Correlation between the deformation of mineral crystal structures and fault activity: A case study of the Yingxiu-Beichuan fault and the Milin fault

Chao Xie; Meng Zheng; Lei Liu; Baixu Chen; Fan Yang; Yongcai Wu; Siyuan He

doi:10.1515/geo-2022-0486

Article Open Access

Correlation between the deformation of mineral crystal structures and fault activity: A case study of the Yingxiu-Beichuan fault and the Milin fault

Chao Xie , Meng Zheng , Lei Liu , Baixu Chen , Fan Yang , Yongcai Wu and Siyuan He

Published/Copyright: July 10, 2023

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information

From the journal Open Geosciences Volume 15 Issue 1

Abstract

The build-up and occurrence of earthquakes are due to the accumulation and release of stress in fault zones. When subjected to tectonic extrusion stress, the crystal structure of the minerals within a fault zone will change. In this study, Raman spectroscopy analysis was conducted on the concurrently deposited quartz veins from Shenxigou, along the Yingxiu-Beichuan fault, and from Niyang River mouth, in the southern section of the Milin fault. The test results reveal a 3.29 cm⁻¹ shift in the characteristic 464 cm⁻¹ peak of the quartz in the veins along the fault plane of the Yingxiu-Beichuan fault, which was significantly lower than the shifts in the quartz peaks of the quartz on both sides of the vein. The 464 cm⁻¹ peak shifts of the samples collected 10 m to the NW and 21 m to the SE of the fault plane were approximately 4.40 and 4.62 cm⁻¹. In the veins from the Milin fault, considerable shifts of the 464 cm⁻¹ quartz peaks occurred at the fault plane and to both sides within 5.5 m of the fault plane. No significant change in the 464 cm⁻¹ Raman peak of quartz was observed for the samples 5–28 m to the SE of the fault plane. These results indicate that the tectonic extrusion stress accumulated more easily in proximity to the fault plane, resulting in significant changes in the crystals near the fault plane. We conclude that there is a correlation between the degree of change in the crystal structures of the minerals in thrust fault zones and fault activity, and such a correlation can provide a new method for studying the activity of thrust faults in areas with bedrock.

Keywords: Yinxiu-Beichuan fault; Milin fault; Raman peak of quartz; shift of the 464 cm−1 peak; crystal structure; thrust fault activity

1 Introduction

When a material is subjected to compressive stress, its crystal structure (parameters of the crystal lattice) will undergo significant changes, including shortening of bond lengths, reduction of bond angles, and even destruction of the original crystal structure [1–5]. Experiments on the mechanics of materials have shown that stresses accumulate and are released more easily near the interface between two materials, resulting in significant deformation [6–8]. The structures of fault zones in rock masses consist of a hanging wall, a footwall, and a fault plane (similar to the two-material interface). Fault zones are also susceptible to the accumulation and release of tectonic stress. The build-up and occurrence of earthquakes are due to processes related to the long-term accumulation and sudden release of tectonic stress along fault zones, respectively [9–12]. In thrust fault zones subjected to tectonic extrusion stress, the crystal structure of minerals, especially those near the fault plane, also changes. In addition, fault activity (earthquakes) can release the extrusion stress in the fault zone, and to a certain extent, the deformed crystal structure in the fault zone can be restored to its original state. Based on this, there is a certain correlation between the changes in the crystal structure and distribution characteristics and the activity in the thrust fault zone. Exploring the relationship between the two can provide new ideas for studying the activity of thrust faults in bedrock.

The electrons in molecules of matter are generally in the ground state. When exposed to incident light, elastic and inelastic scattering will occur. Scattered light from elastic scattering has the same wavelength as the excitation light, while scattered light from inelastic scattering has a longer and shorter component than the wavelength of the excitation light. This phenomenon is called the Raman effect. The Raman effect originates from the reactions of the vibration and rotation of molecules to incident light; therefore, Raman spectroscopy is commonly used to identify the structures of molecules. Every material capable of experiencing the Raman effect has a set of specific Raman peaks. When subjected to external forces, the molecular structure and Raman peak of the material change. The position of the peak (Raman shift) (v _p) in the parameters of Raman spectrometry reflects the vibrational dynamic properties of the ground state of the electrons in the sample molecules, which is expressed in terms of the difference in the wavelength between the incident and scattered light. The change in the position of the peak (peak shift) is mainly determined by the change in the molecular structure of the material and is closely related to the stress on the material [13–16].

Currently, Raman spectroscopy has been widely used to detect tectonic stress and the occurrence state of deep minerals. The Raman vibrational properties of calcite under hydrostatic and differential stresses studied using a first-principle method revealed that the frequencies of all Raman modes increase with increasing pressure [17]. The experimental and calculated Raman frequencies of 3C–SiC materials also linearly increase with increasing pressure [18,19]. A new Raman peak around 517 cm⁻¹ appeared at 2.9 GPa in a diamond anvil cell experiment, implying that a new phase of oligoclase was produced [1]. The residual pressures from quartz inclusions in eclogitic garnet in the Sumdo oceanic slab in Tibet obtained via Raman geobarometry had the highest residual pressure of 0.53 ± 0.07 GPa [20]. The nature of the stress percolation patterns of hydraulic fractured Gondwana coal was predicted based on the shift in the disordered carbon band (D band) of Raman spectra using the laboratory hydraulic fracturing method [21]. The characteristic Raman peaks of quartz and calcite grains in fault gouge generated by the Wenchuan earthquake shifted toward higher frequencies due to the tectonic stress during the fault slip [22]. Veins are indicators of shallow magmatic activity close to the surface, relatively rapid cooling and formation on a small scale, and infilling of the veined rock masses in a jointed arrangement [23,24]. The geometry and morphology of the veins reflect the characteristics of the regional stress field [25]. Quartz is the most common mineral in the Earth’s crust and in felsic veins. The crystal structure of quartz is sensitive to pressure, and the relationship between the stress and the shift of the characteristic 464 cm⁻¹ Raman peak of quartz has been studied in detail. Additionally, quartz is widely used for pressure calibration in high pressure in situ experiments [26,27]. In a previous study, we performed Raman spectroscopy on quartz from rock veins in the Shenxigou exposure of the Yingxiu-Beichuan fault [28]. The Yingxiu-Beichuan fault is the central fault of the Longmenshan fault zone, which has been very active since the Late Quaternary [29] (Figure 1). Trench excavation studies have shown that at least three paleoseismic events have occurred on this fault since the Holocene [29]. The Wenchuan earthquake on May 12, 2008, produced a ∼230 km long surface rupture zone along the fault [31]. The 464 cm⁻¹ Raman peak shifts and distribution were obtained for the samples [28]. However, those data were limited, so that using them to represent fault activity is associated with a large uncertainty. In addition, there were no data for comparison. Therefore, in this study, we collected samples from rock veins in the Niyang River mouth in the southern section of the Milin fault. The northeast-trending Milin fault is developed on the northwest side of the Namche Barwa syntaxis (Figure 1). The northeastern segment of the Milin fault is a Holocene active fault with strong activity, and it dislocates the T2 terrace of the Lulang River [32]. The southwestern segment is a less active Late Pleistocene active fault, which dislocates the T4 terrace of the Yarlung Zangbo River [33]. The quartz particles in the samples were analyzed using Raman spectroscopy to determine the 464 cm⁻¹ peak shifts. These results, combined with test results for samples from the Shenxigou exposure of the Yingxiu-Beichuan fault, reveal the relationship between the deformation of crystal structures and fault activity in thrust fault zones.

Figure 1

Geologic map of the Tibetan Plateau and its surrounding areas. WQLF: west Qinling fault; EKLF: east Kunlun fault; MJF: Minjiang fault; YBF: Yingxiu-Beichuan fault; DRF: Dari fault; XSF: Xianshuihe fault; YGF: Yushu-Ganzi fault; LQSF: Longquanshan fault; NJF: Nujiang fault; JLF: Jiali fault; MLF: Milin fault; YRF: Yarlung Zongbo river fault; MCT: Main Central thrust; MBT: Main Boundary thrust; NBS: Nachmche Barwa Syntaxis; SS: Sang Syntaxis; AS: Assam Syntaxis. Active faults are modified from Tapponnier et al. [30].

2 Sample collection, experimental methods, and data processing

2.1 Sampling

Samples of veins were collected from the Shenxigou exposure of the Yingxiu-Beichuan fault. Andesite is developed in this locality. The fault plane dips to the northwest at a 50° angle in this exposure. The rock mass near the fault plane is strongly broken, forming a strong fracture zone with a width of about 20 cm. A thin layer of unconsolidated fault gouge is developed along the fault plane. The cleavage on the northwest side of the fracture zone is developed, and the joints are dense within 10 m on both sides of the zone. A Holocene gravel layer is deposited at about 32 m in the footwall of the fault (Figures 2a and 3a). This locality had no vein development since the veins were small, there was poor continuity (Figures 4a–d), and there was local reticulation (Figure 4d). We collected 12 rock vein samples within a 48 m range across the fault plane, including six from the footwall, five from the hanging wall, and one from the fault plane.

Figure 2

(a) Schematic diagram of the Shenxigou exposure of the Yingxiu-Beichuan fault (modified after Xie et al. [28]). (b) Schematic diagram of the Niyang River mouth exposure of the Milin fault.

Figure 3

Pictures of the fault exposure: (a) Shenxigou exposure of the Yingxiu-Beichuan fault and (b) river mouth exposure of the southern section of the Milin fault. The red triangles with yellow edges indicate the fault plane. The blue rectangles denote the sampling points.

Figure 4

Pictures of exposures of the sampled veins: (a)–(d) pictures of the veins in the Shenxigou exposure of the Yingxiu-Beichuan fault showing that the veins are fine and poorly elongated and (e)–(h) pictures of the veins in the Niyang River mouth exposure of the Milin fault showing that the veins are well-developed and have good continuity.

Vein samples were collected from the mouth of the Niyang River in the southwestern segment of the Milin fault. The bedrock developed in this exposure is gneiss. The exposed fault plane dips to the NW, forming a strong fracture zone with a width of about 15 cm. The breccia in the fracture zone and the thin layer of fault gouge on the fault plane are semi-consolidated. The joints on both sides of the fracture zone are dense and well-developed (Figures 2b and 3b). At this locality, rock veins were developed, with widths of generally >10 cm and relatively good extension (Figure 4e–h). We collected 26 vein samples at 1–3 m intervals across the fault plane, including 13 from the SE wall (hanging wall), 12 from the NW wall (footwall), and 1 in the fault plane. The samples from this locality were collected within a sampling range of 54 m. To avoid external forces acting on the samples, the sample test site was marked and excavated at the periphery during the sample collection.

2.2 Experimental methods and data processing

The Raman spectroscopy analysis of the quartz particles from the Yingxiu-Beichuan fault was performed using the SP2300 Laser Confocal Micro Raman Spectrometer at the Institute of Earthquake Forecasting. The light source was an Ar+ plasma laser, with a wavelength of 523 nm, a power of 5 mW, a spot size of 10 µm, a slit width of 50 µm, a 50× objective lens, and a peak collection time of 20 s. The instrument was calibrated using monocrystalline silicon wafers before the samples were analyzed. The Winspec32 software was used to obtain the Raman spectrum in the range of 100–1,000 cm⁻¹. Multiple tests were carried out on each sample, and the spectrum with the strongest spectral peak and the weakest back-bottom baseline (the baseline was a straight line) was selected as the final test result. The Peakfit software was used to process the data, the “Autofit and subtract baseline” function was used to establish and fit the baseline. A Gaussian function was selected for the spectral peak fitting, and iterative calculation was required during the fitting until the correlation coefficient R ² was >0.99. The “Line model” and “Line + Symbol” in OriginPro 2015 software were used to draw the figures.

The Raman spectroscopy analysis of the samples collected from the Milin fault was carried out using the HORIBA instrument at the State Key Laboratory of Continental Dynamics of Northwest University. The instrument had a solid-state laser with a laser wavelength of 523 nm, a spot size of 2 µm, a 50× objective lens, and a peak collection time of 2 s. This instrument was also calibrated using monocrystalline silicon wafers before the samples were analyzed. During this test, the “linear fitting” tool embedded in the LabSpec6 software was used to conduct backbottom deduction, and the fitting was repeated several times until a suitable backbottom line was obtained. During the Raman test, the automatic peak location method was adopted to determine the position of the spectral peak. A Gaussian function was selected for the spectral peak fitting. The spectral peak baseline also needed to be deducted during the fitting to make its intensity 0, and an accurate spectral peak value was obtained after multiple fitting steps.

3 Results

We conducted Raman spectroscopy analysis of quartz from 12 samples from veins in the Shenxigou exposure of the Yingxiu-Beichuan fault and 26 samples from veins in the Niyang River mouth exposure of the Milin fault. The characteristic 464 cm⁻¹ peak shifts and distribution of the samples were compared and interpreted.

3.1 Raman spectroscopy results for the quartz from the Shenxigou exposure of the Yingxiu-Beichuan fault

The Raman peaks of the quartz samples from 12 rock veins in the fault zone were obtained. Except for sample YB-5, the peak intensity was relatively weak, and all of the samples exhibited the standard peaks for quartz at 210, 264, 354, and 464 cm⁻¹; while stronger peaks appeared near 464 cm⁻¹ and weaker peaks appeared near 410 cm⁻¹ (Figure 5). The Gaussian fitting results for the sample with the strongest characteristic 464 cm⁻¹ Raman peak revealed that the 464 cm⁻¹ peaks of all the samples were shifted to a higher wavenumber due to the effect of the tectonic extrusion stress (Figure 6). The 464 cm⁻¹ peak for sample YB-0, which was collected from the fault plane, had the smallest shift (Δv _p/cm⁻¹) toward a high wavelength (3.92 cm⁻¹) (Table 1). Further away from the fault plane, the shift increased, and the 464 cm⁻¹ peak shift of the sample (YB-5) collected 21 m to the SE of the fault plane was 4.40 cm⁻¹ (Table 1). For sample YB-c 464 cm⁻¹, which was collected 10 m to the NW of the fault plane, the peak shift decreased after reaching the maximum value of 4.62 cm⁻¹ (Table 1). Moreover, from 15 m (YB-d) to 17 m (YB-e) to the NW of the fault plane, there was also a sharp decrease in the shift from 4.53 to 3.57 cm⁻¹ (Table 1, Figure 7).

Figure 5

Raman peaks of quartz crystals from veins in the Shenxigou exposure (modified after Xie et al. [28]).

Figure 6

Characteristic 464 cm⁻¹ Raman peaks of quartz crystals from veins in the Shenxigou exposure (modified after Xie et al. [28]).

Table 1

Shifts in the 464 cm⁻¹ Raman peak of quartz crystals from veins in the Shenxigou exposure of the Yingxiu-Beichuan fault (modified after Xie et al. [28])

Sample	Raman mode (v _p) (cm⁻¹)	Distance from displacement surface (m)	Wavenumber shift (Δv _p) (cm⁻¹)
YB-5	468.4	−21	4.40
YB-4	468.13	−18	4.13
YB-3	467.92	−10	3.92
YB-2	468.10	−7	4.10
YB-1	467.62	−4	3.62
YB-0	467.29	0	3.29
YB-a	467.74	2	3.74
YB-b	468.38	5	4.38
YB-c	468.62	10	4.62
YB-d	468.53	15	4.53
YB-e	467.57	17	3.57
YB-f	467.21	27	3.21

Note: ¹‘−’ indicates that the sample was collected from the SE wall of the fault.

Figure 7

Distribution of characteristic 464 cm⁻¹ Raman peak shifts of quartz crystals from veins in the Shenxigou exposure (modified after Xie et al. [28]).

3.2 Raman spectroscopy results for the quartz from the veins in the Niyang River mouth exposure of the Milin fault

We performed Raman spectroscopy analysis of 26 samples from the Niyang mouth exposure of the Milin Fault Zone, including 13 from the SE wall, 12 from the NW wall, and 1 from the fault plane. The test results were similar to those observed for the Shenxigou quartz samples, and apart from individual samples, the peak intensity was weak (such as: ML-12, ML-11, ML-6, and ML-m). The remaining samples exhibited stronger peaks near the standard quartz peaks at 210, 264, 354, and 464 cm⁻¹ and weaker peaks near 410 cm⁻¹ (Figures 8 and 9). The Gaussian fitting results for the sample with the strongest characteristic 464 cm⁻¹ Raman peak showed the effect of the tectonic extrusion stress, and for all of the samples, the 464 cm⁻¹ peak was shifted toward higher wavelengths (Figures 10 and 11); however, the overall shifts were less than the peak shifts of the samples from the Shenxigou exposure. The shifts in this exposure occurred within 5.5 m on both sides of the fault plane (ML-2 to ML-b range), and the maximum value was 2.1 cm⁻¹ (ML-0) (Table 2). However, the peak shifts of the samples collected 1 m to the SE (ML-a) and 1 m to the NW (ML-1) of the fault plane decreased sharply. There was no significant change in the position of the 464 cm⁻¹ peak in the samples collected from 5 m (ML-c) to 28 m (ML-m) to the SE of the fault plane, and the shifts were small and concentrated from 0.72 to 0.42 cm⁻¹ (Table 2). However, there was a significantly larger peak shift in the samples collected to the NW of the fault plane. The peak shifts of the samples collected 5 m (ML-3) to 21 m (ML-10) to the NW of the fault plane gradually increased, but the peak shift of the sample collected 24 m (ML-11) to the NW of the fault plane declined sharply to 0.75 cm⁻¹ (Table 2, Figure 12).

Figure 8

Raman peaks of quartz crystals from veins in the NW wall of the fault in the Niyang River mouth exposure.

Figure 9

Raman peaks of quartz crystals from veins in the SE wall of the fault in the Niyang River mouth exposure.

Figure 10

Characteristic 464 cm⁻¹ Raman peaks of quartz crystals from veins in the NW wall of the fault in the Niyang River mouth exposure.

Figure 11

Characteristic 464 cm⁻¹ Raman peaks of quartz crystals from veins in the SE wall of the fault in the Niyang River mouth exposure.

Table 2

Shifts in the 464 cm⁻¹ Raman peak of quartz crystals from veins in the Niyang River mouth exposure of the Milin fault

Sample	Raman mode (v _p) (cm⁻¹)	Distance from displacement surface (m)	Wavenumber shift (Δv _p) (cm⁻¹)
ML-12	464.68	−26	0.68
ML-11	464.75	−24	0.75
ML-10	465.41	−21	1.41
ML-9	464.75	−19	1.75
ML-8	465.22	−17	1.22
ML-7	465.36	−14	1.36
ML-6	465.10	−11	1.10
ML-5	464.84	−9	0.84
ML-4	465.09	−7	1.09
ML-3	465.12	−5	1.12
ML-2	467.57	−3	1.98
ML-1	467.21	−1	0.91
ML-0	466.1	0	2.1
ML-a	464.71	1	0.71
ML-b	465.19	2.5	1.90
ML-c	464.72	5	0.72
ML-d	464.65	8	0.65
ML-e	464.66	11	0.66
ML-f	464.58	13	0.58
ML-g	464.55	15	0.55
ML-h	464.66	18	0.66
ML-i	464.54	20	0.54
ML-j	464.60	22	0.60
ML-k	464.54	24	0.54
ML-l	464.42	26	0.42
ML-m	464.71	28	0.71

Note: ¹‘−’ indicates that the sample was collected from the NW wall of the fault.

Figure 12

Distribution of characteristic 464 cm⁻¹ Raman peak shifts of quartz crystals from veins in the Niyang River mouth exposure.

4 Discussion

4.1 Constraints on the deformation of the crystal structures in the fault zones

When subjected to various stresses, the characteristic 464 cm⁻¹ Raman peak of the quartz samples from the veins in the Yingxiu-Beichuan and Milin fault, shifted toward a high frequency on average. This stress originated from two components. The first was the original internal stress that remained in the crystal structure during mineral crystallization due to the influence of a high temperature environment and a limited growth space, and this internal stress could not be released later [6,34]. The second was the tectonic extrusion stress exerted by the crustal movement. High-pressure in situ experimental studies have shown that quartz can bear up to <2.0 GPa of stress, and that after the pressure stress is released, the crystal structure can recover to its original state [26]. The rock veins in the Yingxiu-Beichuan fault have poor elongation, display local reticulation, have small individual crystals, and exhibit a low level of euhedral crystal development (Figure 4a–d). However, the veins in the Milin fault are larger and wider, with larger mineral particles and highly euhedral crystals (Figure 4e–h). Therefore, we believe that the growth space of the veins in the Shenxigou exposure was very small during the crystallization process, and thus, more of the original residual stress was retained. This predominantly accounted for the frequency shifts of the 464 cm⁻¹ characteristic Raman peak of the quartz in the veins of that the Shenxigou exposure, which was generally greater than the shifts of this characteristic peak in the samples from the Niyang River mouth exposure.

4.2 Relationship between crystal structure deformation and fault activity in fault zones

The transition pressure of quartz at room temperature is 2 GPa, and the 464 cm⁻¹ Raman frequency of quartz has a linear relationship with pressure below 2 GPa [26]. The tectonic stress of the fault zone is far less than 2 GPa [35]. The samples we collected from the same exposure (Shenxigou exposure or Niyang River mouth exposure) belong to contemporaneous veins, which have experienced the same deep thermal action and retained the same residual thermal stress in their crystal structure. In addition, the samples have been exposed to the surface and room temperature for a long time. Therefore, it can be concluded that the differences in the quartz crystal structures (indicated by changes in the 464 cm⁻¹ Raman peak) collected from the same fault zone were mainly caused by tectonic stress. In particular, we collected as many samples as possible according to the degree of vein development. The sampling interval was 2–10 m in the Shenxigou exposure and 1–3 m in the Niyang River mouth exposure (Figure 2). Since only a very small amount of sample (2 and 10 μm light spots) was required for the analysis, samples with a diameter of 1–2 cm were collected at each sampling point. The sampling interval and sampling proportion can fully reflect the spatial variation trend of the 464 cm⁻¹ Raman spectrum peak of the quartz samples and residual stress.

The Yingxiu-Beichuan fault, which is the central fault in the Longmenshan fault zone, is subjected to strong tectonic extrusion stress from both the Bayan Har block and the Sichuan Basin (Figure 1), causing the accumulation of a high amount of tectonic stress in and around the fault zone. As a result, in the Shenxigou, for the samples collected from 5 m (YB-b) to 15 m (YB-d) to the NW of the fault plane and the samples collected from 7 m (YB-2) to 21 m (YB-5) to the SE of the fault plane, the 464 cm⁻¹ characteristic Raman peak was shifted toward a higher frequency. These shifts were significantly greater than the peak shifts for the samples (YB-e and YB-f) collected farther to the NW of the fault plane (Table 1, Figure 7). Since the area to the southeast of YB-5 was covered by flood fan deposits, vein samples were not collected. However, we speculate that in the areas far from the fault zone, the crystal structure of the quartz in the veins will, to some extent, return to its original state due to the lower tectonic extrusion stress. The Yingxiu-Beichuan fault, which has been very active since the Late Quaternary [29,31], was also the fault that triggered the magnitude 8.0 Wenchuan earthquake in 2008 [31,36]. Multiple displacement events have occurred along the fault plane, which has resulted in a sustained release of tectonic extrusion stresses accumulated on and nearby the Shenxigou fault plane (YB-0, YB-a, and YB-1). Therefore, the crystal structure of the quartz in the veins in this exposure had largely been restored to its original state, and the shifts in the 464 cm⁻¹ Raman peak toward higher frequencies were significantly smaller than the shifts observed on either side of the fault plane (Table 1, Figure 7). High-pressure in situ experimental studies have shown that the crystal structure of quartz can be restored to its original state without being destroyed after being subjected to pressures of <2.0 GPa [26]. At room conditions, when the pressure is less than 2.0 GPa, the crystal structure of quartz only undergoes elastic deformation, and the Raman frequency of 464 cm⁻¹ exhibits a linear relationship with pressure. The stress in the fault zone is often less than 100 MPa [35]. In the Shenxigou exposure, the crystal structure of the quartz vein samples collected from 5 m (YB-b) to 15 m (YB-d) to the NW of the fault plane and from 7 m (YB-2) to 21 m (YB-5) to the SE of the fault plane was significantly deformed (Table 1, Figure 7). We believe that the tectonic extrusion stress borne by the fault zone is different from that in the high-pressure in situ experiments conducted in the lab. Instead, what is observed is the result of long-term loading stress (far less than 100 Ma). That is, under long-term tectonic extrusion stress (far less than 100 Ma), the crystal structure of the quartz in the veins within this range has undergone creep without being destroyed. Therefore, unlike in elastic deformation, the resulting deformation of the lattice cannot be restored to the original state after the stress is released.

Due to vein development, numerous samples were collected from the southern section of the Milin fault at the mouth of the Niyang River. Raman spectroscopy analysis was conducted on these quartz samples, and the distribution of the shifts in the characteristic 464 cm⁻¹ peak at this location were compared with the shifts of the samples collected from the Shenxigou exposure of the Yingxiu-Beichuan fault. The Milin fault is affected by the interaction between the Namcha Barwa block to the SE and the Lhasa block to the NW, both of which exert tectonic extrusion stress on the Milin fault [33,37]. The southern section of the Milin fault has been very weakly active since the Late Pleistocene [33]. Thus, tectonic stress has been accumulating on and near the fault plane, and this stress has not yet been released. The structure of the quartz crystals in the veins (ML-0, ML-2, and ML-b) exhibited a large amount of deformation, and shifts in the 464 cm⁻¹ Raman peak also occurred in the samples from this area (Figure 12). The shifts in the 464 cm⁻¹ peak within 1 m of the fault plane (ML-1 and ML-a) were relatively small (Table 2, Figure 12). We believe that a series of secondary displacement surfaces have developed near the main fault plane (ML-0), and that creep and slip along these minute displacement surfaces have caused the continuous release of the tectonic stress near these surfaces.

In the Niyang River mouth exposure, for the samples collected from 5 m (ML-c) to 28 m (ML-m) to the SE of the fault plane, there were small shifts (concentrated between 0.72 and 0.42 cm⁻¹) in the 464 cm⁻¹ Raman peak toward higher wavelengths (Table 2, Figure 12). However, the shifts gradually increased from 5 m (ML-3) to 21 m (ML-10) to the NW of the fault plane, and the largest shift was 1.75 cm⁻¹ (Table 2, Figure 12). There are two possible scenarios that can explain these trends. First, tectonic extrusion stress normally accumulates more easily on a passive thrust fault wall (the NW wall of the southern section of the Milin fault at the mouth of the Niyang River) than in an active wall (the SE wall of the southern section of the Milin fault at the mouth of the Niyang River). Second, although there is no significant displacement along the strike of the fault plane of the thrust fault, if there is fault creep, the active wall gradually releases tectonic stress during long-term creep and slip. For the samples collected 24 m (ML-11) and 26 m (ML-12) to the NW of the fault plane, the 464 cm⁻¹ Raman peak shifted to 464.75 and 464.68 cm⁻¹, respectively (Table 2). This peak shift phenomenon is similar to that observed in the NW wall of the Yingxiu-Beichuan fault in the Shenxigou exposure, indicating that the stress of the thrust fault was concentrated within a certain range on both sides of the fault plane (Figures 7 and 12).

4.3 New investigative method for studying thrust fault activity

Recognizing fault activity is important for the reasonable assessment of the future risk posed by the occurrence of a major earthquake. At present, assessments are mainly based on signs of paleo-earthquake events in Quaternary strata, such as stratigraphic faults, deformation features, colluvial wedges, growth strata, and liquefaction of paleo-sand and soil layers. When combined with the chronology of the related strata, constraints for the occurrence of paleo-earthquakes along faults can be obtained and used to assess the fault activity [38,39]. However, regarding the development of faults in bedrock, often, there is a lack of effective means of determining the activity. Fault zones are susceptible to the accumulation and release of tectonic stress. The build-up and occurrence of earthquakes are due to processes related to the long-term accumulation and sudden release of tectonic stress along fault zones, respectively.

In particular, in thrust fault zones subjected to tectonic extrusion stress, the crystal structure of minerals, especially those on and near the fault plane, will be changed. Thus, the characteristics of the 464 cm⁻¹ Raman spectra peak of quartz collected from the Shenxigou exposure along the Yingxiu-Beichuan fault, and from the Niyang River mouth exposure in the southern section of the Milin fault exhibit shifts toward higher frequencies. The southern section of the Milin fault has experienced weak activity, so the tectonic stress accumulated near the fault plane and the materials with the greatest deformation usually accumulate at this location. Most of the shifts of the 464 cm⁻¹ peak of quartz are also distributed near the fault plane. However, the Yingxiu-Beichuan fault has experienced strong activity, the tectonic stress near the fault plane has been continuously released during multiple displacements, and the deformed crystal structure at this location has largely been restored to its original state, so the smallest shifts of the 464 cm⁻¹ peak of quartz were also measured near the fault plane. The correlation between the changes in the crystal structure and the distribution characteristics and activity of the thrust fault provides a new method for studying the activity of thrust faults in bedrock.

5 Conclusions

In this study, Raman spectroscopy analysis was conducted on samples of concurrently deposited quartz veins collected from the Shenxigou exposure of the Yingxiu-Beichuan fault and from the Niyang River mouth exposure in the southern section of the Milin fault.

In the highly active Shenxigou exposure of the Yingxiu-Beichuan fault, the continuous release of the tectonic stress caused the shifts in the 464 cm⁻¹ Raman peak (toward a higher wavelength) of the quartz in the veins in the fault plane to be significantly smaller than the shifts in the quartz from the samples collected from either side of the fault plane. However, the southern section of the Milin fault is very weakly active. Due to the continuous accumulation of tectonic extrusion stress, the shifts in the 464 cm⁻¹ Raman peak (toward a higher wavelength) of the quartz from the veins in the Niyang River mouth exposure were largest near the fault plane. Therefore, the relative degree of deformation of the mineral crystal structure in the thrust fault zone has a certain correlation with the fault activity, which provides us with new ideas and methods for studying the activity of thrust faults in bedrock.

Acknowledgements

This work was supported by the Open Foundation of the United Laboratory of High-Pressure Physics and Earthquake Science (2019HPPES10) and the State key laboratory of earthquake dynamics (Project No. LED2021B02).

Author contributions: C.X. and L.L. conducted the field work and wrote the manuscript. M.Z., Y.C.W., and S.Y.H. carried out the laboratory work and processed some of the data. F.Y., B.X.C., and S.Y.H. processed some of the data.
Conflict of interest: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.

References

[1] Xie C, Du JG, Cui YJ, Chen Z, Zhang WB, Yi L, et al. Variation of Raman spectra of oligoclase under 1.0–4.4 GPa. Spectrosc Spectr Anal. 2012;32(3):691–4.Search in Google Scholar

[2] Iizuka-Oku R, Soustelle V, Miyajima N, Walte NP, Frost DJ, Yagi T. Experimentally deformed lawsonite at high pressure and high temperature: Implication for low velocity layers in subduction zones. Phys Earth Planet Inter. 2019;295:160282.Search in Google Scholar

[3] Hao M, Zhang JS, Pierotti CE, Ren ZY, Zhang D. High-pressure single-crystal elasticity and thermal equation of state of omphacite and their implications for the seismic properties of eclogite in the Earth’s interior. J Geophys Res: Solid Earth. 2019;124:2368–77.Search in Google Scholar

[4] Young G, Fan LL, Zhao B, Chen X, Liu X, Huang HJ. Equation of state for Fe-9.0 wt% 0 up to 246GPa: implications for oxygen in the Earth’s outer core. J Geophys Res: Solid Earth. 2021;126:e2020JB021056.Search in Google Scholar

[5] Shephard GE, Houser C, Hernlund JW, Valencia-Cardona JJ, Trønnes RG, Wentzcovitch RM. Seismological expression of the iron spin crossover in ferropericlase in the Earth’s lower mantle. Nat Commun. 2021;12:5905.Search in Google Scholar

[6] Qiu Y, Lei ZK, Kang YL, Hu M, Xu H, Niu HP. Micro-Raman spectroscopy and applications to measure residual stress in micro-structure. J Mech Strength. 2004;26(4):389–92 (in Chinese with English Abstract).Search in Google Scholar

[7] Mu MF, Osswald S, Yury G, Karen IW. An in situ Raman spectroscopy study of stress transfer between carbon nanotubes and polymer. Nanotechnology. 2009;20:335703.Search in Google Scholar

[8] Deng WL, Qiu W, Jiao YZ, Zhang QC, Hang YL. Residual stress measurement and analysis in multilayer film Si-substrate structure by micro-Raman spectroscopy technique. J Exp Mech. 2012;27(1):1–9 (in Chinese with English Abstract).Search in Google Scholar

[9] Harris RS. Introduction to special section: stress triggers, stress shadows, and implications for seismic hazard. J Geophys Res. 1998;103:24347–58.Search in Google Scholar

[10] Niu LL, Du JJ, Feng CJ, Meng W. In-situ stress measurement of deep borehole in east of Hebei and its significance. Acta Seismol Sin. 2015;37(6):89–102 (in Chinese with English Abstract).Search in Google Scholar

[11] Meng W, Chen QC, Wu ML, Li GQ, Qi XH, Feng CJ. Research on segmentation and acteristic of tectonic stress field of Longmenshan Fault Zone. Prog Geophys. 2013;28(3):1150–60 (in Chinese with English Abstract).Search in Google Scholar

[12] Blas JMD, Iaffaldano G. Using rigid microplate motions to detect the stress buildup preceding large earthquake: a feasibility based on synthetic models. J Geophys Res. 2019;12:13468–85.Search in Google Scholar

[13] Mao HK, Hemley RJ. Ultrahigh-pressure transitions in solid hydrogen. Rev Mod Phys. 1994;66:671–92.Search in Google Scholar

[14] Huang XL, Fang FL, Huang YP, Wu G, Li X, Zhou Q, et al. High-pressure Raman study of solid hydrogen up to 300 GPa. Chin Phys B. 2016;25(3):037401.Search in Google Scholar

[15] Yan SY, Zhou YQ, Chen Y. The laser Raman spectroscopic study of pyroxene and olivine at the high-temperature. Chin J Light Scattering. 2005;16:325–31 (in Chinese with English Abstract).Search in Google Scholar

[16] Zeng H, Ye TT, Cheng P, Yao DY, Ding JF. Raman spectroscopy investigation on the pressure-induced structural and magnetic phase transition in two-dimensional antiferromagnet FePS3. Chin Phys B. 2022;31(5):056159 (in Chinese with English Abstract).Search in Google Scholar

[17] Liu L, Lv CJ, Zhuang CQ, Yi L, Liu H, Du JG. Effects of differential stress on the structure and Raman spectra of calcite from first-principles calculations. Am Mineralogist. 2016;101:1892–7.Search in Google Scholar

[18] Liu L, Yi L, Liu H. First principles calculation of the nonhydrostatic efects on structure and Raman frequency of 3C-SiC. Sci Rep. 2018;8:11279.Search in Google Scholar

[19] Debernardi A, Syassen K, Cardona M, Ulrich C. Raman line widths of optical phonons in 3C-SiC under pressure: first-principles calculations and experimental results. Phys Rev B. 1999;59(10):6774.Search in Google Scholar

[20] Zhang C, Thomas B, Liu XY, Shen TT, Xiang ZQ, Qiu T. Raman geobarometry of quartz inclusions in eclogitic garnet constrains the metamorphic evolution and exhumation of the Sumdo oceanic slab in Tibet. Gondwana Res. 2022;104:172–84.Search in Google Scholar

[21] Manjunath GL. Estimating stress percolation patterns in hydraulic fractured Gondwana coal using Raman spectroscopy. J Nat Gas Sci Eng. 2022;100:104469.Search in Google Scholar

[22] Xie C, Zhou BG, Du JG, Yi L, Chen ZW. Characteristics of Raman spectra of minerals in gouge of Wenchuan earthquake fault zone. Spectrosc Spectr Anal. 2013;33(6):1562–5 (in Chinese with English Abstract).Search in Google Scholar

[23] Wang XE, Lin J, Cheng J. Introduction to earth science. Beijing: Geological Publishing House; 1999 (in Chinese).Search in Google Scholar

[24] d’Alessio M, Martel SJ. Development of strike-slip faults from dikes, Sequoia National Park, California. J Struct Geol. 2005;27(1):35–49.Search in Google Scholar

[25] Zhang GS, Wen HJ, Hu RZ, Qiu YZ, Yang GS, Yu WX. Geological implications of tectonic stress field formed in mafic dikes in southeastern Fujian. Geotectonica et Metallogenia. 2006;30:142–8 (in Chinese with English Abstract).Search in Google Scholar

[26] Christian S, Ziemann MA. In-situ Raman spectroscopy of quartz: a pressure sensor for hydrothermal diamond-anvil cell experiments at elevated temperatures. Am Mineral. 2000;85:1725–34.Search in Google Scholar

[27] Zheng HF, Sun Q, Zhao J, Duan TY. Comment on the pressure gauge for the experiments at high temperature and high pressure with DAC. Chin J High Press Phys. 2004;18(1):78–82 (in Chinese with English Abstract).Search in Google Scholar

[28] Xie C, Zhou BG, Liu L, Zhou XC, Yi L. Characteristics of Raman spectra of minerals in the veins of Wenchuan earthquake fault zone. Spectrosc Spectr Anal. 2015;35:118–21 (in Chinese with English Abstract).Search in Google Scholar

[29] Ran YK, Xu XW, Chen WS, Chen LC, Dong SP, Wang H. Paleo earthquake and large earthquakes recurrence interval along Yingxiu-Beichuan fault of Longmenshan fault zone, Sichuan, China. Recent Dev World Seismolog. 2010;6:28–9.Search in Google Scholar

[30] Tapponnier P, Xu ZQ, Roger F, Meyer B, Arnaud N, Wittlinger G, et al. Oblique stepwise rise and growth of the Tibet Plateau. Science. 2001;5547(23):1671–7.Search in Google Scholar

[31] Xu XW, Wen XZ, Yu GH. Co-seismic reverse and oblique slip surface faulting generated by the 2008 Mw 7.9 Wenchuan earthquake, China. Geology. 2009;37(6):515–8.Search in Google Scholar

[32] Wang XN. The current movement characters of main fault zones surrounding the Namcha Barwa Syntaxis. PhD thesis, Institute of Geophysics, China Earthquake Administration, Beijing; 2018 (in Chinese with English Abstract).Search in Google Scholar

[33] Xie C. A study on tectonic geomorphology of Namche Barwa and activity of the faults. PhD thesis, Institution of Geology, China Earthquake Administration, Beijing; 2018 (in Chinese with English Abstract).Search in Google Scholar

[34] Liang JR, Hu M, Kang YL, Lei ZK, Dou YW. Study on residual in porous silicon. Transducer Microsyst Technol. 2007;26:15–9 (in Chinese with English Abstract).Search in Google Scholar

[35] Peng H, Ma XM, Jiang JJ. Stability and stress measurement near the Qingchuan fault in the northern Longmen mountains. J Geomech. 2009;15(2):114–30.Search in Google Scholar

[36] Xu XW, Yu GH, Chen GH, Ran YK, Li CX, Chen Y, et al. Parameters of coseismic reverse- and oblique-slip surface rupture of the 2008 Wenchuan earthquakeEastern Tibetan Plateau. Acta Geologica Sin. 2010;83(4):673–84.Search in Google Scholar

[37] Li K, Xu XW, Kirby E, Tang FT, Kang WJ. Late quaternary paleoseismology of the Milin fault: implications for active tectonics along the Yarlung Zangbo Suture, Southeastern Tibet Plateau. Tectonophysics. 2017;29:64–72.Search in Google Scholar

[38] Ran YK, Duan RT, Deng QD, Jiao DC, Min W. 3-D trench excavation and paleoseismology at Gaowanzi of the Haiyuan fault. Seismol Geol. 1997;19(2):97–107 (in Chinese with English Abstract).Search in Google Scholar

[39] Liu-Zeng J, Klinger Y, Xu X, Lasserre C, Chen G, Chen W, et al. Millennial recurrence of large earthquakes on the Haiyuan fault near Songshan, Gansu Province, China. Bull Seismol Am. 2007;97:14–34.Search in Google Scholar

Received: 2022-06-24

Revised: 2023-05-01

Accepted: 2023-05-03

Published Online: 2023-07-10

This work is licensed under the Creative Commons Attribution 4.0 International License.

Articles in the same Issue

https://doi.org/10.1515/geo-2022-0486

Keywords for this article

Yinxiu-Beichuan fault; Milin fault; Raman peak of quartz; shift of the 464 cm−1 peak; crystal structure; thrust fault activity

Creative Commons

BY 4.0