The Kaolinite Crystallinity and Influence Factors of Coal-Measure Kaolinite Rock from Datong Coalfield, China

Abstract

In order to ascertain the kaolinite crystallinity of Carboniferous Permian coal-measure kaolinite rocks, seven groups of fresh samples were collected from below the ground in the Xiaoyu mine, Datong coalfield. Microscopy, X-ray diffraction (XRD), differential thermal analysis (DTA), infrared (IR) spectroscopy and X-ray fluorescence (XRF) spectrometry methods were applied to the samples. The petrographic analysis results show that the kaolinite rocks are characterized as compact, phaneritic, clastic, sand-bearing, sandy and silty types; the kaolinite content in the Shanxi formation and upper Taiyuan formations was more than 95%, while it was 60–90% in the middle and lower Taiyuan formations. Based on the Hinckley index and the features of XRD, DTA and IR of kaolinites, crystallinity was classified as having three grades: ordered, slightly disordered and disordered. The kaolinites’ SiO₂/Al₂O₃ molar ratio was about 1.9–5.7, with a chemical index of alteration (CIA) of about 95.4–99.5. This research suggests that the kaolinite crystallinity correlates positively to its clay mineral content, purity and particle size, which are also related to the SiO₂/Al₂O₃ molar ratio and CIA. The original sedimentary environment and weathering have a direct influence on kaolinite crystallinity, and the existence of organic matter is conducive to the stable existence of kaolinite. The study results have significance for the extraction and utilization of coal-measure kaolinite and the development of kaolinite crystallography and mineralogy.

1. Introduction

Coal-measure kaolinite rocks in China occur in all major coal deposits from the late Paleozoic to Cenozoic eras, but especially in the Carboniferous Permian of the late Paleozoic [1]. The Datong coalfield is one of the most important coal fields in Shanxi, China, where coal-measure kaolinite rocks are of superior thickness, quality, reserves and utilization value. The Datong coalfield is located in the north of Shanxi Province, China, in the central northern region of the North China Craton (NCC) (Figure 1). Here, the Carboniferous Permian coal-measure kaolinite rocks are referred to as “Datong black sandstone” for their nearly mono-mineralic composition with low impurity content [1,2,3,4]. At Datong, coal-measure kaolinite is commonly developed in the roof, floor and coal gangue of the No.4 coal seam of the Shanxi Formation, as well as in the No.2, No.3, No.5, No.8 and No.9 coal seams of the Taiyuan formation [2].

In recent years, coal-measure kaolinite rocks and their application have garnered increasing research attention [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19], particularly in relation to mineral content and composition, rock type, formation conditions, provenance and geochemistry using diverse methods [1,2,3,4,5,6,7,8,9,10], and kaolinite in mineral synthesis has been used in various applications [14,15,16,17]. In-situ alterations in mafic pyroclastic material may produce kaolinite rock, whereas volcanic tuff is transformed into kaolinite rock by acids from peatification of sphagnum overburdening [5,18]. The petrogenesis of kaolinite rocks is affected by their detrital or volcanic sedimentary origins, as well as their biological interactions and organic matter that promote the formation of clay minerals [5,6,11,18]. The crystallinity of the so-called “purple gangue” kaolinite found at Datong increases from a poorly ordered material with low crystallinity to a well-ordered material with high crystallinity upon increasing interaction with high-temperature geothermal fluids [18,19,20].

Earlier mineralogical and geochemical characterization of the coal-measure kaolinite from Datong indicated an acid plutonic (granitic) or volcanic (rhyolitic) precursor [2,3,4,20], with potential provenance from the Yinshan uplift orogenic belt to the north of NCC (Figure 1), which features shallow surface water with a weak oxidation condition diagenetic environment, formed by protolith in situ leaching and alteration [21,22]. The mineral crystallinity of kaolinite is associated with its microstructure, whereby aphanitic kaolinite reveals a higher crystallization index than clastic granular and worm-like kaolinite [23]. In addition to the primary sedimentary and secondary weathered eluvial, the Datong kaolinite diagenesis is related to organic acid leaching in epigenesis diagenesis, kaolinite recrystallization, crystallinity and other factors [2,3,4,20,22].

As we know, the intercalation composites formed by kaolinite have shown unique properties in mechanics, adsorption, electricity and catalysis, and have broad application prospects in the fields of functional ceramics, nonlinear optical materials, polymer matrix composites and electrorheological fluids [14,17,24]. Additionally, crystallinity has an important effect on the intercalation of kaolinite, and the disordered structure is not conducive to intercalation. Therefore, the study of kaolinite crystallinity would benefit the development of intercalated kaolinites for industrial applications [14,25].

2. Geological Background

The Datong coal-bearing basin was formed under regional dynamics of early continental–continental subduction collision between the northern margins of the NCC and the southern Mongolian macroblock in Siberia in the early Carboniferous era. During that collisional orogenic process, a geotectogene was formed toward the back edge of the subduction zone of the NCC [26,27]. Therefore, depression deposits were formed in the Datong in the mid-northern part of NCC (Figure 1), which began accepting the deposition of the late Paleozoic era. As part of the ate Paleozoic era giant coal-accumulating basin in northern China, the Datong coal-bearing basin has undergone weathering and denudation for about 140 Ma since its uplift in the late Ordovician era. The strata of the middle and late Ordovician, Silurian, Devonian, and Lower Carboniferous deposits were interrupted until their re-deposition in the early Middle Carboniferous era [28,29,30].

The fold structure in the coalfield is undeveloped, and only some small asymmetric synclinal structures exist; the main synclinal axis is NE in the south and SN in the north [29]. The boundary is marked by four tectonic features (Figure 1): the Hongtaoshan anticline (extending NW 50°) to the southeast, the Kouquan–Emaokou thrusting fault belt to the east, the Qingciyao fault to the northeast and the Maihutu–Weiyuanbao fault to the northwest.

Neopaleozoic rocks are commonly composed of conglomerate, sandstone, siltstone, muddy shale, sandy shale, shale, mudstone, claystone, marl and coal seams. There are extensive alkaline mafic–ultramafic lamprophyre and carbonatite dikes intruding into the Permo-Carboniferous coal strata [27]. The K/Ar and Rb/Sr geochronological ages of the lamprophyre biotite are middle–late Triassic (242 ± 53 Ma, 229 ± 11 Ma and 224–198 Ma) [27,28]. The intrusion age of the carbonatite dyke associated with the lamprophyre is lower Cretaceous (132.9 Ma) [31], and the K/Ar age of the whole rock is 131–133 Ma. Additionally, almost 3 km² of early Cretaceous basaltic andesite covers the Zuoyun Formation strata nearby the Zuoyun–Jiugaoshan district, along the western side of the basin [2,32].

3. Materials and Methods

Based on the 53 groups of borehole data analysis in the Xiaoyu mine, at least nine layers of kaolinite-bearing claystone or shale were confirmed. Seven layers of underground fresh kaolinite rocks samples were collected from the Xiaoyu Mine, including the samples from the No.4 coal floor (S4F) of the lower part of Shanxi Formation, No.2 coal floor (C2F), No.3 coal roof (C3R), No.5 coal floor (C5F1 and C5F2), No.5 coal gangue (C5G) and No.8 coal floor (C8F) of the Taiyuan Formation (Figure 1). The S4F was stably distributed throughout the mine area, with average thickness of 1.20 m. The C2F and C3R were both from the upper Taiyuan formation and were stably distributed in the mine area, in which the C2F had an average thickness of 1.00 m and C3R was about 2.00 m. The thickness of C5G was generally 0.10–0.80 m with an average of 0.30 m. The C5FS included two parts: No.5 coal direct floor of C5F1 and stable floor of C5F2; thickness varied greatly from 1.00 to 3.00 m. The C8F is unstably distributed with an average thickness of 2.00–5.00 m, and the outcrops are visible on the ground (Figure 2).

To comprehensively study the petrology, mineralogy and crystallography characteristics of Datong kaolinite rocks, the samples were examined under a polarizing microscope (Leica DM2700P) in Geology Experiment Center of Taiyuan University of Technology (Taiyuan, China) and analyzed by X-ray diffraction analysis (XRD), differential thermal analysis and thermal gravity (DTA–TG), infrared spectroscopic analysis (IR) and X-ray fluorescence (XRF). The sample XRD, DTA–TG and IR analyses were completed at the Shanxi Coal Chemistry Institute of the Chinese Sciences Academy. The element content test was completed at the Nuclear Industry Geological Analysis and Testing Center in Beijing.

The samples were analyzed using a Rikagu D/Max-RA XRD (Rigaku, Tokyo, Japan) with Cu Kα radiation at 40 kV and 30 mA from 3° to 70° (2θ) with step increments of 0.1° and a scanning speed of 4° (2θ) per minute. Samples were protected under an argon atmosphere in a homemade container. The whole-rock samples were powdered using a ball-milling machine and then measured to determine the mineralogy characteristics. X-ray diffractograms of the samples were subjected to quantitative mineralogical analysis using Jade 6.5 software, and the types and forms of texture defects of kaolinite were studied according to the Hinkley index (HI), which is proposed by Hinkley (1963) for measuring the quantitative crystallinity of kaolinite [33].

DTA–TG analyses were performed on the kaolinite rocks sample using a Rigaku TAS 100 E instrument (Rigaku, Tokyo, Japan). The DTA–TG curves were obtained from 10 mg of powdered sample in a Pt sample holder, which was heated at an average rate of 10 °C/min with an alumina reference. IR spectroscopic analysis [34] was performed on pressed pellets of powdered clay samples (<2 µm) mixed with KBr, while scans were made at a wave-number resolution of 4 cm⁻¹.

Samples were crushed and ground to less than 200 mesh (75 μm) for element analyses. X-ray fluorescence (XRF) spectrometry was used to determine the major oxides, including silicon oxide (SiO₂), aluminum oxide (Al₂O₃), calcium oxide (CaO), potassium oxide (K₂O), sodium oxide (Na₂O), iron oxide (Fe₂O₃), magnesium oxide (MgO), manganese oxide (MnO), titanium oxide (TiO₂) and phosphoric oxide (P₂O₅) in the powdered samples.

4. Results and Discussion

4.1. Petrography and Mineralogy

A total of 32 thin sections were obtained from the aforementioned seven layers of kaolinite rocks to identify the petrographic features using a microscope of Leica 2700 (

Table 1. The HI and mineral composition of coal-measure kaolinite rock in Xiaoyu mine.

Samples	Lithology	Thin Section	HI	Minerals
S4F	Compact kaolinite rock	4	1.12	Kaolinite, quartz
C2F	Phaneritic kaolinite rock	3	1.13	Kaolinite
C3R	Clastic kaolinite rock	5	1.29	Kaolinite, diaspore, boehmite, anatase
C5G	Clastic kaolinite rock	3	0.81	Kaolinite, quartz, pyrite
C5F1	Sand-bearing kaolinite rock	4	0.82	Kaolinite, quartz
C5F2	Sandy kaolinite rock	6	0.68	Kaolinite, quartz
C8F	Silty kaolinite rock	7	0.65	Kaolinite, quartz, illite

). According to the observation results, the types of kaolinite rocks in the Xiaoyu mine include compact, phaneritic, clastic, sand-bearing, sandy and silty kaolinite rocks. The first three are named according to the texture characteristics of kaolinite, while the last three are named by the texture features of the rocks. The petrological characteristics of kaolinite rocks were described from top to bottom as below.

The petrology type of S4F was brownish-black compact kaolinite rock with a conchoidal fracture, mainly composed of a scaly texture and flaky kaolinite with a content of more than 95 vol.%, as well as a few silt-grade quartz fragments and carbonaceous debris (Figure 3a(−)). The flaky kaolinites were pale brown with uniform extinction (Figure 3a(+)). A small amount of curved worm-like kaolinite was also observed, with a maximum length of 1.015 mm.

The petrology type of C2F was blackish-grey and brownish-grey phaneritic worm-like kaolinite rock, mainly composed of vermicular kaolinite with contents over 98%, as well as a thimbleful of silt-grade quartz fragment and carbonaceous debris (Figure 3b). The length of vermicular kaolinite was generally between 0.36 and 2.00 mm, with the longest length being 4.00 mm. The aggregate extinction of small vermicular kaolinite and scaly kaolinite and carbonaceous debris (Figure 3b) was distributed among large vermicular kaolinite.

The petrology type of C3R was greyish-brown clastic kaolinite rocks with a syngenetic sandy clay texture and jagged fracture, which was composed of 98% fragmental kaolinite and a thimbleful of flaky kaolinite and carbonaceous debris (Figure 3c). Most of the kaolinite debris was subcircular, whereas some of it was subangular, with different particle sizes ranging from 1.16 to 1.35 mm for the large ones and 0.45 to 0.62 mm for the small ones. It was composed of scaly kaolinite, recrystallized vermicular kaolinite, and flaky kaolinite or a few vermicular kaolinites; in particular, the pores between fragments were filled with scaly kaolinite and carbonaceous debris. This texture demonstrates that the weakly consolidated clayey sediments in the basin experienced the process of water impact, whereby they were crushed and redeposited.

The petrology type of C5G was brownish-grey clastic kaolinite rock with conchoidal fracture, and a syngenetic sandy–mud debris texture (Figure 4a). It was composed of 65–70% kaolinite debris and its interspace was filled with fine scaly kaolinite (15–20%), silty quartz (3–5%), carbonaceous or ferriferous debris (3–5%). The kaolinite debris was subcircular and circular, with a particle size ranging from 0.14 and 0.31 mm and arranged in the long axis direction. The kaolinite debris was mainly clastic kaolinite and fine scaly kaolinite (Figure 4a), which presented a uniform extinction. This kind of texture shows that the weakly consolidated clay debris underwent the processes of crushing, short-distance transportation and redeposition via the action of temporary water flow.

The C5F1 samples were greyish-black, sandy-bearing kaolinite rocks from the direct floor of No. 5 coal with clastic texture (Figure 4b). The fragments included monocrystalline kaolinite and scaly kaolinite aggregate, as well as some quartz fragments, in which the kaolinite accounted for 80–85% with a size of 0.13–0.62 mm, and subangular quartz accounted for 15–20% with a size of 0.07–0.36 mm. Intergranular pores were filled with scaly kaolinite, black carbonaceous and ferriferous debris. The C5F2 samples were sandy kaolinite rocks located under the C5F1 samples, mainly composed of quartz (25–30%) and scaly kaolinite (65–70%), along with very little carbonaceous and ferriferous debris (3–5%) (Figure 4c). The quartz fragment was subangular, with a particle size ranging from 0.18 to 0.36 mm. A few clastic kaolinites were occasionally observed. The scaly kaolinite featured uniform extinction, while the carbonaceous debris was distributed in the debris interspace.

The petrology type of C8F was silty kaolinite rock with an aleuritic texture, which was mainly composed of flaky kaolinite (60–65%) with a size of 0.03–0.2 mm, silty quartz (35–40%) and a little carbonaceous debris, clastic kaolinites and some microcrystal clay minerals (Figure 4d).

The composition of kaolinite rocks in different layers from the Carboniferous Permian coal-measure in the Xiaoyu mine was characterized by a little carbonaceous debris in each layer. The C2F and C3R kaolinite rocks in the Taiyuan formation were mainly composed of high-purity kaolinite (>98%). The S4F and C5G samples contained about 95% and 85–90% kaolinite, respectively, and all contained some silty quartz. The kaolinite content is 80–85% and 65–70%, while debris from other minerals is in ranges of 15–20% and 30–35% in the C5F1 and C5F2, respectively. The clay minerals content is 60–65% with a higher debris content of 35–40% in C8F kaolinite rocks.

Additionally, according to the XRD (Figure 5), a few other minerals, such as illite, diaspore, boehmite, anatase and pyrite, were found in the different kaolinite rocks (Table 1).

4.2. Crystallinity

4.2.1. X-ray Diffraction Analysis

X-ray diffraction analysis is one of the most extensive and important clay mineral research methods, and it can be used to study the types and forms of kaolinite structural defects, complex interlaminar interaction, polytype identification and clay mixtures through quantitative or semi-quantitative analysis. X-ray diffraction can be used to study mineral crystallinity and obtain the chemical composition information of major clay mineral components in samples [35,36]. Here, XRD data were used to calculate the Hinkley index (HI) of kaolinite to judge its relative crystallinity (Table 1).

HI = (A + B)/At

(1)

where At is the height from peak to baseline, A and B are the height of ($\stackrel{-}{1}$10) and (11$\stackrel{-}{1}$) peaks, respectively (Figure 5h). When using a Hinkley index analysis, comparisons can only be made if the experimental data are under the same experimental conditions. In addition, the HI index is more suitable for studying kaolinite, which is from clay deposits, but it has some limitations when there are multiple minerals in the samples due to many factors affecting the structural defects of kaolinite, such as the stacking of layers, ion space, grain size, crimping of crystal layers and the influence of experimental conditions [35,36].

A set of diffraction peaks with 2θ within the scope of 20°–30° was selected to represent standard kaolinite crystallinity, and it was analyzed using the Hinckley calculation method [33]. The HI parameter of kaolinite in different layers is listed in Table 1, and the XRD pattern of kaolinite rock is shown in Figure 5. These kaolinites are found in relatively pure deposits but contain trace quantities of quartz, diaspore, anatase, boehmite, illite, pyrite and Fe-oxides (Table 1).

The HI of S4F compact kaolinite rock (Figure 5a) was similar to that of C2F (Figure 5b) and C3R (Figure 5c), with a strong and sharp reflection peak in the (001) and (002) basal planes. The number of diffraction peaks was greater with a higher intensity within the scope of 20°–30°, in which the reflection peak of (02) and (11) crystal band obviously splits as well as the (1$\stackrel{-}{1}$$\stackrel{-}{1}$) reflection peak of 4.17 Å. The Hinckley index calculated from the 02, 11 bands was 1.12, 1.13 and 1.29, respectively, i.e., ranging between 1.1 and 1.3 (Table 1).

The C5F1 sandy-bearing kaolinite rocks XRD pattern (Figure 5e) was similar to that of C5G (Figure 5d), with a low reflection peak intensity but a strong reflection peak in the (020) basal plane. The triplet peak within the scope of 20°–30° was gradually replaced by a double peak, and the (1$\stackrel{-}{1}$$\stackrel{-}{1}$) reflection peak of 4.17 Å was blurred. The HI values of the two samples were 0.81 and 0.82, respectively (Table 1), i.e., ranging between 0.8 and 1.1.

The reflection peak of C5F2 sandy kaolinite rocks (Figure 5f) was wide and its intensity was strong. The different levels of (02) and (11) reflection peaks within the scope of 20°–30° were split and unclear, and they tended to synthesize an asymmetric wide peak, with the weak peak basically disappearing. The HI value of this sample was 0.68.

The XRD pattern (Figure 5g) of C8F silty kaolinite rocks showed the obvious characteristic peak of illite, whereas the (1$\stackrel{-}{1}$$\stackrel{-}{1}$) reflection peak of 4.17 Å disappeared. Its HI value was 0.65.

The experimental analysis shows that the HI can reflect the different genesis of kaolinite, the higher the HI, the better kaolinite crystallinity; on the contrary, the disorder increases [1]. Liu et al. [11] noted that kaolinite crystallinity can be analyzed according to HI value, and the discrimination criteria are as follows: (1) highly ordered: HI ≧ 1.3; (2) ordered: HI = 1.1–1.3; (3) relatively disordered: HI = 0.8–1.1; (4) disordered: HI < 0.8. Additionally, in the X-ray diffraction pattern, with the decrease in crystallinity, the sharpness and intensity of the (001) and (002) basal surface reflection peaks decrease [35,36]. The HI result indicates that the kaolinite crystallinity in C3R, C2F and S4F kaolinite rocks was ordered, the crystallinity of C5F1 and C5G was slightly disordered, and it was disordered in C5F2 and C8F.

4.2.2. Differential Thermal Analysis

Thermal analysis is a method for identifying and studying minerals by using their endothermic and exothermic effects at different temperatures. Since all clay minerals contain hydroxyl groups and most of them contain interlayer water and zeolite water, thermal analysis can be used to analyze the crystalline order of clay minerals.

The common feature of different layers of kaolinite rocks from the Xiaoyu mine (Figure 6) was the presence of two obvious heat-absorbing valleys in the differential heat curves, which were generally at 100 °C. A low-temperature heat-absorbing valley also appeared because of water desorption of the layers, showing that the heat was steadily released during the heating process.

In the medium-temperature phase, the differences were manifested in the strong and deep dehydroxy heat-absorbing valleys in the differential heat curves of C2F, C3R and S4F. The dehydroxy temperature was 510–650 °C and the peak of the dehydroxy valley temperatures was 561.7 °C, 557.3 °C and 552.6 °C, respectively (

Table 2. Kaolinite rocks’ dehydroxy temperature and heat-absorption valley peak values.

Sample	Dehydroxy Temperature (°C)	Heat Absorption Valley Peak Value (°C)
S4F	520–590	552.6
C2F	516–650	561.7
C3R	510–600	557.3
C5G	510–560	544.7
C5F1	510–600	546.4
C5F2	510–560	535.8
C8F	516–580	549.4

).

The dehydroxy valleys of C5G and C5F1 were wide and shallow in the differential thermal curve, with dehydroxy valley temperature values of 544.7 °C and 546.4 °C, respectively. The dehydroxy valley of C5F2 was relatively wide and shallow, and its trough peak value was 535.8 °C. The dehydroxy desorption valley of C8F was strong but shallow, whereas the dihydroxy valley peak temperature value was 549.4 °C, with a dehydroxy temperature of 516–580 °C.

It is generally believed that, from ordered to disordered kaolinite, the medium-temperature hydroxy valley gradually decreases. However, the kaolinite DTA peak of the dehydroxylated endothermic valley is positively correlated with the strength of the OH. Therefore, if the dehydroxyl endothermic valley is wide and the peak value of the thermal absorption valley is small, the kaolinite is disordered. Additionally, the temperature of the dehydroxy endothermic valley can be used as the identification characteristics of kaolinite minerals [37,38]. The dehydroxy temperature of kaolinite is generally 550–600 °C. In the DTA curve of kaolinite, the endothermic valley in the range of 500–600 °C indicates better kaolinite crystallization. Generally speaking, the temperature of the main dehydroxylation peak decreases with the decrease in structural order [38]. DTA analysis shows that the kaolinite crystallinity of C3R, C2F and S4F is ordered, while C5G and C5F1 are relatively disordered and C5F2 is disordered. The dehydroxylation temperature of C8F was 549.4 °C, indicating a relatively ordered crystallinity, but the HI was only 0.65 and the dehydroxyl endothermal trough was shallow. Therefore, C8F crystallinity should be between slightly disordered and disordered, and the relatively high dehydroxylation temperature may be caused by the presence of illite or halloysite minerals.

4.2.3. Infrared Spectroscopy

Infrared spectroscopy (IR) was utilized to confirm kaolinite crystallinity. IR analysis can obtain the characteristic absorption peaks of several main minerals at once without the need for a single mineral separation. According to the position, shape and intensity of these characteristic absorption peaks, the mineral composition in the sample and their relative content can be determined [39,40,41]. In the infrared absorption spectrum, the absorption is generally recorded in the band range of 4000–200 cm⁻¹. The absorption of clay minerals mainly occurs in the high-frequency region of 4000–3000 cm⁻¹ and the middle and low-frequency region of 1200–200 cm⁻¹. In clay minerals, 400–600 cm⁻¹ belongs to Si-O bending vibration, 1000–1200 cm⁻¹ belongs to Si-O stretching vibration and 1600–1700 cm⁻¹ belongs to molecular water (OH) bending vibration. The stretching vibration of hydroxyl (OH) and water (OH) is absorbed in a 3000–4000 cm⁻¹ band. The regional absorption zones, 3700–3600 cm⁻¹, can be used to distinguish the high-ordered kaolinites (HOK) from-low ordered kaolinites (LOK). Kaolinite with good crystallinity has four obvious absorption peaks, which are 3700 cm⁻¹, 3669 cm⁻¹, 3652 cm⁻¹ and 3620 cm⁻¹, respectively, while the middle two absorption peaks of disordered kaolinite merge into one absorption peak of 3653 cm⁻¹.

The results of DTA show that all kaolinite rock samples contain adsorbed water, which will affect the analysis of its crystallinity. Therefore, the IR analysis of its crystallinity should mainly be based on the stretching vibration of the Si-O bond and the bending vibration of the Si-O bond. The more wave numbers of Si-O bond vibration shift to the high-frequency region, the better the crystallinity of kaolinite; on the contrary, the worse the crystallization order.

Kaolinite crystallinity was defined by analyzing the stretching vibration and bending vibration of the Si–O bond of kaolinite (

Table 3. The infrared absorption frequency of coal-measure kaolinite rocks (cm⁻¹).

Sample	υOH and H2O Stretching Vibration	H2O Bending Vibration	Si–O Stretching Vibration	Si–O Bending Vibration
Theoretical wave number	3603	-	1100	468
S4F	3620	1656	1101	469
C2F	3618	1693	1101	471
C3R	3620	1629	1103	473
C5G	3618	1631	1095	465
C5F1	3618	1623	1101	469
C5F2	3620	1629	1093	465
C8F	3620	1638	1092	464

). Generally, when the number of vibration waves of the Si–O bond shifted to the high-frequency region, this denoted better crystallinity and vice versa [39,40,41]. The infrared spectra of S4F (Figure 7a) and C2F (Figure 7b) illustrate that these kaolinites had absorption peaks in the four frequency bands of 3693 cm⁻¹ (S4F) or 3695 cm⁻¹ (C2F), 3670 cm⁻¹, 3656 cm⁻¹ and 3620 cm⁻¹. The bending vibration of the Si–O bond of kaolinite in S4F, C2F and C3R shifted to the high-frequency region by 1, 3 and 5 cm⁻¹ from the theoretical value. The stretching vibration of the Si–O bond of kaolinite shifted to the high-frequency region by 1, 1 and 3 cm⁻¹.

In the infrared spectrum of C5F1, the absorption peaks were in the frequency bands of 3693 cm⁻¹, 3658 cm⁻¹ and 3618 cm⁻¹. The stretching and bending vibrations of the Si–O bond in C5F1 all shifted to the high-frequency region by 1 cm⁻¹. The stretching vibration of the Si–O bond in C5G shifted to the low-frequency region by 5 cm⁻¹, and the bending vibration shifted to the low-frequency region by 3 cm⁻¹ (Table 3).

The infrared spectrum curve of C5F2, which also represents the C5G sample, was similar to that of C8F (Figure 7d), with absorption peaks in the frequency bands of 3693 cm⁻¹, 3653 cm⁻¹ and 3620 cm⁻¹. The stretching vibration of the Si–O bond in C5G, C5F2 and C8F shifted to the low-frequency region by 5, 7 and 8 cm⁻¹, respectively, whereas the bending vibration of the Si–O bond shifted to the low-frequency region by 3, 3 and 4 cm⁻¹ (Table 3).

Based on the IR method, it is generally believed that the temperature of the mesothermal dehydroxylation valley decreases gradually from ordered to disordered kaolinite. If the hydroxyl bonding force in the crystal layer is strong and the dehydroxy heat-absorbing valley in the differential thermal curve of kaolinite is strong and deep, the crystallinity is well ordered. On the contrary, if the dehydroxy heat-absorbing valley is wide and shallow and the temperature of the dehydroxy heat-absorbing valley is low [39,42], kaolinite is poorly crystallized. The heat absorption valley temperature range and the peak value of the differential heat curve fluctuated as a function of the structure and integrity of kaolinite itself and its mineral combinations [42]. Therefore, the kaolinite crystallinity in S4F, C2F and C3R is ordered and relatively disordered in C5F1 and disordered in C5G, C5F2 and C8F.

4.3. Geochemistry

The ideal chemical formula of kaolinite is Al₂O₃·2SiO₂·2H₂O or Al₄·[Si₄O₁₀] (OH)₈, which is expressed in weight percentage as 46.54% SiO₂, 39.5% Al₂O₃ and 13.96% H₂O, with a SiO₂/Al₂O₃ molar ratio of two [43,44,45]. However, the chemical composition in the impurity-bearing kaolinite rocks is different; thus, the quality of kaolinite rocks and their impurity content can be roughly inferred according to the chemical composition. For example, when the molar ratio of SiO₂/Al₂O₃ is equal to two, this means that the silicate minerals are generally represented by kaolinite, with almost no quartz or other impurities. When the molar ratio is less than two, this indicates that there are aluminiferous minerals in the kaolinite rocks, such as boehmite (boehmite), diaspore or gibbsite, which can increase the refractoriness of the kaolinite rocks. When the molar ratio is greater than two, this means there may be quartz or other silicate minerals, such as feldspar, mica, chlorite, illite or montmorillonite. A certain amount of K₂O, Na₂O, CaO or MgO in the chemical composition can validate this interpretation [46,47].

The elemental analysis results of the kaolinite rock showed that the SiO₂ content ranged from 42.63% to 67.97%, the Al₂O₃ content ranged from 19.32% to 38.87% and the SiO₂/Al₂O₃ molar ratio ranged from 1.90 to 5.73 (

Table 4. The element oxides (%), SiO₂/Al₂O₃ molar ratio and CIA (%) in kaolinite samples.

Sample	SiO2	Al2O3	Fe2O3	CaO	Na2O	K2O	MnO	TiO2	P2O5	FeO	$\frac{\mathbf{S}\mathbf{i}{\mathbf{O}}_2}{\mathbf{A}{\mathbf{l}}_2{\mathbf{O}}_3}$	CIA
S4F	52.97	31.49	0.56	0.115	0.054	0.645	0.015	0.460	0.032	0.38	2.85	97.48
C2F	44.97	38.48	0.16	0.079	0.025	0.080	0.005	0.386	0.016	0.12	1.99	99.52
C3R	42.63	38.87	0.47	0.099	0.086	0.202	0.002	1.270	0.049	0.30	1.90	99.01
C5G	52.17	30.76	2.28	0.115	0.060	0.636	0.037	0.849	0.029	1.25	2.88	95.80
C5F1	61.48	24.03	1.51	0.098	0.103	0.972	0.026	0.923	0.033	0.78	4.35	97.31
C5F2	57.17	25.92	2.73	0.265	0.088	0.783	0.014	0.488	0.024	2.10	3.75	95.35
C8F	67.97	20.15	0.86	0.122	0.015	0.720	0.015	1.170	0.029	0.70	5.73	95.92

). The SiO₂/Al₂O₃ molar ratio was mostly greater than two, indicating that these kaolinite rocks mainly contained quartz or illite and other clay minerals. From the previous observation under the microscope, it can be seen that the percentage of quartz was relatively high (~20–40%). The SiO₂/Al₂O₃ molar ratios in the samples were as follows: C3R, 1.90; C2F, 1.99 (close to 2); S4F, 2.85; C5G, 2.88; C5F2, 3.75; C5F1, 4.35; C8F, 5.73.

The chemical index of alteration (CIA) is an important indicator of the intensity of weathering of a parent rock, and it can be used to evaluate the chemical weathering degrees in the provenance and infer the paleoclimate [48,49,50]. The higher the CIA value, the greater the loss of Ca, Na and K, which refers to a more humid and hotter climate environment; alternatively, a lower CIA value refers to a cold environment. As the CIA change pattern has nothing to do with lithological changes, it can be a good indicator of climate change [48,49,50,51,52]. The calculation formula of CIA for kaolinite rock is as follows:

CIA = (Al₂O₃/(Al₂O₃ + CaO* + Na₂O + K₂O)) × 100%,

(2)

where all elements are represented by their mole fraction. CaO* represents the mole fraction of CaO in silicate minerals, and the correction adopts the formula CaO* = CaO − (10/3) P₂O₅ when the calculated mole number is greater than the mole number of Na₂O. The number of moles of CaO* is equal to the number of moles of Na₂O.

The CIA index values of kaolinite rocks in Xiaoyu mine were relatively high, ranging from 95.35 to 99.52. Specifically, in C2F and C3R samples, the CIA was greater than 99.0, while it was 97.31 and 97.48 in C5F1 and S4F samples, and it was equal to 95.35–95.92 in C5G, C5F2 and C8F samples. These results indicate that the kaolinite rocks in the Xiaoyu mine had undergone chemical weathering.

4.4. Geological Effect Factors

Generally, the kaolinite crystallinity in C3R, C2F and S4F was ordered, while it was slightly disordered in C5F1 and disordered in C5F2; however, C5G and C8F are debatable. Based on the DTA, the kaolinite crystallinity in C5G and C8F is slightly disordered, but it is disordered by the IR (

Table 5. Kaolinite crystallinity by petrography, XRD, DTA, IR results and geochemistry index.

Kaolinite Crystallinity				Kaolinite Content		SiO2/Al2O3		CIA (%)
Type	XRD	DTA	IR
Ordered	C3R C2F S4F	C3R C2F S4F	C3R C2F S4F	C2F C3R S4F	>95%	C2F C3R	1.90–1.99	C2F C3R	99.01–99.5
Slightly disordered	C5F1 C5G	C5G C5F1 C8F	C5F1	C5G C5F1	80–90%	S4F C5G	2.85–2.88	S4F C5F1	97.31–97.48
Disordered	C5F2 C8F	C5F2	C5G C5F2 C8F	C5F2 C8F	60–70%	C5F1 C5F2 C8F	3.75–5.73	C5G C5F2 C8F	95.35–95.92

). Although C5G, C5F2 and C8F all show a negative deviation, the IR offset in C5G is similar to C3R when only considering absolute value. However, C5G crystallinity is more ordered than C5F2 and C8F due to its smallest offset. That said, it is unknown whether there were other clay minerals in the three samples that led them to show negative deviations [39]. Depending on the three methods, the crystallinity of C5G and C5F1 is better than that in C5F2 and C8F. However, what are the key factors contributing to kaolinite crystallinity?

Based on the microscopy method of planimetry, the kaolinite content was more than 95% in C2F, C3R and S4F kaolinite rocks with a little carbonaceous debris and silt-grade quartz debris. It was 80–90% in C5G and C5F1 with little silt-grade quartzes, carbonaceous matrix and ferriferous debris. Additionally, there was 60–70% kaolinite with 35–40% quartz and few carbonaceous matrices and ferriferous debris in C5F2 and C8F kaolinite rocks (Table 5). The mineral content statistic showed that the content of kaolinite in each kaolinite rocks coincided with the order of crystallinity, which meant that, the purer the original clay deposits, the fewer non-clay impurities would result in kaolinite being crystallized with high mineral content and kaolinite purity in the claystone of the original sedimentary basin [11,38,53,54,55]. Lu et al. [54] noted that kaolinite crystallinity is related to crystal size. In this paper, the sample with a larger kaolinite crystal size has better crystallinity; however, the crystallinity of C3R and C2F is similar while the crystalline particle size is significantly different. This means that the purity of clay minerals may have a greater influence on crystallinity. To this end, what factors are associated with the purity of clay mineral kaolinite?

Given the particularity of the diagenetic process of coal-measure kaolinite, the following three factors are worth considering: the original sedimentary environment, the weathering after diagenesis and the influence of organic matter.

4.4.1. Original Sedimentary Environment

In the early and middle Taiyuan Period, thick and stable No.8 coal and No.5 coal were formed in the deltaic sedimentary environment, stable No.3 coal and No.2 coal were formed in the continental meandering river sedimentary environment and the No.4 coal seam was formed in the continental braided meandering river sedimentary of the early Permian Shanxi Period [2,21,22,28]. The characteristics of the original sedimentary environment indicated that the sediment of the roof, floor and gangue of No.8 and No.5 coal seams would undergo a stronger filtering effect and retain more stable minerals debris and less clay debris. This feature is more obvious as it is closer to the lower stratum., Due to the higher Si-bearing sedimentary resource background, which struggles to form stable TOa-type clay mineral kaolinite, the static water and low energy environment formed close to the peat swamp [12,14,15,56]. As other clay minerals may also form, kaolinite crystallinity will be biased to being disordered [40,41]. The content and structure of clastic and interstitial matters of C8F, C5F2, C5F1 and C5G confirm this view (Figure 4).

The structure of S4F is unique, with a large amount of cryptocrystalline kaolinite and a small amount of well-crystallized kaolinite. There is no special advantage in particle size but there is an advantage in relation to content. C2F had a worm-like texture, suggesting that it was recrystallized during diagenesis without the reactivation of water bodies after deposition in sedimentary basins with macrocrystalline kaolinite grains [1,2]. The clastic structure of C5F1 is similar to that of C2F, but kaolinite particles are smaller in size and content. The results indicate that, although C5F1 and C2F may have similar hydrodynamic conditions in the early deposition stage, clay minerals with different particle sizes and contents are formed under different sedimentary environments, which directly affects the crystallization order of kaolinite.

C3R had a syngenetic sandy mud texture, suggesting that it was formed following the re-crystallization of weakly consolidated clay deposits in the basin after undergoing water impact, fracture and redeposition [1,2,12]. C5G also had the same sandy mud structure as C3R, but its crystallization index (HI) was 0.81 (Table 1), indicating a lower crystallinity than C3R. The contrast in kaolinite content vs. other minerals in C3R and C5G proves that the original clay mineral content is the most important factor contributing to kaolinite crystallinity. This is caused by the difference between fluvial and deltaic sedimentary environments [1,2,12,57]. Moreover, water impact and redeposition are beneficial for sieving out clay impurities, so the overall crystallinity of C3R is slightly higher than that of C2F.

As for C8F and C52F, due to formation in the delta sedimentary facies, large fine clastic minerals are deposited in this turbulent environment; thus, the content of clay minerals is only 60–70%. The material composition and hydrodynamic environment are not conducive to forming the ordered and larger particle-sized kaolinite crystals. Generally, the particle size of C5F2 and C5G is similar, but C5G is gangue in the coal seam, so the formation environment tends towards static water and weak acid conditions, which may be conducive to the crystallization of kaolinite [2,11,38].

4.4.2. Weathering

In the chemical weathering process of the rock surface, elements are easily weathered and some unstable elements (such as Ca, Na and K) are preferentially removed [3,4,58]. Then, other less mobile elements (such as Al and Ti) are enriched [3,4,46,58]. Therefore, these elements can be used to reflect chemical weathering intensity.

Referring to sample geochemistry, the SiO₂/Al₂O₃ molar ratio and CIA are listed in Table 4; the C2F and C3R with the highest SiO₂/Al₂O₃ molar ratio and CIA values may be due to stronger weathering oxidation during the earlier diagenesis stage. As such, the SiO₂ /Al₂O₃ molar ratio was closer to two and better ordered crystallinity than the other samples [1,45,46,58]. On the other hand, S4F and C5F1 had the same CIA value, but S4F had better crystallinity due to its SiO₂/Al₂O₃ molar ratio being closer to pure kaolinite. Likewise, given the similar CIA for C5G, C5F2 and C8F, their crystallinity order was also consistent with kaolinite purity (SiO₂/Al₂O₃ molar ratio). As for S4F, the sedimentary environment of the basin changed from delta facies to fluvial facies [1,2], meaning that epidiagenesis may be similar to C5G. Its kaolinite crystallinity matches its kaolinite content first and its purity and CIA second.

4.4.3. Organic Matter

These kaolinite rocks were all collected from the roof and floor of the coal seam or gangue and formed in the early or late stage of peat bog formation. Petrological observation shows that there is little organic matter in all of them. Due to the lack of relevant evidence, it is impossible to determine the influence of organic matter on kaolinite crystallinity. The only certainty is that the biochemical process of humic acid generation will lead to an acidic environment, which is conducive to the stable existence of kaolinite minerals [3,7,18]. The fact that the crystallinity of C3R is better than that of C2F and S4F, and that C5G is more disordered than C5F1, C5F2 and C8F, proves this hypothesis.

As clay minerals are authigenic minerals in sedimentary rocks, the above-mentioned geological factors may affect the texture, structure and cementation type of kaolinite rock, as well as the grain size and crystallinity of kaolinite, and then affect the arrangement of structural components and crystallinity in the diagenetic process. Additionally, the effect of the original sedimentary environment is the most obvious, and weathering and organic matter would be secondary.

5. Conclusions

(1)

The kaolinite crystallinity from the Datong coalfield can be divided into ordered, slightly disordered and disordered. Kaolinite crystallinity is positively correlated with its clay minerals content, purity and particle size, which are all affected by the original sedimentary environment and hydrodynamic conditions.

(2)

The kaolinite that formed in continental river facies was ordered, while that from delta facies was relatively disordered and disordered. The existence of organic matter is conducive to the stable existence and crystallinity of kaolinite.

(3)

Crystallinity is related to the rock SiO₂/Al₂O₃ molar ratio and CIA; a higher weathering intensity creates a better order. With the same weathering intensity, a high purity creates a better order.