A Multi-Method Archaeometric Investigation of Warring States to Han Dynasty Jades from Beijing: Materiality, Alteration, and Provenance

Abstract

Jade artefacts from the Warring States to Han periods (c. 5th century BCE – 2nd century CE) in the Beijing region constitute a critical yet understudied corpus for understanding cultural interaction along China’s ancient northern frontier. This study presents a comprehensive scientific reassessment of jades and associated materials from the Houtun cemetery, employing an integrated, non-destructive protocol including X-ray fluorescence (XRF), micro-XRF mapping, Raman spectroscopy, FTIR, and multispectral imaging. Our analyses first confirm a predominance of tremolite, yet also identify a diverse material spectrum including agate and a bone artefact, the latter exhibiting significant post-depositional metallic contamination. More critically, trace-element signatures (notably Ni vs. Fe/Mn) and submicroscopic fabrics reveal multiple geological sources for the nephrite. Even within a single ritual set, such as an Eastern Han-period jade facial covering, individual components could be traced to different raw materials or different parts of larger blocks, revealing a pragmatic, composite mode of assembling funerary assemblages. Elemental mapping and Raman data further clarify post-depositional processes, distinctly separating endogenous weathering (whitening via microstructural change) from exogenous deposition (brown Fe–Mn–Al-rich crusts and black carbonaceous films). Notably, the frequent occurrence of fragmented jades with fresh break surfaces suggests intentional ritual breakage prior to burial, particularly in non-elite contexts. Together, these findings demonstrate how communities in the Beijing frontier actively negotiated their cultural position through selective material use, flexible technological practices, and distinctive mortuary behaviours. This multi-method archaeometric study establishes a robust analytical framework for future research on jade circulation and ritual practice in ancient cultural borderlands.

Graphical abstract

Keywords

Jade / Beijing archaeology / Non-destructive analysis / Surface alteration / Archaeometry

1 Introduction

The archaeological study of jade artefacts has long served as a vital lens for understanding the social complexity, ritual practices, and intercultural dynamics of early China (Barnes 2018). Yet despite this broad scholarly importance, jade assemblages dating to the Warring States (c. 475–221 BCE) and Han (206 BCE–220 CE) periods from the Beijing region remain comparatively underexplored, particularly from a scientific and materials-oriented perspective. Historically situated at the core of the Yan state and later the Guangyang Kingdom/Commandery, the Beijing region functioned as a politically volatile and culturally dynamic frontier, where Central Plains traditions intersected with northern steppe societies through sustained exchange, migration, and conflict (Zhou 2008). Within this context, jade artefacts constitute one of the most sensitive material proxies for tracing long-term continuities and transformations in cultural affiliation, resource selection, and technological practice. A diachronic examination of jade materials and assemblage composition is therefore essential for assessing how local traditions interacted with, absorbed, or reconfigured influences from the Central Plains and the northern cultural sphere across successive periods.

Archaeological evidence indicates that the use of tremolite in the Beijing region can be traced back to the Shang and Western Zhou periods, with representative finds from sites such as the Shang tombs at Liujiahe in Pinggu, Western Zhou Yan cemeteries, and burial grounds at Baifu in Changping and the Jundushan area in Yanqing (Yuan and Zhang 1977, Zhang and Hu 2017, Beijing Institute of Archaeology 1995, 2007). Jade materials from this early phase include tremolite (nephrite), turquoise, agate, talc, and various stone beads. These assemblages reflect both the transmission of Central Plains ritual jade traditions and the strong imprint of northern steppe cultural aesthetics, consistent with the broader Shanrong cultural milieu (Qi 2002). During the Warring States period, jade finds in the Beijing area are comparatively sparse, yet recent discoveries in the Tongzhou district suggest that the region occupied a more active position within contemporary jade networks than previously recognized. Jade production and consumption reached a clear peak in the Han dynasty, exemplified by high-status assemblages from the Dabaotai and Laoshan tombs, where finely crafted tremolite jades attest to the firm integration of Central Plains mortuary and ritual norms among local elites (Wang and Cheng 2002, Dabaotai Han Tomb Excavation Team and IACASS 1989).

Scholarly research on Warring States–Han jade in the Beijing region has long been dominated by typological studies and archaeological reporting, with complementary discussions of ritual function, cultural affiliation, and symbolic association (Gu 2005, Yu 2002). Only in recent years have scientific and multidisciplinary approaches begun to play a more substantive role in addressing questions of raw-material selection, manufacturing technology, and provenance. Recent work has demonstrated the value of such approaches, for example through integrated analyses of turquoise ornaments from the Xingong site, which have explored material characteristics, production techniques, and potential sources, thereby providing a new perspective on material connectivity between northern settlements and wider regional networks during the early state period (Yang et al. 2025, Chen et al. 2025). To date, the most substantial body of scientific data for Warring States–Han jade in Beijing derives from the appendix to The Houtun Cemetery (Warring States volume), which reports analytical results for 21 samples, including beads, tubular ornaments, jade pendants, and rings. The materials identified encompass quartz, amazonite, talc, enstatite, tremolite, and sandstone, substantially expanding the recognized material spectrum of local jade use (Beijing Institute of Archaeology 2024). Particularly noteworthy is the first secure identification of amazonite beads, whose close morphological resemblance to turquoise ornaments highlights the mediating role of northern steppe cultures in the circulation of jade-like materials. Comparable materials documented in northeast China and eastern Inner Mongolia further underscore Beijing’s position within broader northern interaction networks. In parallel, microscopic studies of jade artefacts from the Dabaotai Han tombs have revealed the use of fine steel rotary tools, characterized by thin cutting edges, small diameters, and high rotational speeds, attesting to a high level of technical proficiency in jade working during the Han period (Zhao et al. 2019).

Building on recent advances, it remains evident that research on Beijing’s Warring States–Han jades has been disproportionately focused on elite, often royal, contexts, while assemblages from middle- and lower-status tombs have rarely been subjected to comparable scientific scrutiny. This imbalance has produced a structurally biased understanding of jade’s social distribution, consumption practices, and functional diversity. To address these interrelated issues, this paper proposes a multi-phase research framework designed to systematically advance the study of regional jade assemblages. The initial phase establishes a regionally integrated scientific database based on large-scale, non-destructive analyses using portable X-ray fluorescence (pXRF), Raman spectroscopy, FTIR, and multispectral imaging. This approach aims not only to refine material identification and mineralogical baselines, but also to resolve a series of outstanding problems, including the holistic assessment of jade face coverings, the question of multi-source tremolite, variations in colour and patination, the distinction between use-life damage, deliberate destruction, and burial-related alteration, and the separation of pre-burial technological features from post-burial secondary transformations, thereby providing a more balanced and analytically grounded perspective on Warring States–Han jade use in the Beijing region.

2 Samples and Background

The Houtun cemetery is located in Lucheng Town, Tongzhou District, Beijing, approximately 850 meters south of the Han-period Luxian Ancient City, one of the “Top Ten Archaeological Discoveries of China in 2016”, and separated from it by the modern Yunchaojian River. It is a multi-period burial ground containing graves dating from the Warring States period through the Ming–Qing dynasties.

The Warring States burials form a large, well-organized group characterized by orderly arrangement, consistent orientation, and minimal intercutting, suggesting a managed cemetery used for clan-based interment. A total of 202 late Warring States graves were identified, all rectangular vertical shaft pit tombs of modest scale, most with extended supine burials and a predominantly northward orientation. The spatial distribution of grave goods further indicates internal differentiation, with assemblages reflecting strong Yan cultural traits concentrated in the central–eastern sector, while burials displaying Central Plains traditions are mainly located in the western part of the cemetery. Occasional graves combining both assemblage types point to cultural interaction within the community and suggest prolonged use of the cemetery over time (Liu et al. 2020). However, materials from the Western and Eastern Han periods excavated at Houtun have not yet been fully published.

The present study examines a total of 14 samples (or sample groups) derived from ten tombs, spanning the Warring States period to the Eastern Han dynasty (Fig. 1). The analyzed artefacts include an Eastern Han jade facial covering set, as well as jade bi discs, pendants, huang, and sword ornaments dating to the Warring States and Western Han periods. Detailed information on the samples is provided in Table 1. To ensure the safety and preservation of the artefacts, analytical measurements were conducted only after gently cleaning selected areas with a small amount of ethanol.

3 Analytical methods

3.1 FTIR

The infrared spectra were acquired using Agilent 4300 (Agilent Technologies, CA) handheld portable FTIR spectrometer. No additional sampling was required; instead, samples were firmly pressed onto the sample stage, and the angle was adjusted to optimize detection signals. Adequate contact was confirmed by observing the expected shape of the mineral spectrum in real-time on the connected computer. Spectra were recorded in the range of 4000 to 650 cm⁻¹ with 32 co-added background scans every 15 min and 32 co-added sample scans in diffuse reflectance mode (DRIFTS) with a resolution of 4 cm⁻¹. The spectra are reported in Reflectance units. The data were then analyzed by OMNIC 8.0 software.

3.2 Raman

All samples were analyzed in situ using a portable Raman spectroscopy setup based on a WITec Alpha Cart system (WITec, Ulm, Germany). The system is equipped with a 532 nm air-cooled solid-state laser, a low-noise CCD detector (Andor DU401A-BV-352), and an inverted Nikon Ti-E microscope coupled with a Nikon A1 confocal system (Nikon, Tokyo, Japan). Measurements were conducted using a fiber-optic Raman probe fitted with a Zeiss air objective (10× magnification, NA = 0.23; working distance 11.1 mm) and a spectrograph with a 600 lines/mm grating. High-quality spectra were collected from multiple locations on each sample at a maximum laser power of 28 mW, with each spectrum acquired using 15 accumulations and an integration time of 2 s.

3.3 Elemental analyses

All samples were first analyzed in situ using portable X-ray fluorescence (pXRF), with three to four measurement spots selected on each artefact to account for surface heterogeneity. Analyses were carried out using a Thermo Niton XL3t 950 GOLDD+ spectrometer equipped with a silver (Ag) anode X-ray tube, a geometrically optimised large-area silicon drift detector (SDD), and a protective ultra-thin polymer window. Measurements were mainly conducted in “Soil” mode, which is effective for trace-element detection. The instrument was operated at a voltage of 50 kV and a current of 100 μA, with a detection window diameter of 8 mm and an analysis spot diameter of approximately 3 mm. For each analysis, the total acquisition time was set to 120 s, comprising 40 s for each of the Main, High, and Low filter conditions (Table S1,2).

Owing to time constraints associated with the loan of the artefacts, only a subset of samples (M136:18, M563:1, M564:1, M736:1, and M1002:1) was further analyzed by bench-top energy-dispersive X-ray fluorescence (EDXRF). Measurements were conducted using an EDAX Eagle III XLL μ-Probe spectrometer in the super-large sample laboratory at the Institute of High Energy Physics, Chinese Academy of Sciences (Beijing). The instrument is equipped with a molybdenum X-ray tube, a 125 μm beryllium window, and a silicon–lithium detector with an energy resolution of 160.3 eV. Analyses were performed under vacuum to enhance the detection of low-Z elements, using a primary beam spot of 1 mm diameter, an operating voltage of 40 kV, and a current of 250 mA. Instrument calibration was carried out using laboratory-developed ceramic standard reference materials to ensure analytical accuracy.

Elemental distribution mapping was performed on samples M564:1, M624:1, and M1002:1 using non-invasive quantitative μ-EDXRF analysis with a Bruker M4 TORNADO spectrometer. The system is equipped with a rhodium X-ray tube, polycapillary optics providing a spot size of approximately 30 µm, and a Peltier-cooled silicon drift detector with an active area of 30 mm². Measurements were conducted on selected smooth surfaces at an accelerating voltage of 50 kV and a current of 600 µA, with a pixel size of 30 µm and an acquisition time of 8 ms per pixel. Quantitative elemental maps were generated by correcting the raw data using the Bruker M4 TORNADO software.

3.4 Multispectral imaging

To resolve sub-microstructural features of the samples, multispectral imaging was conducted using a CRi Nuance imaging system. Images were acquired across a wavelength range of 450–800 nm at 10 nm intervals, encompassing both the visible (VIS) and near-infrared (NIR) regions. A transmission-based illumination configuration was employed, with an LED light source (80 CRI, 4000 K, 1600 lumens) positioned beneath the samples to ensure homogeneous illumination while minimizing thermal exposure and preserving the integrity of the archaeological materials. Imaging was performed within an enclosed dark environment to eliminate stray light, and exposure parameters were preset with automatic adjustment to optimize image quality.

4 Results and Discussion

4.1 Material identification

Despite the limited spectral resolution of the portable FTIR instrument, the data are adequate for preliminary material classification. The samples can be divided into three groups: predominantly nephrite, a single agate sample, and one sample showing phosphate-related features, tentatively interpreted as bone-derived (Fig. 2).

Nephrite, a typical chain silicate mineral belonging to the amphibole group, exhibits infrared absorption features dominated by Si–O-related vibrational modes (Fig. 2). Across the wavenumber range of 4000–650 cm⁻¹, a consistent set of absorption bands is observed at approximately 1090, 1040, 995, 918, 758, 682, and 661 cm⁻¹. Among these, the most intense absorption region lies between 1200 and 900 cm⁻¹, corresponding to the stretching vibrations of Si–O, Si–O–Si, and O–Si–O linkages within the silicate chains. In addition, sharp peaks near 758 and 682 cm⁻¹, together with a narrow band at ~661 cm⁻¹, are particularly diagnostic and are commonly attributed to Si–O and Si–O–Si bending and stretching modes characteristic of tremolite–actinolite series minerals (Tan et al. 2013). Nevertheless, it should be noted that differences do exist among individual tremolite samples. For example, characteristic absorption bands appear at 1140 or 1152 cm⁻¹ in samples M564:1 and M1002:1. Such variations are typically associated with increased Fe²⁺ content or local lattice distortion, which tend to shift the Si–O stretching bands toward lower wavenumbers, whereas Mg-rich tremolite with a higher degree of structural order generally exhibits absorption at slightly higher frequencies (Ren et al. 2019).

In contrast, the agate sample (M627:4) displays consistent with agate rather than any non-siliceous major phase. The strong band at ~1086 cm⁻¹ corresponds to the asymmetric stretching of Si–O–Si and represents a stable diagnostic feature of chalcedony and agate, commonly broader than in macrocrystalline quartz due to microcrystalline textures and internal strain. The closely spaced doublet at 797 and 779 cm⁻¹ is a well-established fingerprint of α-quartz, indicating that despite its cryptocrystalline nature, agate remains structurally dominated by low-quartz. The bands at 693 and 672 cm⁻¹ are attributed to Si–O–Si bending and lattice vibrations associated with small grain size and structural disorder. A weak band at 1216 cm⁻¹, though not ubiquitous, can be ascribed to high-frequency vibrations involving non-bridging oxygens related to structural defects or trace impurities (Flörke et al. 1991).

The FTIR spectrum of the sample M1045:3, is dominated by a strong absorption at 1035 cm⁻¹, corresponding to the ν₃ asymmetric stretching of PO₄³⁻. This peak is particularly pronounced in phosphates with low structural order or heterogeneous cation substitution and often indicates lattice distortion, heavy metal incorporation, or secondary precipitation (Berzina-Cimdina and Borodajenko 2012). A weaker band at 1209 cm⁻¹ likely arises from PO₄³⁻ combination modes or high-frequency shoulders caused by structural distortion, commonly observed in Pb- or Cu-bearing secondary phosphates (Spevak et al. 2013). The band at 672 cm⁻¹, along with several nearby minor features, may correspond to the ν₄ bending mode of PO₄³⁻, although interference from surface-attached contaminants cannot be ruled out. Beyond these peaks, the spectrum lacks the broad features of well-crystallized hydroxyapatite, consistent with a poorly ordered, cation-substituted secondary phosphate and supporting its tentative identification as bone-derived.

Raman spectroscopy provides more detailed information, foremost fully confirming the material composition of each sample (Fig. 3). The agate sample (M627:4) exhibits peaks at 506, 468, 365, 213, and 135 cm⁻¹, characteristic of cryptocrystalline α-quartz with subtle structural heterogeneity (Kingma et al. 1994). In particular, the 468 cm⁻¹ peak corresponds to the symmetric Si–O–Si stretching of α-quartz, while the low-frequency peaks at 213 and 135 cm⁻¹ indicate collective lattice vibrations influenced by microcrystalline size and structural disorder. The presence of the 506 and 365 cm⁻¹ peaks suggests local lattice distortions or minor impurities.

The Raman spectrum of the phosphate-bearing sample exhibits characteristic peaks at 947, 636, 434, and 90 cm⁻¹, consistent with apatite-group minerals (Fig. 3a). The strong band at 947 cm⁻¹ corresponds to the ν₁ symmetric stretching mode of PO₄³⁻, confirms hydroxyapatite and, together with the absence of peaks in the 1000–1100 cm⁻¹ region, indicates that the sample is not a carbonate-substituted apatite. The bands at 636 and 434 cm⁻¹ are attributable to the ν₄ and ν₂ bending modes of the phosphate tetrahedron, respectively, while the low-frequency peak at 90 cm⁻¹ may reflect lattice vibrations within the apatite crystal structure. The slightly lower ν₁ position relative to ideal fluorapatite (~964 cm⁻¹) is consistent with biological or poorly crystalline phosphate phases, such as bone or bone-ash–related material (Penel et al. 1998). The absence of detectable OH-related signals further suggests that the sample has undergone notable weathering.

The tremolite samples (Fig. 3a,b) show a prominent peak at ~680 cm⁻¹, corresponding to symmetric stretching vibrations of bridging oxygens (O_br1) linking Q₂ and Q₃ tetrahedral units (Arslanlar et al. 2011), accompanied by a weaker peak at 534 cm⁻¹ from symmetric bending of O_br1. In the 800–1200 cm⁻¹ region, a doublet at 1035 and 1065 cm⁻¹ reflects distinct Q₂ and Q₃ vibrational modes (Korybska-Sadło et al., 2018), while a broader band at 936 cm⁻¹ is attributed to Si–O non-bridging stretching. Low-frequency peaks below 400 cm⁻¹, including 128, 166, 185, 230, 376, 399, and 423 cm⁻¹, correspond to M–O vibrations and lattice modes, indicative of the long-range order in the silicate framework (Zhang et al., 2017). Additional weak bands out structural side these ranges, such as a minor feature at ~815 cm⁻¹ observed in sample M612:2, may reflect local variations or cationic substitutions within the double‑chain silicate framework, consistent with reported sensitivity of amphibole Raman bands to compositional and crystallographic heterogeneity (Waeselmann et al. 2019).

Sample M1002:1 shows an additional Raman band at ~1202 cm⁻¹ that is absent in other spots (Fig. 3c). As this band is not characteristic of well-crystallized tremolite, it is more reasonably attributed to localized structural disorder or surface-related alteration, possibly involving minor secondary silica-rich phases formed during weathering or burial. Its occurrence in only one sample suggests a sample-specific post-depositional overprint rather than a fundamental mineralogical difference (Sharma et al. 1997). In addition, Raman spectra from the blackened area display broad bands near ~1380 and ~1610 cm⁻¹, corresponding to the D and G bands of disordered carbon. Their broad profiles indicate poorly ordered, weakly graphitized carbonaceous matter, consistent with carbon black infiltration derived from low-temperature combustion or organic matter alteration.

In another jade artifact M624:1, Raman spectra collected from the white, opaque area show bands at ~1089 and ~677 cm⁻¹ (Fig. 3d). The 1089 cm⁻¹ band is consistent with carbonate minerals, most likely calcium carbonate, while the 677 cm⁻¹ peak likely originates from the tremolite matrix itself (Fig. 3d). Their confinement to the opaque zones indicates a secondary deposition rather than a primary component of the jade. In archaeological contexts, such white alteration is commonly linked to burial processes, during which carbonate-bearing groundwater infiltrates microcracks and altered domains in nephrite, followed by CaCO₃ precipitation (Wang et al., 2018). This secondary infilling enhances light scattering, producing the characteristic white, opaque appearance and reflecting post-depositional chemical alteration of the jade (Edwards et al. 2005).

4.2 Elemental characteristic/ Secondary alteration

Based on the integrated use of different XRF techniques for analysis, we can extend our judgment on the material and deepen our understanding of the areas of interest. The most intuitive finding is that XRF mapping can visually display variations in the concentration of specific elements across different regions through color gradients.

Interestingly, the elemental distribution on the jade pendant does not correspond to the blackened areas but instead coincides with the brown surface accretions (Fig. 4), which are locally superimposed on the black zones. In these areas, the major tremolite components (Ca, Si, and Mg) are relatively depleted, whereas Al, K, Fe, and Mn are enriched, a pattern characteristic of exogenous, soil-derived sediment adhering to the surface. By contrast, the blackened areas themselves show little elemental deviation from the white jade matrix, apart from a slight Fe enrichment, indicating that the black coloration is dominated by organic or carbonaceous material, while the brown accretions are primarily inorganic in origin (Fig. 4).

The jade sword ornament has an uneven, strongly curved surface; therefore, μ-XRF mapping was carried out on a relatively flat area (Fig. 5). Within this zone, numerous white opaque patches are visible, together with localized brown surface accretions. Elemental mapping shows that the white areas are compositionally consistent with the tremolite matrix, particularly in Ca, Si, and Mg, indicating that the whitening reflects intrinsic surface alteration of the jade, such as microstructural degradation or physical weathering, rather than the introduction of exogenous materials (Fig. 5).

In contrast, the brown accretions display relative enrichment in Al, K, Fe, and Mn, while the major jade-forming elements remain broadly comparable to the host. This elemental pattern closely parallels that observed in sample M1002:1 and is characteristic of soil-derived or iron–manganese-rich deposits adhering to the surface during burial. The clear chemical contrast between the white opaque zones and the brown accretions points to different formation mechanisms, with the former representing endogenous jade alteration and the latter recording exogenous sedimentary overprinting.

Furthermore, the jade dragon provides a useful comparison (Fig. 6). As in the jade sword ornament, the major elements of tremolite jade (Si, Mg, and Ca) are largely homogeneous, indicating limited disturbance of the primary mineral matrix, although Ca is locally enriched in altered areas. Aluminum is concentrated in carved recesses and relief, likely due to the accumulation of fine sediments, with K showing a similar distribution and thus a shared depositional origin. A conspicuously whitened area in the lower right displays enrichment in Fe and Mn. While brown “sugar-coloured” zones are commonly linked to iron enrichment, their elemental contrast with the fresh jade is weaker than that observed between the white altered zones and the unaltered body. As in the jade sword ornament, the white areas also show slightly elevated Ca, suggesting that whitening reflects intrinsic alteration processes, such as Ca redistribution or secondary precipitation, rather than simple iron staining or surface contamination. In addition, the jade dragon shows faint, localized darkening along its edges, forming continuous but spatially limited zones, likely produced by processes similar to those responsible for the darkened areas observed on the jade pendant (Fig. 7).

While elemental mapping offers a qualitative overview of surface heterogeneity, benchtop XRF provides more robust quantitative constraints. The bulk tremolite compositions of the analyzed samples are broadly comparable (Table 2), particularly in MgO, SiO₂, and CaO, which closely match the expected proportions of ideal tremolite and indicate limited alteration of the primary jade matrix. Notably, the brown surface material identified on M1002:1 also occurs on M563:1 and is characterized by strongly elevated Al₂O₃ contents (Fig. 8), reaching approximately five to ten times the average values measured elsewhere, together with Fe₂O₃ levels several times higher than those of the jade matrix. Correspondingly, the major jade-forming components are systematically depleted in these brown areas. Within individual samples, zones displaying more yellowish, greenish, or darker coloration likewise show a concurrent increase in MnO and Fe₂O₃ content. In samples with relatively subtle colour variation, such as M136:18 and M564:1, Fe₂O₃ concentrations can still be more than 20% higher in these darker-toned areas.

Comparable patterns have been reported in previous archaeometric studies of buried jade artifacts, where exogenous elements are introduced through post-depositional processes (Chen et al. 2024). In burial environments, soil organic matter can form stable complexes with multivalent metal ions, enhancing the mobilization of Fe and Al oxides and hydroxides far more effectively than that of silicate minerals, while simultaneously inhibiting their reprecipitation as crystalline phases. The coupled enrichment of Al and K therefore strongly suggests migration and physical adherence of clay minerals, such as illite or kaolinite, from the surrounding sediments. In contrast, the enrichment of Fe and Mn is best explained by the precipitation of iron–manganese oxide or hydroxide colloids from groundwater, which preferentially accumulate on exposed or micro-porous jade surfaces (Zhang et al. 2019).

The bone-derived sample M1045:3 is further characterized by its distinctive elemental composition. Portable XRF analysis reveals exceptionally high concentrations of P, Ca, and S, together with elevated Sr among the trace elements (Table S1), a geochemical pattern that is consistent with apatite-rich bone material. Notably, however, the sample also contains anomalously high levels of metallic elements, particularly Pb, Cu, and As, with Pb exceeding 10% and locally approaching 20%. Such concentrations are far beyond those expected for unaltered bone and cannot be attributed to biogenic uptake alone. When considered alongside the surface coloration and burial context, these enrichments are best explained by post-depositional contamination related to co-buried bronze artifacts. During burial, corrosion products and metal-rich solutions released from bronzes likely migrated into the porous bone matrix, leading to secondary metal incorporation (Pike and Richards 2002). Such processes are well documented in Chinese archaeological contexts, where tremolite jades frequently exhibit greenish discoloration induced by copper corrosion. Representative examples include jade ge from Jiwanggu (M1:54), Yishui; a jade bird from Yinxu, Anyang (M89:16); a jade ge from the Fu Hao Tomb (M5:977); and a jade ge from Sanxingdui (no. 000127), all exhibiting copper-related secondary greening (Wang 2020).

For the slightly yellow agate sample M627:4, the pXRF results indicate a composition overwhelmingly dominated by SiO₂. Minor amounts of Al₂O₃ and MgO fall within the compositional range commonly reported for natural agate and are best interpreted as inherent features of the raw material, reflecting limited trace substitution or the presence of micro-inclusions. The detection of Fe and Mn is likewise consistent with their role as primary chromophoric elements incorporated during agate formation and accounts for the observed yellow coloration. All other elements occur only at trace levels, providing no evidence for significant post-depositional contamination or secondary chemical overprinting.

4.3 Provenance inference

Portable XRF analysis provides an informative insight: measurements taken from areas with colour variation within each sample closely follow the same geochemical patterns discussed above, while the compositions of the primary bodies of the artefacts remain remarkably stable (Table S2).

Notably, the jade objects comprising the facial covering set, including the eye mask, mouth cicada, nasal ornaments, and short jade rod, each display strong internal elemental consistency, indicating minimal post-burial chemical disturbance. At the same time, systematic compositional differences are evident among the components, most clearly illustrated by the Ni content versus Fe/Mn ratio (Fig. 9). In this comparison, the two eye-mask samples are clearly distinguished from the other elements of the set, exhibiting very low Ni contents but comparatively high Fe/Mn ratios, whereas the remaining components show the opposite tendency. Even where visible colour contrasts exist, such as between M136:14 and the markedly whiter M136:15, their elemental characteristics across different zones remain closely comparable. Taken together, these patterns indicate that the facial covering set was not produced from a single block of raw material, but rather assembled from multiple sources or from different portions of larger jade pieces.

This pattern points to a pragmatic approach to raw-material use, in which smaller fragments or offcuts from the working of larger blocks were incorporated to assemble a complete set, rather than selecting and carving all elements from a single, homogeneous jade mass. This observation is archaeologically significant, as it implies that such funerary jade assemblages, which lack practical function and were prepared specifically for burial, did not necessarily constitute an integrated production set but could represent a composite arrangement brought together for mortuary purposes.

In addition, the white jade samples M1002:1 and M136:12/16/18 plot in close proximity to each other, with M664:3 and M624:1 showing slightly looser but still comparable affinities, indicating broadly similar trace-element characteristics. In contrast, the green jade M563:1 is clearly separated from M612:2, despite their similar macroscopic appearance. This distinction is primarily driven by Ni content: M563:1 contains markedly low Ni, at approximately 20 ppm, whereas M612:2 falls within a compositional group defined by Fe/Mn ratios exceeding 8 and Ni concentrations above 40 ppm, a pattern shared with M564:1. A further comparison shows that M736:1 plots somewhat close to the white jade M136:14/15, yet discernible differences in Ni content remain, suggesting only partial overlap in raw-material signatures rather than true compositional equivalence.

As previously noted in other studies, visually similar jades may nonetheless exhibit distinct trace-element signatures (Chen et al. 2022). When considered together with the strong internal compositional stability observed within individual artefacts, the Ni–Fe/Mn relationships reinforce the interpretation that the Houtun jades derive from multiple raw-material sources rather than a single, uniform provenance.

The submicroscopic structural characteristics observed under multispectral imaging provide independent support for a multi-source interpretation of the tremolite jades (Chen et al. 2018, 2023). Among the well-preserved white nephrite samples, the two eye covers M136:14 and M136:15 exhibit closely comparable microstructural patterns (Fig. 10). In both cases, high-brightness features are distributed in a fine, punctate manner, with pronounced contrast differences, forming a densely dispersed texture reminiscent of a millet- or porridge-like fabric at a very small spatial scale. By contrast, M136:18 presents a markedly different internal appearance. Its overall texture appears more homogeneous and optically “clean”, lacking the dense, high-frequency bright points seen in M136:14 and M136:15. Only in areas close to the margins are small white spots observable, whereas the interior is dominated by larger, weakly contrasted circular patches. Although the marginal white spots show some resemblance to the punctate features of M136:14 and M136:15, the overall discrepancy in microstructural organization indicates that these samples were not fashioned from the same block of tremolite.

Moreover, the internal pattern of M136:18 suggests a deliberate selection strategy during manufacture. The artisan appears to have preferentially extracted higher-quality material from the raw jade, carefully excluding zones of inferior texture. As a result, areas of relatively lower material quality are confined to the edges, while the interior preserves a more uniform and refined structure. This selective use of raw material not only reinforces the inference of multiple sources but also highlights a nuanced awareness of material heterogeneity in the working of tremolite jades.

Turning to the remaining samples, the degree of microstructural variability becomes even more pronounced (Fig. 11). The jade dragon M564:1 does not exhibit a punctate fabric but instead shows low-frequency, weakly contrasted block-like aggregates, indicating a textural organization distinct from finely dispersed tremolite. M664:3 similarly departs from a granular pattern, with high-brightness features forming elongated, interconnected bands that are locally obscured by the matrix, suggesting a different internal fabric and formation history. M1002:1, although also a white tremolite jade, contrasts with both the jade face covering assemblage and the other samples, as its microstructure is dominated by relatively continuous linear high-brightness features with minimal interstitial heterogeneity. Taken together, this progression from blocky to banded and linear microstructural expressions cannot be accommodated within a single, homogeneous source framework. Instead, these differences point to the use of tremolite jades derived from geologically distinct sources or from texturally differentiated portions of source bodies, underscoring the heterogeneity of jade procurement practices in the Beijing region during the Warring States-Han period.

4.4 Cultural significance

This fragmented bone object from the Houtun site poses a clear diagnostic challenge. Its annular form with a break could reflect either a damaged ring or a jue originally made with an intentional opening. Although a definitive typological attribution is not currently possible, the antiquity of the fracture is evident. Rather than forcing a categorical identification, this uncertainty is analytically meaningful, underscoring the broader difficulty of distinguishing rings from jue in fragmentary assemblages, particularly when diagnostic traces around the opening are poorly preserved or absent.

Such caution is warranted given the object’s potential significance. Should future, more detailed analyses confirm it as a jue, it would represent an important addition to the limited corpus of securely identified bone jue. Even at present, this sample invites reassessment of the long and discontinuous trajectory of bone jue within the archaeological record.

The earliest bone jue in China date to the Early Neolithic. Two serrated bone ornaments from House F2 at the Xiaojingshan site in Zhangqiu, Shandong, originally reported as rings, are better interpreted as jue based on their openings and are dated to ca. 8000–7000 BP (Jinan Municipal Bureau of Culture and Zhangqiu Museum 2014). In the lower Yangtze region, bone jue appear slightly later, with single examples from the Hemudu site, three burials at the Majiabang site, and a burial at Xuecheng in Gaochun, Jiangsu, indicating continuity from the early Hemudu culture to the early–middle Songze period (ZPICRA 2003, ZPICRA and Jiaxing Museum 2019, Zhou et al. 2000). By around 6000 BP, bone jue are also attested at the Qianjiaping site in the middle Yangtze region (Fig. 12), suggesting their spread across major cultural zones by the Middle Neolithic (Yin 2020).

In contrast, the northeastern region yields no confirmed bone jue until much later. A single example from the Ping’anbao site in Zhangwu, Liaoning, dated to the Bronze Age, currently represents the earliest known case from this area (Zhu et al. 1992). From the Spring and Autumn to Warring States periods, however, bone jue appear more frequently in frontier regions, particularly the northern steppe and the southwest. In the north, examples are documented from the Maoqinggou cemetery in the Daihai region of Inner Mongolia, where a bone jue was placed on the left chest of the deceased in M81 (Li 2019), as well as from the Maojiaping site in Gangu, Gansu, where two morphologically similar samples were recovered (Liang et al. 2022). Additional finds include a pair of finely decorated double-hook, coiled hui-motif bone jue from M30 at the Wayaopo cemetery in Xixian, Shanxi (Wang et al. 2017), and a single example from M49 at the Shangguo cemetery in Wenxi (Zhou 2022).

In the southwest, bone jue are especially numerous during the Warring States and Han periods. At the Kele cemetery in Hezhang, Guizhou, 25 bone jue were recovered, including seven symmetrically placed beside the head of a child in M373 (Fig. 12); other burials, mostly male, contained between one and six examples, reflecting variability in wearing practices (Wu et al. 2015). Comparable finds include a bone jue worn on the right ear of a female individual at the Yinzitan cemetery in Weining, Guizhou (Li 2006), a single example from M575 at the Laolongtou cemetery in the Yanyuan Basin (Ye 2022), and four bone jue recovered from a megalithic tomb at the Bahebaozi cemetery in Xichang (Sichuan 1976).

Rather than representing an exceptional case, the object is best understood as a modest addition to the limited and uneven record of bone annular ornaments, offering incremental evidence for variability in production and use across different regions and periods, particularly in areas where multiple cultural traditions intersect, without exceeding what the material evidence can securely support.

This context of material variability and ritual practice extends meaningfully to the wider phenomenon of intentional breakage, which is archaeologically attested not only in bone but also, and more prominently, in jade. The ritual significance of jade fragmentation, or the burial of intentionally broken jade, has long attracted scholarly attention. Its deep roots are explicitly documented as early as the oracle-bone inscriptions, which record a wide range of sacrificial practices, including liao yu (burning jade), chen yu (submerging jade), mai yu (burying jade), hui yu (destroying jade), zou yu (offering jade), cheng yu, and zun yu. These divinations indicate that different forms of jade sacrifice were directed toward distinct ritual recipients: liao yu was performed for former lords and kings as well as for mountains and rivers; chen yu was directed predominantly toward rivers; mai yu toward mountains; and hui yu specifically toward ancestral rulers (Wang 2020). Such practices continued into the Spring and Autumn and Warring States periods, as illustrated by the Zuo zhuan account (Duke Zhao, eighth year), which records that Yuan Ke “slaughtered horses and destroyed jade for burial” following the fall of the state of Chen, underscoring the continued funerary relevance of jade destruction (Hou 2024).

During the Warring States and Han periods, a time when the system of jade burial reached a high point amid profound social and ritual transformation, jade objects in elite tombs, such as those at the Dabaotai Western Han mausoleum in Beijing, are often well preserved and complete (Dabaotai Han Tomb Excavation Team and IACASS 1989). By contrast, the frequent occurrence of fragmented jade in small- and medium-sized burials has received far less systematic attention and may reflect differences in how funerary norms were practiced across social strata. At the Houtun cemetery, jade fragments recovered from Warring States and Western Han tombs exhibit fracture surfaces characteristic of sudden impact rather than long-term use or post-depositional stress, most clearly exemplified by the broken jade bi M563:1. The fact that these fragments cannot be reassembled into complete objects further suggests deliberate breakage prior to interment. Such acts may be interpreted as part of a ritualized process of offering or apotropaic practice, intended to effect the symbolic transformation of jade from the realm of the living to that of the dead (Chapman 2000). In this sense, the Houtun material provides a micro-scale perspective on the role of jade in mortuary ritual in the Beijing region and highlights the dynamic interplay between central ritual prescriptions and local funerary practices during the Warring States and early Han periods.

Therefore, the intertwined phenomena of fragmentation, compositeness, and secondary alteration observed in the Houtun jades collectively form a coherent set of material evidence. They reflect not only pragmatic resource management strategies but also the development of locally adapted mortuary behaviors within Beijing’s frontier communities, even as they engaged with Central Plains ritual norms. By situating the material trajectories revealed through scientific analysis within this broader cultural and ritual context, this study offers a nuanced, micro-scale case for understanding the complex interplay between material culture and social practice in borderland regions during the formative period of early imperial China.

5 Conclusions

This study presents a comprehensive scientific assessment of jade artefacts from the Houtun cemetery dating from the Warring States to the Han dynasty, aiming to clarify material composition, surface alteration, provenance variability, and associated mortuary practices in the Beijing region. Through the combined application of Raman spectroscopy, FTIR, X-ray fluorescence analyses, and multispectral imaging, the research demonstrates the value of integrating complementary non-destructive techniques to resolve complex archaeological questions that cannot be addressed through typology alone.

The results confirm tremolite nephrite as the dominant raw material, while also identifying agate and a bone-derived object that has undergone substantial post-depositional chemical modification. Elemental and spectroscopic data allow a clear distinction between intrinsic alteration of jade, such as whitening related to microstructural degradation and secondary mineral precipitation, and extrinsic surface accretions dominated by iron-, manganese-, and aluminium-rich components derived from the burial environment. In the case of the bone artefact, anomalously high concentrations of metallic elements are best explained by contamination from nearby bronze corrosion products, underscoring the need for caution when interpreting chemically altered materials in mixed burial contexts.

More importantly, trace-element characteristics and submicroscopic structural features reveal that the analyzed nephrite jades were not derived from a single geological source. Distinct compositional groupings, particularly evident in the relationship between nickel and iron–manganese ratios, demonstrate the use of multiple raw-material sources or texturally differentiated portions of larger jade bodies. This pattern is especially clear within the Eastern Han jade facial covering set, whose individual components show strong internal consistency but clear inter-object divergence, indicating a pragmatic mode of assembly rather than production from a single homogeneous block.

From a cultural perspective, the frequent occurrence of deliberately fragmented jade objects in Warring States and Western Han burials suggests that intentional breakage formed part of funerary practice, particularly in non-elite contexts. Together, these findings indicate that jade use in the Beijing frontier was characterized by flexible material selection, variable technological choices, and locally specific ritual behaviour. This study therefore contributes a materially grounded perspective on how frontier communities actively engaged with, adapted, and reinterpreted broader jade traditions during the Warring States–Han transition.

References

[1] Arslanlar Y. T., J. Garcia-Guinea, R. Kibar, A. Çetin, M. Ayvacıklı , N. Can. . Luminescence Behavior and Raman Characterization of Jade from Turkey. Applied Radiation & Isotopes, 2011, 69(9): 1299–1306

[2] Barnes G. L. . Understanding Chinese Jade in a World Context. Journal of British Academy, 2018, 6: 1–63

[3] Beijing Institute of Archaeology北京市文物研究所. 1995. The Western Zhou Yan State Cemetery at Liulihe (1973–1977) 琉璃河西周燕国墓地(1973-1977). Beijing: Cultural Relics Press (in Chinese).

[4] Beijing Institute of Archaeology北京市文物研究所. 2007. The Jundushan Cemetery: Yuhuangmiao军都山墓地:玉皇庙. Beijing: Cultural Relics Press (in Chinese).

[5] Beijing Institute of Archaeology北京市文物研究所. 2024. The Houtun Cemetery (Warring States Volume)后屯墓地(战国卷). Beijing: Science Press (in Chinese).

[6] Berzina-Cimdina, L., and N. Borodajenko. 2012. “Research of Calcium Phosphates Using Fourier Transform Infrared Spectroscopy.” InTech. doi:10.5772/36942.

[7] Chapman, J. 2000. Fragmentation in Archaeology: People, Places and Broken Objects in the Prehistory of South Eastern Europe. London: Routledge.

[8] Chen D., M. Pan, W. Huang, W. Luo , C. Wang. . The Provenance of Nephrite in China Based on Multi-spectral Imaging Technology and Gray-Level Co-occurrence Matrix. Analytical Methods, 2018, 10: 4053–4062

[9] Chen D., J. Tang, M. Yu, Y. Yang , C. Wang. . Identification of a Source of Nephrite from Late Shang Yinxu by Multispectral Imaging. Heritage Science, 2023, 11: 4

[10] Chen D., Y. Yang, B. Qiao, J. Li , W. Luo. . Integrated Interpretation of pXRF Data on Ancient Nephrite Artifacts Excavated from Tomb No.1 in Yuehe Town, Henan Province, China. Heritage Science, 2022, 1: 69

[11] Chen D., B. Qiao , W. Luo. . Concentration-number (C-N) fractal models reveal the distribution pattern of the elements in ancient nephrite measured by portable X-ray fluorescence: Based on nephrite objects excavated from different sites in Nanyang, Henan Province. Archaeometry, 2024, 66(4): 931–948

[12] Chen D., J. Yang, H. Y. Han, J. H. Zhang, C. Li , W. G. Luo. . From Mines to Tombs: Decoding the Journey of Turquoise Artifacts at the Xingong Site (1500-1300 BC), Beijing. Archaeological and Anthropological Sciences, 2025, 17: 102

[13] Dabaotai Han Tomb Excavation Team, and Institute of Archaeology大葆台汉墓发掘组, Chinese Academy of Social Sciences (IACASS)中国社会科学院考古研究所. 1989. The Dabaotai Han Tombs in Beijing北京大葆台汉墓. Beijing: Cultural Relics Press (in Chinese).

[14] Edwards H. G. M., S. E. J. Villar, J. Jehlicka , T. Munshi. . FT-Raman Spectroscopic Study of Calcium-Rich and Magnesium-Rich Carbonate Minerals. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2005, 61(10): 2273–2280

[15] Flörke W. , O. Graetsch , H. Martin , B. Röller , K. Wirth. . Nomenclature of Micro- and Non-crystalline Silica Minerals, Based on Structure and Microstructure. Neues Jahrbuch für Mineralogie, 1991, 163(1): 19–42

[16] Gu, F.古方, ed. 2005. The Complete Collection of Jades Unearthed in China: Beijing, Tianjin, Hebei Volume中国出土玉器全集·北京、天津、河北卷. Beijing: Science Press (in Chinese).

[17] Hou, Y.侯雅兰. 2024. “A Study on the Burial System of the Beibaie Cemetery北白鹅墓地丧葬制度研究.” Master’s thesis, Henan University (in Chinese).

[18] The First Excavation of the Xiaojingshan Site in Zhangqiu, Shandong山东章丘小荆山遗址第一次发掘. Eastern Archaeology东方考古, 2014, (1): 405–449

[19] Kingma K.J. , R. J. Hemley. . Raman Spectroscopic Study of Microcrystalline Silica. American Mineralogist, 1994, 79(3-4): 269–273

[20] Korybska-Sadło Gil , I. Gunia , G. Horszowski , P. Sitarz. . Raman and FTIR Spectra of Nephrites from the Złoty Stok and Jordanów Śląski (the Sudetes and Fore-Sudetic Block, SW Poland). Journal of Molecular Structure, 2018, 1166: 40–47

[21] Li, F.李飞. 2006. “An Analysis of the Yinzitan Cemetery in Weining, Guizhou贵州威宁银子坛墓地分析.” Master’s thesis, Sichuan University (in Chinese).

[22] Li, Y.李永珍. 2019. “A Study on the Personal Adornments Unearthed from Female Burials of the Spring-Autumn and Warring States Periods in the Daihai Region, Inner Mongolia内蒙古岱海地区春秋战国时期女性墓葬出土装饰品研究.” Master’s thesis, Inner Mongolia Normal University (in Chinese).

[23] Preliminary Report on the 2012–2014 Excavation of the Gouxi Cemetery at the Maojiaping Site, Gangu, Gansu甘肃甘谷毛家坪遗址沟西墓地2012~2014年发掘简报. Archaeology and Cultural Relics考古与文物, 2022, (03): 27–46

[24] et al. New Archaeological Discoveries at the Warring States to Western Han Cemetery in Houtun Village, Lucheng Town, Tongzhou District, in 2019 2019年通州区潞城镇后屯村战国至西汉墓地考古新发现. Beijing Cultural Relics and Museums Review北京文博文丛, 2020, (03): 71–84

[25] Penel G., G. Leroy, C. Rey , E. Bres. . MicroRaman Spectral Study of the PO₄ and CO₃ Vibrational Modes in Synthetic and Biological Apatites. Calcified Tissue International, 1998, 63(6): 475–481

[26] Pike A.W.G. , M. P. Richards. . Diagenetic Arsenic Uptake in Archaeological Bone: Can We Really Identify Copper Smelters?. Journal of Archaeological Science, 2002, 29(6): 607–611

[27] Q i , X .齐心 . A Preliminary Exploration of Jade Culture in Pre-Qin Beijing北京先秦玉器文化初探. Beijing Cultural Relics and Archaeology北京文物与考古, 2002, (1): 92–98

[28] Ren J., G. Shi, J. Zhang,. . et al. Infrared Spectra of Grayish Green Nephrite and Gray Nephrite: Characteristics and Significance. Spectroscopy and Spectral Analysis, 2019, 39(3): 772–777

[29] Sharma S. K., S. Misra , P. G. Lucey. . Raman Spectroscopy of Silicate Glasses. Journal of Non-Crystalline Solids, 1997, 223: 109–120

[30] Preliminary Report on the Excavation of the Dolmen at Bahebaozi, Xichang西昌坝河堡子大石墓发掘简报. Archaeology考古, 1976, (5): 357–330

[31] Spevak L., C. R. Flach, T. Hunter, R. Mendelsohn , A. Boskey. . Fourier Transform Infrared Spectroscopic Imaging Parameters Describing Acid Phosphate Substitution in Biologic Hydroxyapatite. Calcified Tissue International, 2013, 92(5): 418–428

[32] Tan T., L. L. Ng, L. C. Lim,. . et al. Studies on Nephrite and Jadeite Jades by Fourier Transform Infrared (FTIR) and Raman Spectroscopic Techniques. Cosmos, 2013, 47–56

[33] Waeselmann N., J. Schlüter, T. Malcherek, G. D. Ventura , B. Mihailova. . Nondestructive Determination of the Amphibole Crystal-chemical Formulae by Raman Spectroscopy: One Step Closer. Journal of Raman Spectroscopy, 2019, 51(9): 1530–1548

[34] Wang, R.王荣. 2020. Research on the Scientific Archaeology and Conservation of Early Chinese Jade. Shanghai: Fudan University Press (in Chinese).

[35] A Study on the Natural Whitening Phenomenon of Jades of Tremolite and Serpentine from Yinxu: with a Re-examination of the 'Calcification' Phenomenon殷墟透闪石和蛇纹石玉器自然白化现象研究——兼谈重识“钙化”现象. Cultural Relics in Southern China南方文物, 2018, (03): 79–87

[36] Wang, X.王鑫, and L. Cheng程利. 2002. “The Laoshan Han tomb in Shijingshan District石景山区老山汉墓.” In Chinese Archaeology Yearbook 2001中国考古学年鉴2001. Beijing: Cultural Relics Press (in Chinese).

[37] Two Tombs of the Spring-Autumn Period at the Wayaopo Cemetery in Xi County, Shanxi山西隰县瓦窑坡墓地的两座春秋时期墓葬. Archaeology考古, 2017, (05): 2–53

[38] Excavation of Two Han Dynasty Tombs at the Kele Cemetery in Hezhang County, Guizhou贵州赫章县可乐墓地两座汉代墓葬的发掘. Archaeology考古, 2015, (02): 19–31

[39] Yang J., W. G. Luo, H. Y. Han , D. Chen. . On-site Analysis of the Turquoise Artifacts Excavated from the Xingong Site in Beijing. Journal of Archaeological Science: Reports, 2025, 61: 104981

[40] Ye, X. 叶小青. 2022. “A Study on the Bronze Knives Unearthed from the Yanyuan Basin盐源盆地出土青铜刀研究.” Master’s thesis, Sichuan University (in Chinese).

[41] Preliminary Report on the Archaeological Excavation of the Qianjiaping Neolithic Site in Guiyang, Hunan湖南桂阳千家坪新石器时代遗址考古发掘简报. Hunan Archaeology Review湖南考古辑刊, 2020, (00): 324–336

[42] Y u , P .于平 . Jade Artefacts from the Beijing Region北京地区的玉器. Beijing Cultural Relics and Museums北京文博, 2002, 30(4): 42–51

[43] Discovery of a Shang Dynasty Tomb in Pinggu County, Beijing北京市平谷县发现商代墓葬. Cultural Relics文物, 1977, (11): 96–99

[44] Analysis of the Ethnic Affiliation and Related Issues of the Baifu Western Zhou Tomb in Changping, Beijing北京昌平白浮西周墓族属及相关问题辨析. Research on China’s Frontier Archaeology边疆考古研究, 2017, (2): 177–190

[45] Analysis of the Material and Weathering Process of Jades from Moyi Mountain in Zengcheng, Guangzhou广州增城墨依山玉器材质和受沁过程分析. Sciences of Conservation and Archaeology文物保护与考古科学, 2019, 31(03): 68–76

[46] Zhang Y., T. Hui , X. Feng. . Characterization of Mg and Fe Contents in Nephrite Using Raman Spectroscopy. Gems and Gemology, 2017, 53(2): 204–212

[47] Zhejiang Provincial Institute of Cultural Relics and Archaeology浙江省文物考古研究所 (ZPICRA). 2003. Hemudu河姆渡. Beijing: Cultural Relics Press (in Chinese).

[48] Zhejiang Provincial Institute of Cultural Relics and Archaeology浙江省文物考古研究所 (ZPICRA), and Jiaxing Museum嘉兴博物馆. 2019. Majiabang马家浜. Beijing: Cultural Relics Press (in Chinese).

[49] Determination of the Manufacturing Techniques of Jades Unearthed from the Dabaotai Western Han Tomb大葆台西汉墓出土玉器加工工艺判断. Collected Papers of the Capital Museum首都博物馆论丛, 2019, 420–427

[50] Zhou, J.周佳雯. 2022. “Periodization and Ranking Analysis of the Elite Tombs of the Early Jin State during the Zhou Dynasty周代前期晋国贵族墓葬分期与等级探析.” Master’s thesis, Zhengzhou University (in Chinese).

[51] Preliminary Report on the Excavation of the Xuecheng Neolithic Site in Gaochun County, Jiangsu江苏高淳县薛城新石器时代遗址发掘简报. Archaeology考古, 2000, (05): 97–101

[52] Re-exploration of Issues Related to the Han Dynasty Yan State, Guangyang Kingdom, and the Dabaotai Han Tombs关于汉代燕国、广阳国及大葆台汉墓相关问题的再探讨. Qin-Han Studies秦汉研究, 2008, 279–285

[53] The Ping’anbao site in Zhangwu, Liaoning辽宁彰武平安堡遗址. Acta Archaeologica Sinica考古学报, 1992, (04): 529–534

Rights & permissions

The Author(s) 2026. This article is available under open access at journal.hep.com.cn.