New Discoveries in the Maijishan Grottoes: Identification of Blue-Green Pigments and Insights into Green Pigment Application Techniques

Abstract

The application techniques and composition of green and blue-green pigments in the Maijishan Grottoes were explored by utilizing microscopic observation, Raman spectroscopy, and SEM-EDX analysis. For the first time, lavendulan and high-purity botallackite were identified in these grottoes, in addition to the commonly found malachite and atacamite. These discoveries suggest that several caves in the Maijishan Grottoes were originally painted in blue-green tones, which have since altered to the current green or dark green hues. It was also revealed that the application of green mixed pigments involved layering malachite over basic copper chloride, rather than blending them together. Moreover, variations in the composition and placement of white ash layers indicate that the use of mixed pigments was likely due to repainting rather than initial decorative purposes. These findings significantly enhance our understanding of ancient painting techniques and provide crucial data for the conservation and restoration of cultural heritage in the Maijishan Grottoes.

1. Introduction

The Maijishan Grottoes, established during the Later Qin period (AD 384–417) of the Sixteen Kingdoms (AD 304–439), have been built and renovated by over ten dynasties, including the Northern Wei (AD 386–534) and Western Wei (AD 535–556), resulting in a rich history and cultural integration [1,2]. Preserving a vast array of statues, predominantly clay sculptures complemented by exquisite murals (Figure 1), the grottoes have earned the reputation of the ’Oriental Sculpture Exhibition Hall’ [3,4]. The Maijishan Grottoes offer priceless tangible proof of early Chinese painted sculpture and mural techniques, as well as the history of science and technology, and are of great significance when exploring the study of mural paintings in other grottoes along the Silk Road [5].

The craftsmanship and materials used in the Maijishan Grottoes’ painted sculptures and murals have consistently been focus of academic research [6,7]. Wen [8] identified the employment of red pigments, including minium (Pb₃O₄), cinnabar (HgS), and ochre (Fe₂O₃), with a white calcite (CaCO₃) layer in Cave 131 of the Maijishan Grottoes. However, the composition of green pigments was not analyzed. The utilization of green pigments in the Maijishan Grottoes is notably intricate and warrants further investigation. Zhou [9] employed the X-ray diffraction technique to analyze the composition of 103 samples of mural and painted sculpture pigments from 14 caves dating from the Later Qin to the Qing Dynasties (AD 1644–1912) and conducted a comparative study with pigments utilized in the Mogao Grottoes. The research indicated that basic copper chloride (Cu₂(OH)₂Cl₂) was predominant in both the Mogao Grottoes and Maijishan Grottoes, comprising almost half of the green pigments [10,11,12,13], with hydrated basic copper chloride (Cu₂(OH)₃Cl·1.5H₂O) being found in Ming Dynasty (AD 1368–1644) grottoes at both sites [14]. Malachite (Cu₂(OH)₂CO₃) was detected in Caves 5 and 110 during preliminary research investigations. Studies have shown that the green pigments used in painted sculptures and murals in the Maijishan Grottoes are used either alone or in mixtures [15,16]. Early green pigments were dominated by malachite, followed by a mixture of malachite and atacamite, whereas green pigments in caves from the Ming and Qing Dynasties were dominated by atacamite, emphasizing the complexity of green pigment usage in the Maijishan Grottoes.

Recent research indicates that most green pigments used in the Maijishan Grottoes are copper-containing mineral pigments, comprising basic copper carbonate and basic copper chloride [9]. Green basic copper carbonate, also known as malachite, is commonly found in pigments used in cultural artifacts, including murals [17,18,19,20] and painted sculptures [21]. It often coexists with azurite (Cu₃(CO₃)₂(OH)₂), and these two basic copper carbonates can be interconverted. Basic copper chloride, previously referred to as patina or atacamite in early studies, is only one of the isomers of basic copper chloride [22]. In addition to atacamite, other isomers include botallackite, paratacamite, and clinoatacamite [23,24]. Variations in the crystal structures among these isomers lead to differences in stability, color, particle size, and other properties [25]. Early studies on Maijishan mural pigments often failed to fully elucidate the use and chemical composition of green pigments, being hindered by the scientific and technological limitations of that time [26,27]. For instance, the use of mixed pigments like malachite and atacamite remained unclear; also, the composition of blue-green pigments excluding malachite, lapis lazuli ((Na,Ca)₈(AlSiO₄)₆(SO₄,S,Cl)₂), azurite, and atacamite is unidentified [28]. To address these issues, selected pigments from specific grotto murals and painted sculptures are analyzed in this study. We aimed to elucidate the layering and composition of these pigments by employing a range of analytical methods, thereby precisely determining the varieties and application techniques used in the Maijishan Grottoes.

2. Materials and Methods

2.1. Painting Samples Investigated in This Study

After conducting a thorough investigation of all the caves, samples of cultural relics containing green mixed pigments and those differing from typical blue-green pigments were selected, specifically from Caves 48, 49, 78, and 120. The sampling locations and detailed information are presented in

Table 1. Sampling location and appearance of samples.

Sample Source	Cave Info.	No.	Sampling Location	Sample Type and Color
Cave 48	The Northern Zhou Dynasty (AD 557–581)	48		Dark green and light green pigments
Cave 49	Sui Dynasty (AD 581–618)	49-1		Blue-green pigment
Cave 49	Sui Dynasty	49-2		Blue-green pigment
Cave 78	Later Qin Dynast (the Northern Wei Dynasty rebuilt)	78		Blue-green pigment
Cave 120	The West Wei Dynasty	120		Dark green and light green pigments

.

2.2. Analytical Methodology

Given the limited number of samples, our comprehensive analysis faced challenges. The samples were analyzed using the following analytical approaches:

A digital VHX-5000 microscope (Keyence, Osaka, Japan) was utilized to observe surface morphology and cross-sectional layer structures after embedding. This system offers continuous magnification from 0 to 2500× and integrates 2D and 3D image measurement modes, which are essential for visualizing the micro-characteristics of pigments.

A laser-confocal microscope Raman spectrometer Alpha300R (WITec, Ulm, Germany) was employed. The parameters of experiment were as follows: laser wavelength: 532 nm; spectrometer focal length: 300 mm, f/4; luminous flux: 70%; grating: 600 g/mm and 1800 g/mm; EMCCD: 1600 × 200 back-illuminated deep-cooled electron-multiplying spectral CCD with a Zeiss confocal Raman microscope, offering a spatial resolution of up to 200 nm for non-destructive acquisition of chemical information gathering [29].

A Phenom XL system (Thermo Fisher Scientific, Eindhoven, The Netherlands) equipped with an EDS system (Amptek Fast SDD X123) was utilized, and the analysis was conducted in low vacuum mode (20 Pa). Cross-sectional backscattered electron (BSE) images were obtained at a 15 kV accelerating voltage (with a beam current of 2.8 nA) and a 7.0 mm working distance was used to examine the layer relationships and elemental distribution within the sample section [30].

3. Results and Discussion

3.1. The Mixed Samples

3.1.1. Analysis Results for Cave 48

Sample 48, derived from the green pigment on the outer-flared short skirt of a clay sculpture situated between the two niches in Cave 48, exhibits notable pigment stratification that is clearly distinct (Figure 2a). Microscopic examination (Figure 2b,d) reveals a least five layered structures within the sample (the right side of the cross-section includes the upper dark green pigment layer and white ash layer, detached at the right during sample preparation): a dark green pigment layer at the top, followed by a white ash layer, a light green pigment layer, another white ash layer, and finally the ground layer. Raman spectroscopy measurements (Figure 2c) indicate that the top dark green pigment layer has significant characteristic peaks at 114 cm⁻¹, 152 cm⁻¹, 180 cm⁻¹, 220 cm⁻¹, 433 cm⁻¹, 1005 cm⁻¹, 1495 cm⁻¹, 3348 cm⁻¹, 3383 cm⁻¹, and 3437 cm⁻¹. The peak at 1005 cm⁻¹, associated with the symmetric stretching vibration of the sulfate ion, indicates the presence of calcium sulfate (CaSO₄) [31]. Meanwhile, the peaks at 3348 cm⁻¹ and 3437 cm⁻¹, which correspond to the stretching vibrations of OH groups, suggest the presence of atacamite [32]. The other peaks are characteristic of malachite, suggesting that the upper dark green pigment is primarily malachite, containing calcium sulfate and atacamite [33]. The third light green pigment layer shows characteristic peaks at 1007 cm⁻¹, 3353 cm⁻¹, and 3438 cm⁻¹, implying the presence of atacamite and calcium sulfate. The second white ash layer exhibits a distinct peak at 1006 cm⁻¹, confirming its composition as calcium sulfate. The fourth layer has additional characteristic peaks at 896 cm⁻¹, 1463 cm⁻¹, 1490 cm⁻¹, and 1628 cm⁻¹ when compared to the second layer, indicating the presence of calcium oxalate along with calcium sulfate in the white ash layer [34], the C=O stretching vibration peak at 1490 cm⁻¹ suggests that the calcium oxalate exists in the form of whewellite (CaC₂O₄·H₂O) [35]. To elucidate the relationships between the layers within the sample, SEM-EDX analyses were performed on the sample’s cross-section. The results (Figure 2e–g) demonstrate that the Cl and Cu elements do not completely overlap and that there is a significant difference in the thickness of the two green pigment layers; the S and Ca elements exhibit some coexistence in the second white ash layer, implying that this layer consists of calcium sulfate. The fourth white ash layer and the ground layer also contain substantial amounts of Ca, further corroborating the existence of calcium oxalate (Figure S1). Considering the region’s humid environment and abundant microbial life, the calcium oxalate is likely a byproduct of microbial activity. The findings suggest that the two green pigments in Cave 48 were applied sequentially rather than combined, with malachite as the upper layer and atacamite as the lower layer, and both white ash layers are gypsum. The presence of two white ash layers indicates that the pigment application was not a deliberate technique but rather a result of repainting.

3.1.2. Analysis Results for Cave 120

Sample 120 was obtained from the murals in Cave 120. Microscopic observations (Figure 3a) revealed that the green pigment layers were not mixed as previously thought, but exhibited distinct stratification, with a darker upper layer and a lighter lower layer, indicating the use of two green pigments with different compositions. The Raman spectroscopy analysis results (Figure 3b) indicated that the Raman spectra of the light green and dark green pigments were largely consistent, with characteristic peaks present at 145 cm⁻¹, 180 cm⁻¹, 218 cm⁻¹, 269 cm⁻¹, 355 cm⁻¹, 433 cm⁻¹, 506 cm⁻¹, 535 cm⁻¹, 894 cm⁻¹, 1006 cm⁻¹, 1059 cm⁻¹, 1097 cm⁻¹, 1460 cm⁻¹, 1490 cm⁻¹, 1627 cm⁻¹, 3323 cm⁻¹, 3381 cm⁻¹, and 3438 cm⁻¹. The peaks at 3323 cm⁻¹ and 3438 cm⁻¹ are attributed to atacamite; 1006 cm⁻¹ is characteristic of calcium sulfate; 894 cm⁻¹, 1460 cm⁻¹, 1490 cm⁻¹, and 1627 cm⁻¹ are associated with calcium oxalate (whewellite); while the remaining peaks are characteristic of malachite. Microscopic observations and SEM-EDX were employed to clarify the layer structure and elemental distribution of the sample. The cross-section analysis (Figure 3c) reveals that the sample consists of three separate layers: the upper layer is a dark green pigment, the middle layer is a light green pigment, and the lower layer is the ground layer, lacking a distinct white ash layer. The scanning electron microscopy data (Figure 3d–f) demonstrate the presence of elements including Cu, Cl, Ca, and S within the sample. The detection of Ca and S elements suggests that the white layer is calcium sulfate (Figure S2). Microscopic observation did not show any distinct white ash layer, indicating that it may have migrated with water and mixed with the pigment layers. Moreover, the distribution of Cu and Cl elements is not totally coincidental, with Cu concentrations markedly exceeding those of Cl, particularly evident in the top dark green pigment layer. The analytical results from Cave 120 are similar to those from Cave 48, confirming that the superimposition of the two green pigment layers resulted from repainting rather than the mixture of two different green pigments.

3.2. Blue-Green Pigment Samples

3.2.1. Analysis Results for Cave 49

Sample 49-1 was collected from the left armpit of the Buddha in the niche on the primary wall of Cave 49, and its visual appearance to the naked eye is a composite hue of blue and yellow. Microscopic observations indicated that the surface of sample 49-1 consists of a yellowish-white ash layer, beneath which lies the blue-green pigment (Figure 4a). The Raman spectroscopy analysis results (Figure 4b) reveal that the yellowish-white ash layer on the surface exhibits characteristic absorption peaks near 241 cm⁻¹, 305 cm⁻¹, 399 cm⁻¹, 488 cm⁻¹, 1005 cm⁻¹, and 3404 cm⁻¹, with the absorption peak at 1005 cm⁻¹ corresponding to calcium sulfate, and the characteristic peaks at 241 cm⁻¹, 305 cm⁻¹, 399 cm⁻¹, and 488 cm⁻¹ indicating the possible presence of goethite (α-FeOOH) in the white ash layer [36]; the blue-green pigment has absorption peaks near 410 cm⁻¹, 493 cm⁻¹, 616 cm⁻¹, 669 cm⁻¹, 1005 cm⁻¹, 1135 cm⁻¹, 3351 cm⁻¹, 3438 cm⁻¹, 3407 cm⁻¹, and 3497 cm⁻¹, suggesting that it is a composite of botallackite (3497 cm⁻¹) and atacamite (3351 cm⁻¹, 3438 cm⁻¹), with the other characteristic peaks being similar to those of the yellowish-white ash layer [37,38]. The sample’s stratification was evaluated using microscopic observations and SEM-EDX. The cross-section reveals that the sample comprises five layers (Figure 4c): the uppermost layer is the white ash layer, succeeded by the second blue-green pigment layer, a further white ash layer, the ground layer as the fourth layer, and the final layer is the blue-green pigment layer. Notably, the thicknesses of the two white ash layers and the two pigment layers exhibit significant similarities. The scanning electron microscopy results (Figure 4d–f) indicate the presence of elements such as Cu, Al, Si, S, Cl, and Ca in the sample, with Cu and Cl elements being largely coincident and exclusively dispersed within the blue-green pigment layers, further confirming the identification of the blue-green pigment as basic copper chloride; the Ca and S elements partially overlap, suggesting the presence of other components containing Ca alongside calcium sulfate. Based on the analytical results of the sample from Cave 120, we infer that the sample also contains calcium oxalate; the presence of the Pb element is noted in the white ash layer, but no lead carbonate is detected in the Raman results, so its origin cannot be determined. The Pb element is only found in the white ash layer, suggesting it may be related to the layer’s production process or repainting of the statue (Figure S3).

Sample 49-2, also extracted from the left armpit of the Buddha in the niche on the main wall, showed a superimposed relationship with sample 49-1. Microscopic observation reveals that the surface of this specimen exhibits a blue-green tint (Figure 5a), identical to the final layer of sample 49-1. Beneath it, the second layer is a white ash layer, followed by the third layer being the ground layer, and under that lies an orange-yellow pigment (Figure 5b). The Raman spectroscopy analysis shows that the Raman spectra of the blue-green pigment are essentially uniform across all locations, with characteristic peaks at 156 cm⁻¹, 398 cm⁻¹, 449 cm⁻¹, 498 cm⁻¹, 895 cm⁻¹, 3352 cm⁻¹, 3425 cm⁻¹, and 3509 cm⁻¹ (Figure 5c). Based on the peaks between 300 cm⁻¹ and 500 cm⁻¹, along with the peak at 3509 cm⁻¹, it suggests that the blue-green pigment is primarily high-purity botallackite with a negligible presence of atacamite [37,38]. The orange-yellow pigment shows characteristic peaks at 136 cm⁻¹ and 264 cm⁻¹ (Figure 5d), indicating the presence of a lead oxide, and during the Raman analysis, laser-induced heating destabilized it, causing peak shifts that confirmed its identification as lead yellow (massicot, PbO) [39]. The analysis results show that the pigments on the Buddha statue in Cave 49 display a multi-layered structure: The outermost layer of pigments may have detached due to weathering and other factors, leaving a visible top layer composed of a calcium sulfate-based white ash layer. Given the presence of goethite detected within the white ash layer, we infer that the surface initially included iron-containing pigments. Beneath the white ash layer lies a relatively rare and high-purity layer of botallackite. The lowest layer comprises yellow lead pigments. Considering the multiple occurrences of the white ash layer, it is inferred that the statue may have undergone repainting.

The Maijishan Grottoes have yielded the initial discovery of botallackite, which is crucial for elucidating the ancient pigment preparation techniques. The formation of basic copper chloride minerals, such as atacamite, requires particular environmental conditions, resulting in their scarcity in natural environments. Botallackite, as a polymorph of basic copper chloride, is exceedingly rare. It predominantly forms in arid or hyperchlorine environments and, given certain conditions, can transform into paratacamite or further recrystallize into clinoatacamite. This instability hinders the existence of these minerals in substantial numbers and with high purity in the natural environment [40]. High-purity botallackite requires controlled reaction and preservation conditions for its synthesis and utilization. Recent studies indicates that high-purity botallackite can be synthesized under high chloride ion concentrations by regulating temperature and reaction duration [41]; furthermore, the thermodynamic stability of botallackite is lower than that of other prevalent copper-containing pigments [42], rendering it vulnerable to transformation influenced by factors such as temperature, moisture, and chloride ions [43]. Analysis revealed that the botallackite purity in the upper layer is extremely high, with only a little quantity of atacamite present. No other copper-containing pigments were detected in the vicinity. This suggests that the botallackite found is likely an early synthetic pigment utilized by humans, rather than a natural mineral or the degradation product of other copper-containing pigments. The higher content of atacamite in the upper layer compared to the lower layer is attributed to a greater degree of transformation in the upper layer due to exposure to the external environment, leading to variations in atacamite content across different strata within the sample. Nonetheless, the reasons for the prolonged stability of the unstable botallackite and its transformation warrant further investigation. Therefore, based on the formation conditions, stability, and analytical findings for botallackite, it is inferred that ancient craftsmen possessed the ability to produce alkaline copper chloride pigments by optimizing specific conditions, such as increasing salt content to establish a high-chlorine environment. The reaction process was regulated based on the color of the reaction products, ultimately synthesizing botallackite with a blue-green hue for decorative purpose in specific areas and conveying particular symbolic significances.

3.2.2. Analysis Results for Cave 78

Sample 78, derived from the left hem of the Buddha’s robe on the principal wall of Cave 78, appeared blue-green in color. Microscopic observation indicates that the sample consists of three layers in total: the lowest layer is the ground layer, and the pigment layer has two distinct strata (Figure 6a); the upper layer is a dark green pigment, and the lower layer is a light blue pigment. Raman spectroscopy analysis reveals that the light blue pigment exhibits characteristic peaks at 121 cm⁻¹, 177 cm⁻¹, 225 cm⁻¹, 545 cm⁻¹, 635 cm⁻¹, 857 cm⁻¹, 882 cm⁻¹, and 1007 cm⁻¹ (Figure 6b), with the peak at 857 cm⁻¹ being attributed to the symmetric and asymmetric stretching vibrations of AsO₄³⁻. Comparisons between databases validate the identification of the light blue pigment as lavendulan [44]. The dark green pigment shows characteristic peaks at 117 cm⁻¹, 152 cm⁻¹, 179 cm⁻¹, 217 cm⁻¹, 349 cm⁻¹, 435 cm⁻¹, 511 cm⁻¹, 639 cm⁻¹, 1008 cm⁻¹, 1092 cm⁻¹, 1365 cm⁻¹, 1492 cm⁻¹, 3353 cm⁻¹, and 3439 cm⁻¹ (Figure 6c), with the peak at 1008 cm⁻¹ indicative of calcium sulfate, the peaks at 3353 cm⁻¹ and 3439 cm⁻¹ being characteristic of atacamite, and the remaining peaks being characteristic of malachite. Preliminary analysis indicates that the hues are superimposed greens pigments, with the upper layer comprising a blend of malachite and atacamite, while the lower layer consists of lavendulan (NaCaCu₅(AsO₄)₄Cl·5H₂O).

Microscopic observation and SEM-EDX analysis were subsequently conducted on the sample’s cross-section to elucidate its stratification. Microscopic observation (Figure 6d) reveals that the sample comprises four distinct layers in total: the uppermost layer is a dark green pigment layer, the second layer is a light blue pigment layer whose microstructure is needle-like, and the third layer is a white ash layer, with the fourth being the ground layer. The scanning electron microscopy results (Figure 6e,f) demonstrate the presence of elements such as Cu, As, Na, Cl, Ca, S, and Pb within the sample. The distribution of Ca and S elements is largely coincidental (Figure S4), indicating that calcium sulfate was used as the white ash layer in this area. The results show that the presence of calcium sulfate is not confined to the white ash layer but is dispersed throughout the entire sample, suggesting that calcium sulfate has migrated and mixed with the pigment layers. This also explains the whitish appearance of the upper blue-green pigment sample. The Pb element is distributed within the white ash layer (Figure S4), suggesting that white lead was used as the white ash layer, and the existence of two different compositions of the white ash layer also indicates that the sculpture has been repainted. The Cu and Cl elements are present in both pigment layers, with a higher concentration in the upper layer compared to the lower layer; the As element and trace amounts of Na element are primarily found in the lower blue-green pigment layer. In conjunction with Raman spectroscopy results, the pigment layers are identified as containing three copper-containing pigments: malachite, atacamite, and lavendulan.

The analysis indicates that the blue-green coloration of the pigment is due to a mixture of malachite (green), atacamite (green), and lavendulan (blue), with a distinct stratigraphic arrangement: the upper layer consists of malachite, the middle layer comprises atacamite, and the bottom layer is lavendulan. Lavendulan, a rare mineral pigment, was first identified in 1837 [45], and it is widely acknowledged that the majority of lavendulan found in cultural relics is the result of the aging of Paris green (Cu(C₂H₃O₂)₂·3Cu(AsO₂)₂) [46,47]. However, the high purity of lavendulan found at this site and no other pigments containing both Cu and As elements being detected in the vicinity suggests that it is not a degradation product of copper-arsenic pigments such as Paris green. Other researchers have found arsenic-containing pigments and their degradation products in the Maijishan Grottoes [48], and based on the micromorphology of the pigments, we hypothesize that lavendulan is formed by a reaction between basic copper chloride and arsenic-containing pigments [49]. Therefore, the green pigments in Cave 78 were also used in a superimposed way with malachite and atacamite. Over time, the underlying atacamite reacted with arsenic-containing substances to produce lavendulan, resulting in the overall blue-green appearance. The presence of two different compositions of the white ash layer suggests that the mixed pigments were not a decorative technique but rather an effect of repainting. The presence of high-purity lavendulan indicates that the reaction was thorough. Among the isomers of basic copper chloride, only botallackite is unstable and can undergo a complete transformation. This indirectly confirms that the early use of basic copper chloride in the Maijishan Grottoes was of unstable botallackite rather than the more stable atacamite. It can thus be inferred that ancient craftsmen had a proficient mastery of the preparation process of botallackite. The SEM-EDX analysis shows that the distribution of As is uniform and only present in lavendulan, which leads us to speculate that the source of As is related to the behavior of ancient people. For example, it could be due to the presence of As-containing impurities in the raw materials used in the pigment preparation process, or the use of As-containing substances during painting or repainting, rather than being a later contaminant. Further in-depth analysis is required to determine the exact source.

This comprehensive research indicates that the green pigments used in the Maijishan Grottoes consist of malachite, atacamite, botallackite, and lavendulan. The white ash layer is primarily composed of calcium sulfate; however, the presence of Pb in some grottoes’ white ash layers requires further analysis to determine its form. Moreover, the results reveal that green pigments were used both individually and in superimposed layers, rather than in the mixed manner described in earlier studies (see Table S1 for details).

4. Conclusions

This work employs microscopic observations, Raman spectroscopy, and SEM-EDX analysis to examine the techniques and composition of green mixed pigments and blue-green pigments in the Maijishan Grottoes. In the Maijishan Grottoes, the green pigments used include not only the commonly found malachite and atacamite but also lavendulan and high-purity botallackite, both identified for the first time in the grottoes. Notably, based on the crystal morphology, stability, and formation conditions of the two pigments, it can be proposed that lavendulan was produced from the reaction of basic copper chloride with arsenic-containing substances, whereas botallackite was synthesized by ancient people under controlled reaction conditions. These findings suggest that several caves in the Maijishan Grottoes were originally adorned with blue-green colors instead of the green or dark green hues seen now. Furthermore, the application of green mixed pigments in the Maijishan Grottoes involves superimposing malachite (upper layer) over basic copper chloride (lower layer), rather than amalgamating them; the differing compositions and placements of the white ash layers also indicate that the use of mixed pigments arose from repainting rather than decorative intent. These findings provide insight into pigment application techniques and color changes in the Maijishan Grottoes, greatly enhancing our comprehension of the evolution of ancient painting techniques and the conservation and restoration of cultural heritage.