products at each stage, and the cryptocrystalline biogenic silica in the Longmaxi shale is integrated into the model diagram as the final product of diagenetic transformation, which makes the model diagram more complete. The diagenetic transformation mechanism of biogenic silica in the Longmaxi shales is shown in Figure 13.  [80]. The individual opal-CT blades and incipient lepisphere are from Riech and von Rad (1979) [74]. The complete lepisphere is from von Rad et al. (1978) [77]. The composite lepispheres are from Flörke et al. (1976) [78]. The cycle and diagenetic transformation of silicas are modified from Heath (1974), von Rad et al. (1978), and Riech and von Rad (1979) [74,75,77]. The cryptocrystalline biogenic silicas are the sketches of the observed silicas in the Longmaxi shales.

Conclusions
Analyses of crystal morphology, major elements, mineral composition, TOC content, and ancient organisms of the Longmaxi shales from Well Yuye 1 were conducted. We obtained the following conclusions: • Numerous cryptocrystalline silicas with poorly crystalline morphology, which differed from that of detrital silica, authigenic silica, and hydrothermal silica, proved that there may be biogenic silicas occurrence in the Longmaxi shale.

•
The lack of correlation between K 2 O/Al 2 O 3 and SiO 2 /Al 2 O 3 and a negative correlation between TiO 2 and SiO 2 /Al 2 O 3 excludes a terrigenous detrital origin of the silicas. Moreover, the lack of correlation between illite and SiO 2 , and negative correlations between illite and quartz and excess Si and between Al 2 O 3 and excess Si proved that the silicas were not derived from clay mineral transformation. In addition, the fact that all samples were located in the non-hydrothermal origin area in the Al-Fe-Mn diagram excludes a hydrothermal origin of the silicas. • Almost all samples plotted above the illite Si/Al line in the cross-plot of Si versus Al, suggesting the existence of biogenic silica. Furthermore, the mean values of Al/(Al + Fe + Mn) and Si/(Si + Al + Fe + Ca) were 0.68 and 0.70, respectively, which were close to the values of biogenic silica. These pieces of evidence prove that the silicas are primarily of biogenic origin.

•
There was a considerable amount of ancient organism microfossils in the Longmaxi shales and positive correlations between TOC and quartz and excess Si, indicating that silicas are associated with siliceous organisms.

•
The postmortem siliceous organisms of the Early Silurian in the southeastern Chongqing area underwent a diagenetic process of opal-A → opal-CT → cryptocrystalline biogenic silica after burial. This sequence proceeded via a dissolution-precipitation mechanism as the burial depth and temperature increased.