Lithosphere Modification Beneath the North China Craton: Geochemical Constraints of Water Contents from the Damaping Peridotite Xenoliths

Abstract

The water contents and geochemical evidence of nominally anhydrous minerals in peridotite xenoliths provide critical insights into lithospheric mantle features, offering a deep understanding of cratonic destruction and mantle evolution processes. Damaping, located in the central part of the intra-North China Craton, hosts abundant mantle peridotite xenoliths’ samples, providing new constraints on lithospheric mantle evolution. In this study, spinel lherzolite samples from Damaping Cenozoic basalts were analyzed for major and trace elements, water content, and oxygen isotope to investigate the factors controlling mantle water distribution and lithospheric mantle modification. The olivines of Damaping spinel lherzolite have a range of Mg# values from 89.73 to 91.01, indicating moderately refractory mantle characteristics. Clinopyroxenes display an LREE-depleted pattern, suggesting a consistency with 1–6% of batch partial melting and 1–5% fractional partial melting. The high (La/Yb)_N (0.20–0.73) and low Ti/Eu (3546.98–5919.48) ratios of Damaping clinopyroxenes reveal that the lithosphere mantle beneath the Damaping has undergone silicate metasomatism. The water contents of Damaping clinopyroxenes and orthopyroxenes range from 13.39 to 19.46 ppm and 4.60 to 7.82 ppm, respectively. The water contents of the olivines are below the detection limit (<2 ppm). The whole-rock water contents can be estimated based on the mineral modes and partition coefficients, with values ranging from 3.21 to 5.44 ppm. Partial melting indicators (Mg# in Ol and Yb_n in Cpx) correlate with the water content in clinopyroxenes and orthopyroxenes but show no correlation with the redox state (Fe³⁺/∑Fe ratios in spinel) or metasomatism ((La/Yb)_N in clinopyroxene). These results suggest that the degree of partial melting primarily controls the heterogeneous water distribution in Damaping spinel lherzolite, rather than the redox state or metasomatism. The δ¹⁸O values of clinopyroxenes from Damaping spinel lherzolites (5.27–5.59‰) fall within the range of mid-ocean ridge basalts (MORB), indicating a mantle source characterized by MORB-like isotopic signatures. The low whole-rock water contents are attributed to lithospheric reheating resulting from asthenospheric upwelling during the Late Mesozoic–Early Cenozoic. Therefore, the lithosphere is predominantly composed of ancient Proterozoic residues, with localized contributions of younger asthenospheric material near deep faults.

1. Introduction

The North China Craton (NCC), bounded by the Central Asian orogenic belt to the north and the Qinling-Dabie-Sulu orogenic belt to the south, is a key region for studying the evolution of the lithosphere. The NCC formed through the assembly of the Eastern and Western Blocks at ~1.85 Ga, resulting in the establishment of the Trans-North China orogen [1,2]. Cenozoic alkaline basalts in the NCC, which contain mantle xenoliths, provide direct insights into the composition, evolution, and fluid activity of the lithospheric mantle [3]. Within the NCC, regions such as Nushan, Hebi, Shanwang, and Hannuoba have exposed abundant peridotite xenoliths (Figure 1a). The constituent minerals and chemical composition of peridotite xenoliths are recognized as high-quality samples for revealing the properties of the lithospheric mantle [4]. Significant spatial heterogeneities exist in mantle properties across the NCC, characterized by fertile mantle beneath Shanwang (Mg#-Ol < 90) and refractory mantle beneath Hebi and Mengyin (Mg#-Ol > 92) [5].

Damaping, located on the northern margin of the NCC, is of particular significance. Re-Os isotopic studies revealed that the Damaping lithospheric mantle was replaced at ~1.9 Ga, with the formation of fertile mantle occurring during the Phanerozoic [9]. Damaping basalts (10–22 Ma) contain diverse xenoliths from the lower crust and upper mantle, including spinel lherzolites formed as partial melting residues at ~1.9 Ga [10,11]. These xenoliths record complex metasomatic processes, including interactions with silicate melts/fluids [12,13], and exhibit low water contents attributed to lithospheric reheating during the Late Mesozoic–Early Cenozoic [14]. These findings establish Damaping as a key region for studying mantle modification and the spatial heterogeneity of the NCC lithosphere.

The main minerals of peridotite xenoliths (olivine, spinel, pyroxenes) are nominally anhydrous minerals (NAMs), which contain no water in their chemical formula, but hydrogen can be incorporated as a lattice defect in the crystal structure [15]. Even trace amounts of water can significantly influence the physical and chemical properties of minerals and rocks, as well as dynamic processes and their mantle domains [16,17,18]. The water content in nominally anhydrous minerals from peridotite mantle xenoliths from Hannuoba ranges from 50 to 155 ppm (H₂O) for clinopyroxene and from 20 to 55 ppm (H₂O) for orthopyroxene in these peridotites [19]. These findings indicate that the distribution of water in the lower crust and upper mantle beneath the NCC is laterally and vertically heterogeneous. Hannuoba peridotites experienced partial melting during the Paleoproterozoic (~1.9 Ga) [10], an age that closely corresponds to the collision of the Eastern and Western Blocks. Additionally, four distinctive REE patterns are observed in the whole-rock compositions. To date, H₂O studies in Damaping have primarily focused on quantifying water contents, and there has been limited in-depth research on the factors controlling water content variations [20,21,22].

This study aimed to address this gap by conducting a detailed and integrated analysis of the major and trace elements, water content, and oxygen isotope of the Damaping lherzolite from the NCC. The primary objective of this analysis was to further elucidate the mantle properties and evolution of the Damaping area through the integration of findings from previous studies (1) to explore the distribution of the water content of the peridotites in the Damaping area and identify the factors influencing these variations and (2) to trace the sources and characteristics of materials that modified the lithospheric mantle beneath the NCC.

2. Geological Setting

The NCC, one of the world’s oldest cratons, contains Archean to Proterozoic crystalline basement rocks [23,24]. It comprises two major tectonic domains: the Eastern and Western Blocks, which were amalgamated during the Paleoproterozoic collisional orogeny (~1.85 Ga) [24,25] (Figure 1a,b). The Eastern Block basement predominantly consists of Paleoarchean to Neoarchean (2.5–3.8 Ga) tonalitic–trondhjemitic–granodioritic (TTG) suites, along with late Archean granitic gneisses and supracrustal sequences, including metavolcanic and metasedimentary rocks [24,26,27]. The Western Block is characterized by high-grade metamorphic assemblages, primarily comprising granulite-facies TTG gneisses and orthopyroxene-bearing charnockites, indicative of deep crustal processes [25,28]. These basement rocks are unconformably overlain by Archean to Paleoproterozoic metasedimentary sequences, which form prominent greenstone belts and banded iron formations [2,24]. Since the Late Mesozoic (~160 Ma), the NCC has experienced significant tectono-thermal reactivation, attributed to the subduction of the surrounding plates, leading to extensive lithospheric thinning and transformation and craton modification [1,29,30]. The NCC has witnessed multiple tectonic episodes including the initial southward subduction of the Paleo-Asian oceanic plate from the Early Paleozoic through Late Permian, followed by northward convergence with the Yangtze Block during the Triassic, and westward subduction of the Pacific Plate commencing in the Early Cretaceous. The stratigraphic hiatus spanning the Silurian to Early Carboniferous along the northern NCC margin suggests significant continental uplift and erosion, likely linked to the southward subduction of the Paleo-Asian oceanic plate.

In the NCC, xenoliths similar to those observed in the Damaping are also documented in several other localities. Prominent examples include the Nushan locality within the Tan-Lu Fault Zone, along with Mingxi and Qilin in the Cathaysia block (Figure 1b), each of which provides unique insights into the cratonic lithospheric architecture. The Damaping Cenozoic basalts (40°58′08″ N, 114°32′58″ E), which were the focus of this study, are located in the northern part of the NCC’s central belt [31,32,33] (Figure 1b). This region occupies a pivotal geotectonic position at the convergence of the Inner Mongolia anteclise and Yanshan subsidence belt, which is adjacent to the major Shangyi-Chicheng deep-seated fault system. The volcanic stratigraphy consists of two interbedded series: tholeiitic and alkaline basalts, with the alkaline varieties predominantly forming the basal units of each eruptive sequence. A striking contrast exists between the two basalt types: alkaline basalts are remarkably enriched in mantle-derived and lower crustal xenoliths, whereas tholeiitic basalts rarely contain such inclusions [7]. The xenolith assemblages include mafic to felsic granulites [7], spinel- and garnet-bearing pyroxenites [7,34,35,36], and a significant proportion of spinel peridotites [7,11,34]. The Damaping basalt comprises intercalated alkali basalt and tholeiite basalt [7,33], with alkali basalt predominantly occurring at the base of most profiles. The Oligocene-Miocene basalts of Damaping (10–24.5 Ma) [37] serve as exceptional carriers of deep-seated xenoliths, providing direct samples from both the lower crust and the upper mantle. The xenolith population is dominated by spinel lherzolites and harzburgites [7,32,33,38,39], with minor occurrences of spinel-garnet lherzolites [7,40]. Complementing this assemblage are minor but petrologically important occurrences of spinel- and garnet- pyroxenites [7,32,34,39], along with mafic to felsic granulite [7,41,42,43,44], which provide crucial insights into crust–mantle interaction processes. This unique combination of features establishes Damaping as a world-class natural laboratory for investigating lithospheric evolution and crust–mantle interaction processes.

3. Materials and Methods

3.1. Sample Description

Three representative rock samples were systematically collected and designated as DMP202, DMP204, and DMP208 for comprehensive petrological analysis. These samples contained abundant peridotite xenoliths, which exhibited a uniform distribution within the host rocks (Figure 2a). Five specific domains were carefully selected for microstructural and geochemical analysis, designated as DMP202-1, DMP204-1, DMP204-2, DMP208-1, and DMP208-2, each representing distinct petrological features within the xenoliths. The host basalts displayed a characteristic gray-black coloration with a dense, massive structure, whereas the peridotite xenoliths exhibited fresh, dark green to yellow-green hues. These xenoliths typically ranged from medium to coarse grain size (0.5–2 mm) and exhibited well-preserved massive structures. The constituent minerals predominantly exhibited a euhedral to subhedral granular texture. The xenoliths ranged in diameter from 4 to 7 mm and were composed of a characteristic mineral assemblage of olivine (Ol), clinopyroxene (Cpx), orthopyroxene (Opx), and spinel (Sp).

Petrographic observations revealed curved mineral boundaries, and the transition zone near the host rock exhibited a fragmentary porphyritic texture. Modal composition analysis identified olivine as the dominant phase, accounting for 47.9 to 61.39% of the xenoliths. Clinopyroxene constituted the second most abundant mineral in the studied sample, ranging from 6.12 to 14.03%. Orthopyroxene content varied between 30.07 and 39.15%, with minor spinel contents ranging from (0.11 to 6.29%) (

Table 1. Rock types and mineral modes from Damaping peridotite xenoliths.

Samples	Rock Type	Mineral Mode (%)
		Cpx	Opx	Ol	Sp
DMP202-1	Spinel lherzolite	6.12	32.38	61.39	0.11
DMP204-1	Spinel lherzolite	10.57	30.07	53.07	6.29
DMP204-2	Spinel lherzolite	6.79	39.15	52	2.06
DMP208-1	Spinel lherzolite	14.03	33.6	47.9	5.47
DMP208-2	Spinel lherzolite	7.17	32.25	54.74	4.84

Mineral modes were estimated by point counting on thin sections. Ol, olivine; Opx, orthopyroxene; Cpx, clinopyroxene; Sp, spinel.

). All samples were classified as spinel lherzolite (Figure 3).

The minerals displayed poikilitic structures. Petrographic observations revealed that spinel was enclosed within orthopyroxene (Figure 2b) and olivine (Figure 2d). Olivine frequently displayed a characteristic 120° triple junction (Figure 2c). CO₂ phase gas–liquid inclusions were preserved within both olivine and pyroxene (Figure 2e). Clinopyroxene exhibited two distinct reaction textures: small fragments’ rims surrounding spinel (Figure 4a) and well-developed reaction coronas along larger clinopyroxene grains (Figure 4b). The distinct reaction rims on clinopyroxene grains provided clear petrographic evidence of metasomatic alteration. However, these reaction features maintained sharp boundaries with the host mineral and showed no detectable influence on the primary clinopyroxene composition. Fragmentation textures were also observed along clinopyroxene margins (Figure 4c). Spinel grains contained well-preserved melt pocket composed of secondary clinopyroxene, spongy-texture rim, and interstitial glass (Figure 4a). Backscattered electron (BSE) imaging revealed significant core-to-rim compositional zoning within spinel melt pockets, particularly in Al, Cr, and O concentrations (Figure 5).

Olivine represented the dominant mineral phase, accounting for more than 40% of the modal composition. Under plane-polarized light, olivine appeared colorless to pale yellow, exhibiting characteristic irregular granular morphology with fractured crystals exhibiting wide margins and distinct surface roughness. Cross-polarized observations revealed second- to third-order interference colors, parallel extinction, and well-developed kink bands, indicative of plastic deformation (Figure 2f). Olivine frequently displayed a characteristic 120° triple junction (Figure 2c).

Orthopyroxene exhibited tabular to short prismatic morphologies. Crystal margins frequently exhibited dissolution features. The mineral exhibited fine, widely spaced cleavage with grain sizes ranging from 0.2 to 3 mm, occasionally reaching >5 mm. In plane-polarized light, it exhibited as pale yellow (Figure 2b), whereas cross-polarized observations revealed first-order gray to yellow interference colors.

Clinopyroxene was mostly irregularly tabular, exhibiting well-developed cleavage systems with prominent fracture networks. The grain size ranged from 0.2 to 2 mm, typically finer than that of associated orthopyroxene. In plane-polarized light, it was generally light green (Figure 2b), while cross-polarized observations revealed high-order interference colors extending beyond second-order blue. Microstructural analysis revealed two genetically distinct clinopyroxene populations in Damaping peridotites—primary magmatic Cpx-1 grains with sharp crystal boundaries (Figure 2b) and metasomatized Cpx-2 grains featuring distinct reaction rims that establish clear phase boundaries with adjacent minerals (Figure 4a,b).

Spinel appeared predominantly brown in plane-polarized light, occurring as irregular grains with rare octahedral forms. The crystals exhibited extensive fracturing, distinct black reaction rims, and positive relief. Under cross-polarized light, spinel exhibited complete extinction (Figure 2d), with characteristic spongy-textured margins (Figure 4a).

3.2. Methods

3.2.1. Microscopic Observation and Spectroscopy

Microscopic observations and laser Raman spectroscopy analyses were performed at the Gemological Experimental Teaching Center, School of Jewelry, China University of Geosciences (Beijing, China). Sample textures were examined using a GI-MP22 binocular microscope (Nanjing Baoguang Technology Co., Ltd., Nanjing, China) under transmitted light at 10–40× magnification. Raman spectra were acquired using a Horiba HR Evolution confocal Raman spectrometer (Kyoto, Japan). Analytical conditions included: 532 nm laser wavelength, 100 mW output power, 1–5 µm spot size, and 1 cm⁻¹ spectral resolution. Each spectrum represented 3 accumulations of 20 s scans. Spectral data (100–2000 cm⁻¹) were compared with reference spectra from the RRUFF mineral database.

3.2.2. Scanning Electron Microscope

Microstructural and chemical analyses were conducted using a Phenom XL benchtop SEM (Phenom-World B.V., Netherlands) equipped with an energy-dispersive X-ray spectroscopy (EDS) system. Two analytical approaches were employed: polished thin section analysis and fresh fracture surface examination. All samples underwent ultrasonic cleaning followed by carbon coating prior to analysis. The SEM operated at 15–20 kV accelerating voltage and 2–5 nA beam current under high vacuum (0.1 mbar), acquiring both backscattered and secondary electron images. The EDS system provided elemental analysis from B (5) to Cf (98).

3.2.3. Chemical Analysis

Major-element compositions (wt%) were determined using a JEOL JXA-8230 electron probe microanalyzer at the Shandong Institute of Geological Sciences, Jinan, China. The microprobe operated at 15 kV accelerating voltage, 10 nA beam current, and 1–10 μm beam diameter. Analyses were calibrated using natural mineral and synthetic standards: jadeite (Si, Na), almandine garnet (Al, Fe), diopside (Ca, Mg), sanidine (K), rutile (Ti), rhodonite (Mn), chromium oxide (Cr), and vanadium (V). ZAF corrections were applied using JEOL Delta NMR standard software.

Trace-element analyses were conducted using laser ablation-inductively coupled plasma mass spectrometry (LA-ICP-MS) at the Key Laboratory of Gold Mineralization Process and Resource Utilization (Ministry of Natural Resources) and Key Laboratory of Metallic Mineralization Geological Process and Resource Utilization, Shandong Institute of Geological Sciences. The system comprised a Resolution SE 193 nm excimer laser coupled to a Thermo Fisher Element XR ICP-MS. Helium carrier gas and argon makeup gas were mixed through a three-way connector prior to ICP introduction to optimize ablation sensitivity. Laser parameters included: 3–5 J/cm² energy density, 40 μm spot size, and 4 Hz repetition rate in single-spot mode. Each analysis comprised 10–20 s background acquisition followed by 50 s sample ablation. Quantification followed the multiple external standard method (Liu et al., 2008 [45]) using USGS reference glasses (BCR-2G, BHVO-2G) and NIST610 for calibration, with NIST610 monitoring instrumental mass bias.

3.2.4. Water Content Analysis

Doubly polished thin sections with a thickness ranging from 0.1 to 0.2 mm were prepared for microscopic Fourier transform infrared spectroscopy (Micro-FTIR). The water content was conducted at the Gemological Experimental Teaching Center, School of Gemology, China University of Geosciences (Beijing, China). Unpolarized spectra were obtained from 2800–4500 cm⁻¹ on a BRUKER TENSOR27 FTIR spectrometer coupled with a Continuum microscope. Tests were performed using the reflectance method at 18–25 °C and <70% humidity with a scan voltage of 85 to 265 V, and 32 scans were accumulated for each spectrum at a 4 cm⁻¹ resolution. Each mineral phase was measured at 9 to 12 points covering the whole section. OMNIC (9.7) software provided by the Thermo Nicolet Co. was used to collect and process the spectrum. H₂O in anhydrous minerals was calculated by the Beer–Lambert law:

C = A/(I × t × γ)

where C is the concentration of hydrogen species (ppm wt. H₂O), A is the integral area (cm⁻²) of absorption bands in the region of interest, I is the integral specific absorption coefficient of minerals (ppm⁻¹ cm⁻²), t is the thickness of the section (cm), and γ is the orientation factor; for minerals with anisotropy, γ = 1/3 [46]. In the study, integration of OH absorption region was 3000 to 3800 cm⁻¹. An integral specific coefficient of 5.32 ppm⁻¹ cm⁻² was used for olivine, 7.09 ppm⁻¹ cm⁻² for clinopyroxene, and 14.84 ppm⁻¹ cm⁻² for orthopyroxene to calculate the H₂O contents.

3.2.5. Isotope Analysis Methods

The δ¹⁸O isotope of the sample was analyzed at Kehui Testing Technology Co., Ltd. (Tianjin, China). The analytical system employs a Delta^Plus XP mass spectrometer to analyze collected oxygen using stable gas isotope ratios. The injection method was dual channel injection. The sample underwent bromine pentafluoride treatment to generate oxygen, which was then collected in a sample tube equipped with a 5 Å molecular sieve under vacuum conditions. The δ¹⁸O of the oxygen was measured on a 253 plus gas isotope mass spectrometer. The external precision of the standard sample was better than +0.2 ‰ with V-SMOW serving as the relative standard. The internal precision for a single sample test was 0.06‰. The laboratory utilizes GBW04421 as the working standard sample; this standard is a national first-class silicon isotope reference material, with a δ³⁰SiNBS-28 value of −0.02. After calibration with the international standard NBS-28, its δ¹⁸O_V-SMOW value was +10.92‰ (A detailed introduction to the instrumental conditions was provided in [47,48]).

4. Results

4.1. Raman Spectra

Micro-Raman spectroscopy was systematically employed to characterize the mineral phases and inclusions within the spinel lherzolite, revealing that olivine consisted of both forsterite (Figure 6a) and chrysolite (Figure 6d), whereas clinopyroxene was composed of diopside (Figure 6b) and augite (Figure 6c). Orthopyroxene was identified as enstatite (Figure 6e). Raman analysis of CO₂ inclusions was primarily conducted on forsterite hosted inclusions (Figure 6a), since the small size of inclusions in other phases precluded reliable spectral acquisition.

4.2. Major Elements

The Mg# of olivine of the studied xenoliths ranged from 89.72 to 91.01. Olivine displayed remarkably consistent major-element compositions, including SiO₂ (39.62–41.769 wt%), Al₂O₃ (0–0.05 wt%), MgO (47.73–49.36 wt%), and CaO (0–0.11 wt%) (

Table A1. Major-element concentrations in olivine from Damaping spinel lherzolite.

Sample	SiO2	Na2O	Cr2O3	TiO2	MgO	FeO	K2O	MnO	Al2O3	NiO	CaO	Total	Mg#
DMP202-1-OL-1	41.76	0.02	0.07	0.00	49.21	8.91	0.01	0.10	0.04	0.29	0.07	100.47	90.78
DMP202-1-OL-2	40.98	0.02	0.00	0.00	49.02	9.38	0.00	0.03	0.00	0.35	0.01	99.78	90.31
DMP202-1-OL-3	41.04	0.03	0.00	0.06	48.61	9.02	0.02	0.03	0.05	0.26	0.04	99.14	90.58
DMP202-1-OL-4	40.79	0.05	0.02	0.00	48.57	9.05	0.00	0.01	0.04	0.29	0.06	98.87	90.54
DMP202-1-OL-5	41.54	0.02	0.02	0.01	48.50	9.28	0.03	0.12	0.00	0.22	0.07	99.82	90.31
DMP204-1-OL-1	40.99	0.04	0.01	0.00	47.88	9.78	0.00	0.08	0.00	0.30	0.00	99.07	89.73
DMP204-1-OL-2	40.37	0.03	0.00	0.00	48.64	9.89	0.00	0.14	0.00	0.24	0.00	99.32	89.76
DMP204-1-OL-3	40.24	0.00	0.02	0.05	48.39	9.88	0.01	0.10	0.02	0.27	0.02	98.98	89.73
DMP204-1-OL-4	40.47	0.03	0.00	0.00	48.15	9.46	0.00	0.14	0.00	0.33	0.04	98.61	90.08
DMP204-1-OL-5	40.43	0.02	0.02	0.00	47.93	9.59	0.01	0.08	0.00	0.27	0.00	98.35	89.91
DMP204-2-OL-1	40.58	0.02	0.03	0.05	48.49	8.86	0.01	0.21	0.01	0.25	0.01	98.52	90.71
DMP204-2-OL-2	41.62	0.01	0.02	0.04	48.91	9.03	0.00	0.13	0.00	0.46	0.07	100.28	90.62
DMP204-2-OL-3	41.41	0.02	0.02	0.01	48.48	8.54	0.00	0.16	0.00	0.34	0.00	98.98	91.01
DMP204-2-OL-4	40.41	0.05	0.00	0.01	48.91	9.20	0.01	0.20	0.05	0.32	0.05	99.21	90.46
DMP204-2-OL-5	40.77	0.01	0.05	0.06	49.24	9.24	0.01	0.16	0.00	0.24	0.05	99.81	90.48
DMP208-1-OL-1	39.62	0.02	0.01	0.00	47.99	9.30	0.00	0.03	0.03	0.32	0.08	97.41	90.20
DMP208-1-OL-2	40.99	0.02	0.04	0.00	48.12	9.34	0.01	0.11	0.00	0.25	0.03	98.90	90.19
DMP208-1-OL-3	41.36	0.03	0.02	0.01	49.36	9.95	0.00	0.18	0.01	0.24	0.05	101.21	89.84
DMP208-1-OL-4	41.17	0.00	0.04	0.02	48.44	9.31	0.03	0.01	0.03	0.26	0.08	99.38	90.27
DMP208-1-OL-5	41.19	0.03	0.00	0.00	48.56	9.52	0.01	0.13	0.03	0.26	0.04	99.76	90.10
DMP208-2-OL-1	40.29	0.02	0.00	0.04	48.08	9.17	0.00	0.21	0.03	0.28	0.11	98.22	90.34
DMP208-2-OL-2	40.16	0.03	0.00	0.00	48.61	9.08	0.00	0.13	0.00	0.35	0.01	98.37	90.51
DMP208-2-OL-3	40.79	0.02	0.00	0.01	48.61	9.13	0.01	0.18	0.00	0.41	0.07	99.23	90.47
DMP208-2-OL-4	41.06	0.03	0.01	0.00	48.17	9.45	0.04	0.10	0.02	0.23	0.07	99.18	90.09
DMP208-2-OL-5	41.07	0.03	0.03	0.03	47.95	9.38	0.00	0.17	0.00	0.21	0.03	98.90	90.11

Mg# = molar 100 × Mg/(Mg + Fe).

). The Mg# of olivines were ranging from 89.73 to 91.01 (Figure 7a), indicating a transition from fertile mantle to transitional mantle . The CaO content, as a fusible element in olivine, was indicative of its genesis: magmatic olivine typically contains high CaO (>0.1%), while mantle-derived olivine typically contains less than 0.1% CaO. The low CaO content (0–0.11 wt%) in Damaping lherzolite indicated a magmatic origin. CaO distribution patterns further confirmed that the studied peridotites were primarily within the fertile mantle (Figure 7a).

Orthopyroxenes were En_0.47–0.51Fs_0.03–0.05Wo_0.44–0.49, classifying them as enstatite (Figure 8). Mg# and Cr# in orthopyroxenes ranged from 90.11 to 93.12 and 7.08 to 9.06, respectively (

Table A2. Major-element concentrations in orthopyroxenes from Damaping spinel lherzolite.

Sample	SiO2	Na2O	Cr2O3	TiO2	MgO	FeO	K2O	MnO	Al2O3	CaO	Total	Mg#	Cr#	En	Fs	Wo
DMP202-1-OPX-1	54.74	0.09	0.55	0.06	33.10	4.99	0.02	0.03	3.78	0.60	97.96	92.20	8.81	0.91	0.08	0.01
DMP202-1-OPX-2	55.65	0.08	0.47	0.03	32.52	5.49	0.01	0.16	3.77	0.61	98.79	91.35	7.73	0.90	0.09	0.01
DMP202-1-OPX-3	54.64	0.10	0.51	0.06	33.10	5.02	0.00	0.18	3.63	0.61	97.84	92.17	8.61	0.91	0.08	0.01
DMP202-1-OPX-4	55.89	0.06	0.54	0.07	33.12	4.88	0.04	0.01	3.82	0.68	99.12	92.37	8.67	0.91	0.08	0.01
DMP202-1-OPX-5	55.09	0.09	0.43	0.04	33.18	5.02	0.00	0.13	3.68	0.71	98.37	92.18	7.22	0.91	0.08	0.01
DMP204-1-OPX-1	56.00	0.12	0.37	0.12	33.27	5.54	0.00	0.09	3.91	0.73	100.15	91.46	5.98	0.90	0.09	0.01
DMP204-1-OPX-2	55.63	0.12	0.43	0.09	33.37	5.79	0.00	0.12	3.83	0.68	100.06	91.13	7.03	0.90	0.09	0.01
DMP204-1-OPX-3	56.41	0.10	0.51	0.16	33.94	5.84	0.00	0.13	3.90	0.64	101.64	91.20	8.06	0.90	0.09	0.01
DMP204-1-OPX-4	55.95	0.07	0.36	0.06	33.71	6.07	0.00	0.05	3.84	0.58	100.68	90.82	5.88	0.90	0.09	0.01
DMP204-2-OPX-1	56.00	0.12	0.37	0.12	33.27	5.54	0.00	0.09	3.91	0.73	100.15	91.46	5.98	0.90	0.09	0.01
DMP204-2-OPX-2	55.63	0.12	0.43	0.09	33.37	5.79	0.00	0.12	3.83	0.68	100.06	91.13	7.03	0.90	0.09	0.01
DMP204-2-OPX-3	56.41	0.10	0.51	0.16	33.94	5.84	0.00	0.13	3.90	0.64	101.64	91.20	8.06	0.90	0.09	0.01
DMP204-2-OPX-4	55.95	0.07	0.36	0.06	33.71	6.07	0.00	0.05	3.84	0.58	100.68	90.82	5.88	0.90	0.09	0.01
DMP208-1-OPX-1	55.27	0.08	0.44	0.15	33.08	5.02	0.01	0.26	3.81	0.68	98.79	92.16	7.18	0.91	0.08	0.01
DMP208-1-OPX-2	55.85	0.07	0.44	0.17	33.02	5.13	0.00	0.20	3.84	0.63	99.36	91.99	7.19	0.91	0.08	0.01
DMP208-1-OPX-3	55.23	0.05	0.39	0.15	33.36	5.22	0.01	0.10	3.90	0.60	99.00	91.94	6.27	0.91	0.08	0.01
DMP208-1-OPX-4	56.39	0.05	0.43	0.11	33.61	5.16	0.00	0.13	3.88	0.58	100.34	92.07	6.85	0.91	0.08	0.01
DMP208-1-OPX-5	55.78	0.07	0.43	0.12	33.27	4.98	0.02	0.17	3.84	0.66	99.34	92.25	6.95	0.91	0.08	0.01
DMP208-2-OPX-2	54.49	0.09	0.43	0.02	33.01	6.03	0.00	0.10	3.64	0.67	98.46	90.71	7.33	0.89	0.09	0.01
DMP208-2-OPX-3	54.72	0.08	0.45	0.04	33.09	5.40	0.00	0.08	3.73	0.75	98.33	91.62	7.50	0.90	0.08	0.01
DMP208-2-OPX-4	55.33	0.10	0.41	0.16	33.36	5.76	0.00	0.12	3.85	0.69	99.77	91.18	6.72	0.90	0.09	0.01

Cr# = molar 100 × Cr/(Cr + Al); En = Mg/(Fe + Mn + Mg + Ca); Fs = (Fe + Mn)/(Fe + Mn + Mg + Ca); Wo = Ca/(Fe + Mn + Mg + Ca).

). A weak negative correlation existed between Mg# and Al₂O₃ content (Figure 7b), consistent with fertile mantle compositions.

This study focused exclusively on analyzing Cpx-1 grains, which represent pristine domains unaffected by metasomatic alteration, to obtain accurate measurements of their major elements, trace elements, and water contents. Clinopyroxenes were En_0.50–0.91Fs_0.04–0.10Wo_0.01–0.46, indicating that most were diopside and a few were augite (Figure 8). Mg# and Cr# in clinopyroxenes ranged from 91.89 to 93.51 and 8.59 to 14.57, respectively. The clinopyroxenes in the Damaping spinel lherzolite exhibited high Al₂O₃ content (4.60–6.72 wt%) (

Table A3. Major-element concentrations in clinopyroxenes from Damaping spinel lherzolite.

Sample	SiO2	Na2O	Cr2O3	TiO2	MgO	FeO	K2O	MnO	Al2O3	CaO	Total	Mg#	Cr#	En	Fs	Wo
DMP202-1-CPX-1	52.96	1.10	1.14	0.15	16.22	2.38	0.00	0.09	4.60	20.61	99.25	92.39	6.55	0.50	0.04	0.46
DMP202-1-CPX-2	52.20	1.03	1.20	0.17	15.87	1.96	0.03	0.01	4.70	21.11	98.27	93.51	14.64	0.49	0.03	0.47
DMP202-1-CPX-3	52.68	1.13	1.11	0.19	16.34	2.22	0.00	0.01	4.92	21.21	99.81	92.92	13.10	0.50	0.04	0.46
DMP204-1-CPX-1	52.22	1.56	0.95	0.63	15.50	2.35	0.00	0.05	6.20	19.39	98.85	92.15	9.36	0.50	0.04	0.45
DMP204-1-CPX-2	51.70	1.57	0.84	0.68	14.59	2.28	0.00	0.10	5.95	21.27	98.97	91.94	8.62	0.47	0.04	0.49
DMP204-1-CPX-3	52.35	1.84	0.97	0.55	14.58	2.18	0.00	0.14	6.72	20.81	100.13	92.28	8.85	0.47	0.04	0.49
DMP204-2-CPX-1	52.62	1.32	1.02	0.26	16.23	2.17	0.01	0.08	5.36	20.59	99.66	93.03	11.29	0.50	0.04	0.46
DMP204-2-CPX-2	52.53	1.27	0.94	0.33	15.76	2.23	0.00	0.13	5.28	20.90	99.37	92.66	10.70	0.49	0.04	0.47
DMP204-2-CPX-3	52.75	1.35	1.23	0.31	15.88	2.22	0.00	0.00	5.51	20.58	99.83	92.72	13.04	0.50	0.04	0.46
DMP208-1-CPX-1	53.30	1.34	1.11	0.38	15.82	2.34	0.02	0.00	5.43	20.51	100.25	92.33	12.07	0.50	0.04	0.46
DMP208-1-CPX-2	51.72	1.33	1.09	0.35	15.91	2.37	0.04	0.03	5.47	20.58	98.88	92.28	11.81	0.50	0.04	0.46
DMP208-1-CPX-3	52.41	1.18	1.29	0.43	15.68	2.39	0.00	0.14	5.95	21.15	100.63	92.14	12.74	0.49	0.04	0.47
DMP208-1-CPX-4	52.95	1.28	1.04	0.40	15.79	2.09	0.04	0.10	5.38	21.09	100.16	93.08	11.52	0.49	0.04	0.47
DMP208-1-CPX-5	52.07	1.13	1.04	0.38	15.87	2.08	0.00	0.07	5.26	20.63	98.53	93.16	11.68	0.50	0.04	0.46
DMP208-1-CPX-6	52.44	1.31	0.88	0.29	16.01	2.49	0.00	0.06	5.18	20.45	99.10	91.98	10.25	0.50	0.04	0.46
DMP208-2-CPX-1	52.27	1.23	1.17	0.33	15.79	2.22	0.00	0.07	5.07	21.04	99.18	92.70	13.41	0.49	0.04	0.47
DMP208-2-CPX-2	53.54	1.15	1.24	0.41	16.09	2.14	0.00	0.05	5.09	20.12	99.83	93.06	14.06	0.51	0.04	0.46
DMP208-2-CPX-3	51.97	1.20	0.82	0.29	15.86	2.28	0.01	0.03	5.13	20.95	98.54	92.55	9.69	0.49	0.04	0.47
DMP208-2-CPX-4	53.63	1.24	0.91	0.30	16.17	2.28	0.01	0.06	4.98	20.10	99.67	92.66	10.89	0.51	0.04	0.45

).

Spinel in the Damaping spinel lherzolite exhibited large variations in Cr₂O₃ (12.19–21.12 wt%) and Cr# (1.64 to 20.08). Mg# in spinel ranged from 73.11 to 76.36. Based on Cr# values, spinel minerals in mantle peridotites were delineated, with all experimental samples classified as low-Cr spinel (

Table A4. Major-element concentrations in spinels from Damaping spinel lherzolite.

Sample	SiO2	Na2O	Cr2O3	TiO2	MgO	FeO	K2O	MnO	Al2O3	CaO	Total	Mg#	Cr#
DMP202-1-SP-1	0.12	0.02	19.19	0.15	17.43	10.70	0.02	0.12	45.65	0.01	93.41	74.38	22.00
DMP202-1-SP-3	0.10	0.00	20.95	0.07	18.72	11.12	0.01	0.11	48.43	0.00	99.50	75.01	22.50
DMP202-1-SP-5	0.12	0.02	21.12	0.08	18.98	10.71	0.00	0.13	49.67	0.00	100.84	75.96	22.20
DMP204-1-SP-1	0.08	0.07	12.19	0.15	19.52	10.77	0.03	0.18	57.42	0.02	100.41	76.36	12.47
DMP204-1-SP-2	0.05	0.04	12.59	0.19	19.49	11.47	0.02	0.07	56.85	0.00	100.77	75.18	12.94
DMP204-2-SP-1	0.07	0.02	18.07	0.08	18.98	11.10	0.00	0.10	50.44	0.02	98.87	75.30	19.38
DMP204-2-SP-2	0.11	0.02	17.76	0.20	19.29	10.89	0.00	0.08	51.16	0.00	99.50	75.94	18.90
DMP208-1-SP-1	0.10	0.06	17.56	0.22	19.16	11.31	0.01	0.07	51.46	0.01	99.96	75.13	18.63
DMP208-1-SP-2	0.08	0.03	19.88	0.19	18.92	11.51	0.00	0.07	49.89	0.00	100.57	74.56	21.10
DMP208-1-SP-3	0.06	0.00	19.38	0.23	19.09	11.89	0.00	0.13	49.66	0.00	100.43	74.11	20.75
DMP208-2-SP-1	0.05	0.03	20.41	0.18	18.44	10.98	0.01	0.10	48.04	0.00	98.24	74.97	22.18
DMP208-2-SP-2	0.07	0.01	20.70	0.21	18.66	12.24	0.00	0.03	48.52	0.02	100.46	73.11	22.26
DMP208-2-SP-3	0.07	0.01	18.82	0.17	19.28	11.85	0.00	0.03	49.36	0.00	99.59	74.37	20.37

).

Whole-rock compositions were reconstructed based on mineral modes (Table 1) and the composition of the minerals (Table A1, Table A2, Table A3 and Table A4). The high Al₂O₃ content primarily resulted from the high spinel proportion within the rock (

Table A5. Major-element compositions of the whole rocks from Damaping spinel lherzolite.

Sample	SiO2	Na2O	Cr2O3	TiO2	MgO	FeO	K2O	MnO	Al2O3	CaO	Total	Mg#
DMP202-1	46.40	0.11	0.20	0.04	41.64	7.39	0.01	0.07	1.57	1.52	99.13	90.94
DMP204-1	43.71	0.21	1.00	0.12	38.48	7.92	0.00	0.11	5.41	2.33	99.45	89.65
DMP204-2	46.80	0.14	0.62	0.08	40.00	7.32	0.00	0.13	2.93	1.68	99.87	90.70
DMP208-1	45.66	0.21	1.34	0.11	33.82	6.46	0.01	0.10	4.82	3.14	95.78	90.32
DMP208-2	43.45	0.22	1.23	0.09	37.79	7.34	0.01	0.12	4.04	3.33	97.77	90.18

). Reconstruction of the whole-rock major elements showed that most of the Damaping lherzolite were quite fertile, with only one sample (DMP202-1) overlapping the refractory region (Figure 9a). A similar pattern is shown in Figure 9b, with the Damaping peridotite being close to the fertile peridotites. The Damaping lherzolite plots within the fertile mantle field in Figure 9a,b shows a similar pattern that is consistent with fertile mantle compositions.

4.3. Trace Elements

Clinopyroxene served as the main trace-element carrier in peridotite xenoliths, reflecting the trace-element signature of the whole rock [54]. A total of 15 clinopyroxene occurrences were analyzed in this study (

Table A6. Trace-element concentrations in clinopyroxenes from Damaping spinel lherzolite (in ppm).

Sample	DMP202-1-1	DMP202-1-2		DMP204-1-2		DMP204-1-3		DMP204-2-2		DMP204-2-3		DMP208-1-1		DMP208-1-2
Sc	100.64	111.24		99.75		95.61		91.75		96.59		93.73		95.25
Ti	1470	1720		6780		5500		3270		3090		3790		3470
V	332.04	333.83		323.19		317.02		326.51		323.44		322.03		319.57
Cr	10,464.22	9400.62		8128.57		8317.80		10,346.61		10,108.11		8964.65		9521.37
Co	27.62	28.87		21.51		24.58		26.20		26.79		27.34		27.35
Ni	1011.07	1070.66		906.16		840.56		627.28		626.59		676.58		647.32
Cu	0.24	4.84		11.90		14.60		4.05		2.92		3.89		2.48
Zn	10.42	11.01		6.74		6.52		10.20		9.94		11.14		10.76
Ga	3.08	3.51		4.65		4.35		3.58		3.41		3.99		4.00
Rb	0.00	0.00		0.08		0.12		0.00		0.00		0.00		0.00
Ba	0.09	0.00		0.00		0.00		0.15		0.12		0.12		0.05
Nb	0.54	0.49		0.04		0.13		0.50		0.53		0.51		0.48
La	1.00	0.99		0.60		0.63		1.27		1.29		1.54		1.49
Ce	2.95	2.79		2.69		2.74		0.00		0.00		0.00		0.00
Pb	0.09	0.04		0.03		0.02		0.10		0.07		0.07		0.08
Pr	0.43	0.47		0.80		0.81		0.79		0.73		1.01		1.00
Sr	56.27	57.99		62.49		70.17		65.65		66.72		86.16		85.82
Nd	2.51	3.12		5.46		5.95		4.34		4.42		6.21		6.51
Zr	13.23	14.22		42.91		44.36		24.62		25.30		27.18		29.08
Hf	0.43	0.28		1.60		1.27		0.77		0.77		0.94		0.98
Sm	0.92	1.08		2.46		2.70		1.38		1.78		2.28		2.29
Eu	0.27	0.42		1.15		1.16		0.64		0.65		0.76		0.75
Gd	1.13	1.10		3.92		3.29		1.98		2.52		2.44		2.48
Tb	0.31	0.25		0.60		0.70		0.41		0.43		0.42		0.44
Dy	2.33	2.31		4.30		4.21		3.10		3.30		2.77		2.84
Y	12.91	13.46		25.65		25.85		17.24		18.31		16.62		17.02
Ho	0.54	0.61		0.95		1.17		0.67		0.67		0.58		0.64
Er	1.59	1.54		2.84		2.76		1.92		2.16		1.89		2.04
Tm	0.17	0.21		0.40		0.40		0.25		0.35		0.23		0.24
Yb	1.77	1.59		2.05		1.74		1.73		1.85		1.62		1.83
Lu	0.26	0.18		0.38		0.35		0.25		0.28		0.28		0.22
Sample	DMP208-1-3		DMP208-1-4		DMP208-1-5		DMP208-1-6		DMP208-2-2		DMP208-2-3		DMP208-2-4
Sc	82.81		92.92		94.82		99.05		82.47		99.13		108.78
Ti	4310		3970		3790		2920		4070		2850		2950
V	309.15		322.15		316.92		326.69		319.77		321.34		331.65
Cr	11,873.09		10,912.96		11,253.50		8890.56		8759.32		9442.91		10,770.85
Co	25.74		27.06		27.04		27.61		28.28		28.44		27.87
Ni	690.06		701.67		672.17		705.92		884.13		705.70		731.64
Cu	2.74		4.94		2.11		4.22		9.31		4.78		3.01
Zn	9.93		10.35		10.51		11.08		10.72		10.42		10.94
Ga	4.21		4.22		4.14		3.75		4.41		4.41		3.99
Rb	0.00		0.00		0.00		0.00		0.07		0.00		0.00
Ba	0.07		0.00		0.01		0.00		0.26		0.00		0.00
Nb	0.60		0.55		0.59		0.48		0.60		0.55		0.60
La	1.58		1.60		1.49		1.61		1.35		1.62		1.66
Ce	0.00		0.00		0.00		0.00		0.00		0.00		0.00
Pb	0.08		0.08		0.06		0.16		0.09		0.05		0.06
Pr	1.14		1.01		1.05		1.02		1.00		0.98		1.08
Sr	85.74		85.33		86.20		88.23		86.89		87.17		90.32
Nd	6.55		6.50		7.04		6.15		5.90		6.64		6.49
Zr	29.83		30.18		31.49		28.81		25.76		30.27		34.15
Hf	0.74		1.10		1.01		0.89		0.79		0.95		0.90
Sm	2.13		2.22		2.33		2.29		1.80		2.23		1.84
Eu	0.84		0.74		0.87		0.82		0.79		0.67		0.67
Gd	2.88		2.71		2.93		2.49		2.07		2.76		2.82
Tb	0.42		0.40		0.51		0.39		0.36		0.40		0.52
Dy	3.13		2.89		3.37		3.15		2.32		2.92		3.04
Y	17.97		17.32		18.25		17.00		13.84		16.54		17.91
Ho	0.69		0.61		0.67		0.61		0.50		0.60		0.67
Er	2.00		2.12		1.94		1.62		1.60		2.15		2.02
Tm	0.27		0.23		0.26		0.29		0.20		0.25		0.25
Yb	1.73		1.94		1.93		1.70		1.26		1.70		1.63
Lu	0.25		0.27		0.32		0.29		0.18		0.24		0.27

). The REE abundances in clinopyroxenes from Damaping mantle xenoliths exhibited chondrite-normalized patterns characterized by LREE depletion and flat HREE ((La/Yb)_N = 0.20–0.73, (Sm/Yb)_N = 0.561.68, (La/Ce)_N = 0.58–0.92) (Figure 10a). In the primitive-mantle-normalized trace-element diagrams, a trend of decreasing incompatibility from left to right is observed, accompanied by slight to moderate negative anomalies for Ba, Pb, Hf, Eu, Tb, and Lu and positive anomalies for Sr, Zr, and Y (Figure 10b).

4.4. Equilibrium Temperature of the Mantle Xenoliths

To minimize systematic uncertainties, equilibration temperatures of Damaping spinel lherzolite were estimated using three geothermometers: the two-pyroxene geothermometer (T_Taylor) [57], the Ca-Na-in-orthopyroxene-clinopyroxene geothermometer (T_BKN) [58], and the olivine-clinopyroxene geothermobarometer (T_Ca-ol/cpx) [58]. The olivine-clinopyroxene geothermobarometer can be used to estimate pressure. Furthermore, the olivine-clinopyroxene geothermobarometer constrains equilibrium pressures to a range of 19 to 29 kbar (

Table 2. Equilibration temperature (°C) and pressure (kbar) for the Damaping spinel lherzolite.

Samples	TTaylor	TBKN	TCa-ol/cpx	Pkb
202-1	1085	1380	1232	22
204-1	963	1307	1209	22
204-2	1068	1392	1199	29
208-1	1064	1390	1209	26
208-2	1071	1391	1213	19

T_Taylor: method by Taylor et al., 1998 [57]; T_BKN, T_Ca-ol/cpx, and P_kb: methods by Köhler and Brey, 1990 [58].

), corresponding to mantle depths of 60 to 93 km.

4.5. Water Contents

All the clinopyroxenes and orthopyroxenes analyzed here exhibited several absorption bands in the typical OH stretching vibration region (2800–4000 cm⁻¹). We selected only intact cpx-1 grains for water content measurements to ensure representativeness of the data. In contrast, olivine exhibited no prominent OH bands (Figure 11c), with water contents below the minimum detectable value. Spinel samples exhibited no significant OH band due to the strong inhibition by Fe³⁺. Representative infrared spectra for clinopyroxene, orthopyroxene, and olivine from Damaping lherzolite are shown in Figure 9. The OH absorption bands of clinopyroxene could be divided into three groups: 3630−3620 cm⁻¹, 3550−3530 cm⁻¹, and 3470−3450 cm⁻¹ (Figure 11a). Three prominent OH absorption bands from orthopyroxene were: 3600−3580 cm⁻¹, 3520−3510 cm⁻¹, and 3420−3410 cm⁻¹ (Figure 11b). These absorption bands are similar to those reported by previous researchers [14,22,59,60,61]. Analysis of the infrared spectral profiles of the large-grained pyroxene revealed almost identical OH absorption peaks from the core to the rim.

The calculated water contents of Damaping spinel lherzolite ranged from 13 to 19 ppm for clinopyroxenes and from 5 to 8 ppm for orthopyroxenes. Notably, the sharp boundaries between clinopyroxene and reaction rims had no significant influence on water contents. The water contents of olivine were below the detection limit (<2 ppm) but could be approximately estimated from the water contents of pyroxene and the previously reported partition coefficients between pyroxene and olivine. The coefficients of water partitioning between pyroxene and olivine measured in experimental studies were highly variable. The values obtained at low pressure (<3 GPa) were much higher than those at high pressure (>8 GPa [18,62,63,64,65]). Calculations showed that, for Damaping samples obtained at low pressure (Table 2), the H₂O partition coefficient between pyroxene and olivine was consistent with low pressure experiments (D_cpx/ol > 10 [60,63,64,65,66]). Following the same method as previously used for peridotites from the NCC [14] we used the D_cpx/ol = 10 to calculate H₂O contents of olivine, which likely represented maximum estimates. The results showed that olivine water contents ranged from 1 to 2 ppm. Spinel had little effect on the water content of the whole rock due to its low abundance. The whole-rock water content was calculated based on the modal abundances of olivine, clinopyroxene, and orthopyroxene (Table 1) and their water contents (

Table 3. Water contents from Damaping spinel lherzolite.

Sample	Water Content (ppm)
	Cpx	Opx	Ol a	WR b
DMP202-1	14	5	1	3
DMP204-1	19	8	2	5
DMP204-2	14	7	1	4
DMP208-1	16	6	2	5
DMP208-2	13	6	1	4

Ol, olivine; Opx, orthopyroxene; Cpx, clinopyroxene; Sp, spinel. ^a Olivine water content was estimated as a water content of ol calculated as 1/10 of Cpx. ^b Whole rock (WR): water contents were reconstructed from measured water contents in minerals and estimated mineral modes (Table 1).

). The whole-rock water contents of Damaping spinel lherzolite ranged from 3 to 5 ppm.

4.6. δ¹⁸O Values

Oxygen isotope analyses were performed on clinopyroxenes from Damaping spinel lherzolite. The δ¹⁸O values of clinopyroxenes ranged from 5.27 to 5.59‰ (mean = 5.40‰) (

Table 4. Oxygen isotope composition of olivine and clinopyroxene from the Damaping spinel lherzolite. The δ¹⁸O values of mantle peridotite from Mattey, D. et al., 1994 [67], Xu et al., 1996 [69,70], and Chaozot et al., 1997 [69,70]. The δ¹⁸O values of MORB are from Elier et al., 2001 [68].

Sample	δ18OCpx‰
Mantle peridotite	5.25–5.90
Mid-ocean ridge basalts	5.3–5.7
DMP202	5.59
DMP204	5.34
DMP208	5.27

), falling within the range of mantle peridotites [67] and consistent with those of mid-ocean ridge basalts (MORB; δ¹⁸O = 5.0–5.7‰ [68]).

5. Discussion

5.1. Melt Depletion and Enrichment Processes

5.1.1. Partial Melting

The clinopyroxenes in the Damaping peridotites had high Al₂O₃ contents (4.60–6.72 wt%). The covariant relationship between Al₂O₃ content and Mg# of clinopyroxene can be used to estimate the degree of melt extraction from the peridotite [5]. The negative correlation between Mg# and Al₂O₃ content suggests that the peridotite is less subject to melt extraction (Figure 7c). Like Mg# in silicates, Cr# in spinel is often used to indicate the degree of partial melting of olivine. Higher Cr# values indicate a higher degree of partial melting, while the lower Cr# values of spinel in this study suggest a low degree of melting. By plotting Mg# vs. Cr# values, the data all fell within the fertile region and exhibited a negative correlation (Figure 7d).

Depletion of clinopyroxene in LREE is indicative of partial melting of the mantle. The positive correlations among these elements suggest they were all removed by partial melting and were not affected by metasomatic enrichment (Figure 12) [71]. The HREE concentration of clinopyroxene is generally less affected by late metasomatism [72,73,74] and can be used to estimate the degree of partial melting of the peridotite based on the Y and Yb contents [72]. Estimates of the degree of partial melting using the Y and Yb contents in clinopyroxene 1–6% batch partial melting and 1–5% fractional partial melting (Figure 13). Furthermore, the REE patterns in clinopyroxene provide additional constraints on partial melting degrees, indicating approximately 5% melting for the Damaping spinel lherzolites (Figure 10).

5.1.2. Silicate Melt Metasomatism

Clinopyroxene from the Damaping spinel lherzolite exhibited characteristic HREE depletion patterns accompanied by negative Hf anomalies. The absence of HFSE depletion in associated basalts [74] precludes host rock-derived metasomatism, suggesting mantle-derived fluid metasomatism [75]. Previous studies suggested that mantle enrichment primarily results from metasomatism by alkali-rich silicate melts/fluids [76,77] and CO₂ + H₂O-rich carbonate melts [78]. It is generally accepted that a good correlation of incompatible elements with La indicates siliciclastic metasomatism, while a good correlation of Ti and V with La suggests carbonatite magma metasomatism. In Damaping clinopyroxene, La showed correlations with Sr, Ce, and Nd but weak correlations with V, indicating that silicate melts dominated the metasomatic process. Clinopyroxene affected by silicate melt displayed low (La/Yb)_N (0.20–0.73) and high Ti/Eu (3546.98–5919.48). This indicated silicate melt-dominated mantle metasomatism (Figure 14) with minimal modification water content. The presence of the CO₂ phase gas–liquid inclusions in olivine and pyroxene further suggests weak carbonate modification.

In summary, the integrated petrological and geochemical evidence demonstrates that: (1) the Damaping spinel lherzolites represent moderately refractory mantle material (Mg# = 89.73–91.01); (2) they have experienced 1–6% partial melting; and (3) the mantle subsequently underwent significant silicate melt metasomatism.

5.2. Preservation of the Initial Water Contents in the Mantle Source

Previous studies indicated that olivine may experience a certain degree of hydrogen (H) diffusion loss as the host magma ascends to the Earth’s surface [79], while pyroxene maintains relatively stable hydrogen concentrations, effectively preserving mantle-derived water signatures [21,60,61,80,81,82,83,84]. This differential behavior primarily results from the differences in hydrogen diffusion rates between olivine and pyroxene, with olivine exhibiting weaker diffusion strength compared to pyroxene [85]. During xenolith ascent with magma, olivine is more prone to disruption, resulting in the diffusion loss of water. In contrast, pyroxenes demonstrate greater water retention capacity during ascent, making their water content a reliable indicator of bulk rock water characteristics.

A positive correlation exists between clinopyroxene and orthopyroxene water contents (Figure 15), with a trend line slope of approximately 2.3, indicating water equilibrium between clinopyroxene and orthopyroxene. This equilibrium indicates that pyroxenes effectively preserve mantle-derived water signatures and represent the dominant hydrous phases in the upper mantle. Olivines exhibit negligible water contents, reflecting either intrinsic anhydrous compositions or complete hydrogen loss during xenolith ascent. Spinel-phase peridotites contain high water content (290 ppm) [59], while the Damaping spinel lherzolite peridotites exhibit a systematic decrease in water content, correlating with higher pyroxene modal abundances in spinel-facies peridotites [60,61,83].

5.3. Variations in H₂O Contents Controlled by Partial Melting

The heterogeneous water contents in the Damaping spinel lherzolite beneath the interior of the NCC are likely the result of variations in the redox state, metasomatism, or partial melting.

The redox state is believed to reduce the solubility of water in peridotite minerals, resulting in the expulsion of more than half of the initial water content [63]. The redox state is reflected in the Fe³⁺/∑Fe ratios of spinels, which serve as a reliable proxy for the redox state in Damaping spinel lherzolite. The water content of peridotite in this region showed no linear relationship with Fe³⁺/∑Fe (Figure 16), indicating that water content heterogeneity in Damaping spinel lherzolite is independent of the redox state.

Damaping spinel lherzolite underwent silicate metasomatism. A comparison of the water content of pyroxene and clinopyroxene (La/Yb)_N ratios revealed no significant correlation (Figure 17), indicating that metasomatic processes have not significantly influenced water content heterogeneity in Damaping spinel lherzolite.

Hydrogen behaves as a highly incompatible element during mantle melting and fractional crystallization, with partition coefficients intermediate between La and Ce [86]. The Mg# of olivine serves as a robust indicator of partial melting degrees in spinel lherzolite, while the Yb content in clinopyroxene remains relatively stable during metasomatism, making it another reliable indicator for assessing the extent of partial melting [71,72,87]. The observed correlations among water content, Mg# of olivine, and Yb content of clinopyroxene demonstrated that water distribution was primarily controlled by partial melting processes. In this study, pyroxene water contents exhibited a negative correlation with olivine Mg# and a positive correlation with clinopyroxene Yb concentrations (Figure 18). These correlations reflect the highly incompatible behavior of water (~0.1), compared to Ce [88], and its tendency to enter the melt phase during melting. As a result, the residual solid phase becomes progressively depleted in H₂O with increasing degrees of partial melting [89]. The whole-rock water content of the Damaping spinel lherzolite (3.21–5.44 ppm) was significantly lower than that of primitive mantle spinel-phase lherzolite (290 ppm) [59], consistent with partial melting residue characteristics and demonstrating that water content is primarily controlled by partial melting processes. Therefore, the heterogeneity of the water contents from the Damaping spinel lherzolite is primarily controlled by partial melting rather than redox state and metasomatism.

5.4. Modification of the Lithospheric Mantle Beneath the Interior of the North China Craton

The δ¹⁸O values of clinopyroxenes from Damaping spinel lherzolites ranged from 5.27 to 5.59‰ (Table 4), falling within the range of mid-ocean ridge basalts (MORB; δ¹⁸O = 5.0–5.7‰) [69]. This similarity suggests that the mantle source of Damaping lherzolites shares isotopic characteristics with the MORB reservoir. However, the whole-rock water contents of Damaping spinel lherzolites (3–5 ppm) (Table 3) were significantly lower than those of the MORB source (50–250 ppm) [22]. This observation necessitates an examination of the potential mechanisms responsible for water depletion in the lithospheric mantle.

Two potential mechanisms could explain the lower δ¹⁸O values in MORB: (1) extreme magmatic differentiation under ultrahigh temperatures (>1500 °C), which could reduce δ¹⁸O through kinetic fractionation [90], and (2) high degrees of partial melting (>20%), which could preferentially extract low-δ¹⁸O melts [90]. However, neither mechanism is applicable to the Damaping spinel lherzolites. First, the calculated mantle equilibration temperatures are inconsistent with the ultrahigh-temperature conditions required for extreme magmatic differentiation. Second, melt extraction models indicate minimal melting degrees (1–6%), ruling out significant partial melting as a contributing factor. The δ¹⁸O values of clinopyroxenes from Damaping lherzolites (5.27–5.59‰) fell within the normal range for MORB (5.0–5.7‰), indicating that neither extreme magmatic differentiation nor high degrees of partial melting have significantly affected the Damaping spinel lherzolite.

The low water contents of Damaping spinel lherzolites are likely attributable to lithospheric reheating caused by asthenospheric upwelling during the Late Mesozoic–Early Cenozoic. This process would have facilitated hydrogen diffusion from nominally anhydrous minerals (NAMs), particularly olivine, which exhibits faster hydrogen diffusivity compared to pyroxenes [91]. The near-zero H₂O contents in olivine and reduced values in pyroxenes reflect partial dehydration during mantle residence. In contrast, oxygen isotopic ratios, which are less sensitive to short-term thermal events due to slower diffusion rates [68], retain primary MORB-like signatures. This preservation of MORB-like δ¹⁸O values, despite significant water loss, highlights the contrasting responses of volatiles and stable isotopes to mantle processes.

The extensive thermal perturbation associated with Mesozoic–Early Cenozoic lithospheric thinning in the NCC likely significantly altered the water budget of the mantle beneath Damaping. We suggest that the low H₂O contents observed in Damaping spinel lherzolites are primarily a consequence of lithospheric reheating driven by asthenospheric upwelling. Based on the geophysical data regarding crustal thickness, the Damaping peridotites exhibited an average Mg#-Ol = 90.1, while the upwelling of the asthenosphere has led to lithospheric destruction and thinning (Xia et al., 2020) [92]. This thermal erosion process has resulted in reduced water contents in the Damaping spinel lherzolites. If this interpretation holds, the current lithospheric mantle beneath Damaping largely consists of relict ancient mantle that experienced thinning during the Late Mesozoic–Early Cenozoic. This model is further corroborated by Re–Os isotopic data from peridotite xenoliths in Cenozoic basalts, which reveal a lithosphere dominated by ancient Proterozoic residues, with localized inputs of younger asthenospheric material near deep faults [11,93,94,95].

6. Conclusions

In this study, the petrology, mineralogy, and geochemistry of the Damaping lherzolite reveals the water content, properties, and evolution processes of the lithospheric mantle beneath the NCC. The main conclusions are as follows:

(1)

The Damaping spinel lherzolite shows moderately refractory characteristics (Mg# = 89.73–91.01). The lithospheric mantle has undergone 1–6% partial melting.

(2)

Water contents of clinopyroxene and orthopyroxene in the xenoliths are 13 to 19 ppm and 5 to 8 ppm. The calculated whole-rock water content ranges from 3 to 5.44 ppm. The heterogeneity in the water content of the lithospheric mantle is attributed to the effects of partial melting.

(3)

The MORB-like δ¹⁸O values (5.27–5.59‰) of Damaping spinel lherzolites reflect a mantle source similar to mid-ocean ridge basalts. Low water contents may result from Late Mesozoic–Early Cenozoic lithospheric reheating, which drove hydrogen diffusion from minerals like olivine.