Petrogenesis of Tholeiitic Basalts from CZK06 Drill Core on the Tianchi Volcano, China–North Korea Border

Abstract

To constrain Tianchi Volcano basalt petrogenesis, this study focuses on tholeiitic basalts from the CZK06 drill core on the northern slope. Using elemental geochemistry and Mg isotope analyses, we investigate magma evolution, petrogenesis, and mantle source properties. The tholeiitic basalts formed during the Pliocene-Early Pleistocene shield-forming stage, recording three stages of basaltic volcanism (Phases I to III). Classified as sodium-series basalts, they exhibit geochemical affinities with EM1-type OIB. Their δ²⁶Mg values (−0.420‰ to −0.150‰) show a substantially wider range than N-MORB. Their geochemical compositions are primarily controlled by source region characteristics and partial melting degree, with minor additional influences from fractional crystallization and crustal contamination. Fractional crystallization intensity shows a progressive increase from Phase I to III. Integrated with geochemical tracing studies of Changbaishan basalts, we propose that the tholeiitic basalts are derived predominantly from the partial melting of carbonatized pyroxenite, which originated from subducted ancient clay-rich altered oceanic crust. The carbonate melts driving the carbonatization were generated by low-pressure melting of recent oceanic sediments, transported by the deeply subducted carbonate-rich Pacific Plate within the Mantle Transition Zone. The tholeiitic magma formed in the Low-Velocity Zone at depths of 160–180 km beneath the lithospheric mantle.

1. Introduction

The Changbai Mountain Volcanic Field (CMVF) is situated along the China–North Korea border, approximately 1400 km from the Japan Trench. Tectonically located on the northeastern margin of the North China Craton, it presently lies within a NE-NNE trending uplift region between the Japan Sea Back-arc Basin and the Songliao Basin (Figure 1a) [1,2]. Geophysical data reveal low-resistivity bodies within the crust and upper mantle above upper interface of the Mantle Transition Zone (MTZ; ~410 km deep) beneath the CMVF [3]. These bodies are associated with the volcanic magma system, indicating an upwelling of damp-hot asthenosphere materials. Consequently, the CMVF is widely recognized as a prototypical back-arc intraplate volcano and has been extensively studied in relation to the deep subduction of the Pacific Plate and mantle convection within the big mantle wedge beneath Northeast Asia [3,4].

The Tianchi volcano (TCV) is one of the three major volcanoes within the vast CMVF (Figure 1b) [1,2]. It is composed of Pliocene-Early Pleistocene basalts, Pleistocene trachyte, and Holocene pyroclastic deposits (Figure 1b). For the basalts, significant advancements have been achieved in understanding their activity history [2], magma evolution [5,6], as well as petrogenesis and geodynamic mechanisms [2,4,7]. However, due to the heterogeneity of the mantle source region and inherent limitations in geochemical and geophysical methods, the complex petrogenesis of the basalts and the variable magma system remains poorly constrained [4]. First, the specific impact of crustal contamination during basaltic magma traversal through the continental crust cannot be definitively determined [4,5,6]. Second, the genetic relationship between the two types of basalts (alkaline and tholeiitic basalts) is not simply magmatic differentiation [5]. Previous studies have often focused on alkaline basalts associated with later trachytes [4,7], while neglecting the petrogenesis research on tholeiitic basalts. Furthermore, the properties of the mantle source region are still inadequately defined. It remains unclear whether the source comprises primitive mantle [2], a mixed composition of Depleted Mantle (DM), and Enriched Mantle I (EM I) [8], or a heterogeneous mantle associated with a mantle plume [4,9,10]. This study primarily focuses on tholeiitic basalts retrieved from the CZK06 drill core on the northern slope of the TCV. Through elemental geochemistry and magnesium isotope analyses, we investigate their compositional variability to further constrain magma evolution, petrogenesis, and mantle source properties.

2. Geological Setting

The CMVF is typified by the outcrops of Cenozoic volcanic rocks, interspersed with scattered exposures of Precambrian schists, Mesozoic volcanic rocks, and Precambrian and Mesozoic granites. These volcanic rocks are predominantly basalts, with a cumulative volume estimated at 1.2 × 10³–4.8 × 10³ km³ [4]. Based on the lithospheric thickness variation trends across NE China, the CMVF is projected to have a lithospheric thickness of 120–160 km [11]; however, influenced by mantle magma upwelling, the actual thickness is relatively reduced (80–100 km) [12]. The Moho depth within this region ranges from 30 to 35 km [13] to approximately 40 km [12].

The TCV is positioned at the intersection of the Liudaogou–Zhenfengshan fault (NE trending) and Huadian–Jince fault (NW trending) (Figure 1b). Below the volcanic edifice, several low-velocity bodies, which could be interpreted as potential magma reservoirs, have been identified [13]. Within the region, sporadic basement rocks, including early Precambrian schists and late Mesozoic volcanic rocks, are exposed in the valleys, reflecting a complex volcanic geological setting. The eruptive history of the volcano can be traced back to the Early Pliocene (~5 Ma) [2]. Its eruptive products primarily consist of basalts, which exhibit a shield-shaped distribution, and trachytes forming a conical structure overlying this shield. Additionally, Early Miocene basalts are also present in the Naitoushan, Feihushanzhuang, and Manjianghegu areas [14]. Notably, unlike other late Cenozoic volcanoes in NE China, the TCV displays a relatively complete compositional sequence, encompassing tholeiitic and alkaline basalts, trachytes, and alkaline rhyolites [2,4].

3. CZK06 Drill Core Stratigraphy and Sampling

The CZK06 drill-hole is positioned on the northern slope of the TCV at coordinates E 128°11′42″ and N 42°9′46″ (Figure 1c). It has an elevation of 1201 m, a vertical drilling angle (90°), a total depth of 403 m, and a recovered core length of 395 m. The surrounding surface residual deposits consist of gray-white brecciated pumice debris, which is essentially consistent with the deposits in the upper section (0–21.6 m) of the core. From 21.6 m depth to the total drilling depth of 403 m, basalts, as the primary geological formation, exhibit distinct features typical of continental lava cooling units. Based on the basalt K-Ar dating, phenocryst characteristics, and development features of cooling units, the core is classified into four phases, comprising a total of 42 units (Figure 1d). Petrogeochemical results indicate that the basalts of Phase I, II, and III display tholeiitic basalt characteristics and constitute the primary focus of this study.

The Phase I basalts, located in the lower section of the core (depth: 307.3–401.5 m), comprise nine lava cooling units with significant thickness variations (4.6–39.9 m). Notably, the uppermost unit (L9) lacks a top oxidation zone. Phase II basalts occur at core depths of 166.4–304.9 m, consisting of sixteen lava units intercalated with one loess-like soil layer. These units exhibit minor thickness variations (4.4–13.6 m). Phase III basalts are distributed at core depths of 84.7–166.4 m, encompassing eight lava units and two loess-like soil layers. Their unit thickness varies significantly (2.4–23.7 m). The top oxidation zone is absent from units L29 and L31, which are below the two soil layers, as well as the uppermost unit (L33). Based on the K-Ar age of the basaltic matrix (unpublished data), the formation ages of these three phases of basalts correspond to the Pliocene, the Early Pleistocene, and the Middle Early Pleistocene. They represent products of volcanic activity during the shield-forming period prior to 1 Ma.

A total of 19 basalt samples were collected from the core, including 6 from Phase I, 8 from Phase II, and 5 from Phase III. The specific sampling positions are shown in Figure 1d. All samples display either massive or sparsely vesicular textures, with fresh to weakly weathered characteristics. Subsequently, all collected samples underwent petrographic analysis, along with major and trace element analyses. Additionally, a subset of seven representative samples was selected for Mg isotope analysis.

4. Analytical Methods

Whole-rock major and trace elemental compositions were determined following petrographic analysis. Fresh samples were carefully selected, crushed, and powdered in an agate mill to a grain size of ~200 mesh for subsequent elemental analysis. Major element analysis was conducted at the National Center for Geological Experimentation and Testing, China Geological Survey in Beijing, China. Trace element analysis was performed at the Key Laboratory of Orogenic Belt and Crustal Evolution, Peking University, Beijing, China. Major element compositions were quantified using X-ray fluorescence spectrometry (XRF), with analytical uncertainties ranging to less than 2‰. Trace element compositions were determined via inductively coupled plasma mass spectrometry (ICP-MS), yielding analytical precision of better than 5% for elements > 10 ppm, better than 8% for elements < 10 ppm, and approximately 10% for transition metals. Results of the analyses for the basalt samples are listed in

Table 1. Whole-rock major element concentrations and CIPW norm calculations for the tholeiitic basalt samples.

Sample	CZK06-16	CZK06-18	CZK06-21	CZK06-24	CZK06-26	CZK06-27	CZK06-32	CZK06-35	CZK06-38	CZK06-39
Phase	III	III	III	III	III	II	II	II	II	II
Major elements (wt. %)
SiO2	51.7	50.6	49.8	51.7	51.3	50.4	58.5	53.9	50.9	52.0
Al2O3	15.7	15.5	15.3	16.6	15.6	14.8	14.5	15.3	15.2	15.5
Fe2O3	3.5	3.2	2.9	3.0	4.6	8.1	1.6	3.1	2.0	2.8
FeO	7.5	8.0	9.2	7.1	6.3	3.7	6.1	5.2	8.1	7.6
CaO	7.9	8.2	8.3	8.1	7.7	7.7	5.7	6.9	8.0	7.5
MgO	4.9	5.4	5.3	4.9	5.1	3.8	4.9	4.1	5.8	4.5
K2O	1.5	1.3	1.2	1.3	1.5	1.4	2.1	1.8	0.9	1.6
Na2O	3.5	3.5	3.5	3.9	3.7	3.5	3.5	3.6	3.5	3.7
TiO2	2.5	2.6	2.7	2.2	2.6	2.8	1.5	2.1	1.9	2.5
P2O5	0.4	0.4	0.5	0.4	0.5	0.6	0.2	0.5	0.3	0.4
MnO	0.2	0.2	0.2	0.1	0.1	0.2	0.1	0.1	0.2	0.1
LOI	0.3	−0.2	−0.4	−0.4	0.5	2.2	1.1	2.9	2.0	0.9
SUM	99.6	98.7	98.5	99	99.6	99.2	99.8	99.4	98.6	99.2
FeOT	10.6	10.9	11.8	9.8	10.5	11.0	7.5	8.0	9.9	10.1
Mg#	45.2	46.8	44.3	47.1	46.5	38.1	54.0	48.0	51.0	44.3
K2O/Na2O	0.4	0.4	0.3	0.3	0.4	0.4	0.6	0.5	0.3	0.4
Na2O + K2O	5.0	4.8	4.7	5.2	5.2	4.9	5.7	5.4	4.4	5.3
FeOT/MgO	2.2	2.0	2.2	2.0	2.1	2.9	1.5	1.9	1.7	2.2
σ	2.8	2.9	3.1	3.0	3.1	2.8	2.0	2.5	2.1	2.9
A.R.	1.54	1.51	1.50	1.53	1.57	1.56	1.78	1.65	1.46	1.59
DI	41.1	38.1	37.2	41.1	42.7	44.6	52.4	49.9	36.7	43.6
Ca/Al (mol)	0.5	0.5	0.5	0.4	0.4	0.5	0.4	0.4	0.5	0.4
FC3MS	1.1	1.0	1.1	0.9	1.1	1.2	1.1	0.9	0.9	1.1
CIPW norm
Qz	2.4	0.3	0	0.3	2.6	5.2	9.4	7.2	0.9	2.4
An	22.8	22.8	22.7	24.2	21.9	21.1	17.7	20.8	24	21.5
Ab	29.7	30.1	30	32.8	31.4	31	30.2	31.8	30.3	31.8
Or	9	7.8	7.2	8	8.9	8.4	12.8	11	5.4	9.4
Af	13.9	12.2	11.4	12.6	14.3	13.6	20.8	17.6	8.5	15.2
Pl	47.6	48.6	48.5	52.4	47.9	46.9	39.9	45.9	51.3	47.6
Nph	0	0	0	0	0	0	0	0	0	0
Di	11.3	13	13	11	9.6	11.5	8.2	9.2	12.4	11
Hy	14	15.4	14.4	14.1	12.7	9.1	16.2	10.3	19.4	14
Ol	0	0	2.2	0	0	0	0	0	0	0
Il	4.8	5.1	5.2	4.2	4.9	5.6	2.8	4	3.8	4.9
Mt	5	4.7	4.2	4.4	6.4	6.7	2.3	4.6	3	4.1
Ap	1	0.9	1.2	1	0.6	1.5	0.4	1.2	0.8	1.1
Sample	CZK06-40	CZK06-44	CZK06-46	CZK06-49	CZK06-50	CZK06-53	CZK06-55	CZK06-56	CZK06-60
Phase	II	II	II	I	I	I	I	I	I
Major elements (wt. %)
SiO2	52.5	48.9	51.1	54.5	54.7	55.8	55.7	50.5	50.4
Al2O3	15.2	14.6	14.3	14.7	14.8	14.7	14.5	14.5	14.7
Fe2O3	2.5	2.6	4.2	2.4	1.9	3.6	3.6	4.0	4.0
FeO	8.0	8.6	7.8	6.7	7.1	6.2	6.6	7.7	7.5
CaO	7.3	8.7	7.3	6.3	6.0	6.6	6.4	7.7	7.8
MgO	4.4	4.9	3.9	4.3	4.1	3.9	4.3	5.8	5.7
K2O	1.7	1.6	1.7	2.0	1.9	1.6	1.6	1.0	1.0
Na2O	3.7	3.3	3.5	3.5	3.8	3.5	3.6	3.3	3.4
TiO2	2.6	2.6	3.0	2.1	2.1	2.1	2.1	2.3	2.3
P2O5	0.5	0.6	0.7	0.5	0.5	0.5	0.5	0.9	0.9
MnO	0.1	0.2	0.2	0.2	0.1	0.1	0.1	0.2	0.2
LOI	0.6	3.0	1.6	1.7	2.0	0.7	0.3	1.3	1.3
SUM	99.2	99.4	99.2	98.9	99	99.2	99.2	99.2	99.2
FeOT	10.2	10.9	11.6	8.9	8.9	9.5	9.9	11.3	11.1
Mg#	43.5	44.5	37.3	46.1	45.3	42.6	43.7	47.7	47.6
K2O/Na2O	0.5	0.5	0.5	0.6	0.5	0.5	0.4	0.3	0.3
Na2O + K2O	5.5	4.9	5.2	5.5	5.7	5.1	5.2	4.4	4.4
FeOT/MgO	2.3	2.2	3.0	2.1	2.1	2.4	2.3	2.0	2.0
σ	3.0	3.3	3.1	2.5	2.6	1.9	2.1	2.3	2.4
A.R.	1.64	1.53	1.64	1.71	1.75	1.62	1.66	1.49	1.48
DI	44.7	38.7	45.7	49.6	50.6	50.1	49.7	38.7	38.7
Ca/Al (mol)	0.4	0.5	0.5	0.4	0.4	0.4	0.4	0.5	0.5
FC3MS	1.1	1.0	1.4	1.2	1.2	1.2	1.3	1.1	1.1
CIPW norm
Qz	2.3	0	4.8	6.7	6	10.8	9.3	3.8	3.4
An	20	20.9	18.6	18.9	18.3	20.1	18.8	21.9	22.5
Ab	32	29	30.5	30.7	32.8	29.9	31.1	28.7	29.1
Or	10.3	9.7	10.4	12.2	11.8	9.4	9.3	6.2	6.2
Af	16.9	15.2	17	19.6	19.7	15.1	15.5	9.8	9.8
Pl	45.5	44.4	42.4	42.1	43.1	44.3	43.7	47.1	48
Nph	0	0	0	0	0	0	0	0	0
Di	11.2	16.3	11.6	8.3	7.7	8.1	8.4	9	9.2
Hy	14.3	11.1	10.6	14.4	15.4	11.3	12.8	17.8	17.1
Ol	0	2.6	0	0	0	0	0	0	0
Il	5	5.1	5.8	4.1	4	4.1	4	4.5	4.5
Mt	3.7	3.9	6.3	3.6	2.9	5.4	5.3	6	6
Ap	1.1	1.4	1.7	1.2	1.2	1.1	1.1	2.2	2.2

Note: FeO^T = FeO + 0.8998 × Fe₂O₃ (wt. %), Mg^# = 100 × (MgO)/(MgO + FeO^T) (mol), Rittmann index (σ) = (Na₂O + K₂O)²/(SiO₂ − 43) (wt. %), alkalinity ratio (A.R.) = [Al₂O₃ + CaO + (Na₂O + K₂O)]/[Al₂O₃ + CaO − (Na₂O + K₂O)] (wt. %), differentiation index (DI) = Qz + Or + Ab + Ne + Lc + Kp, in which Qz, Or, Ab, Ne, Lc and Kp are calculated with CIPW, FC3MS = FeO/CaO − 3 × MgO/SiO₂. Abbreviation: Qz, quartz; An, anorthite; Ab, albite; Or, orthoclase; Af, alkali-feldspar; Pl, plagioclase; Nph, nepheline; Di, diopside; Hy, hypersthene; Ol, olivine; Il, ilmenite; Mt, magnetite; Ap, apatite.

and

Table 2. Whole-rock trace element concentrations (ppm) for the tholeiitic basalt samples.

Sample	CZK06-16	CZK06-18	CZK06-21	CZK06-24	CZK06-26	CZK06-27	CZK06-32	CZK06-35	CZK06-38	CZK06-39
Phase	III	III	III	III	III	II	II	II	II	II
Y	24.8	22.9	23.4	22.3	19	22.2	18	21.8	19.2	28.3
La	24	20.5	18.5	20.6	18.1	29.5	21.2	22.8	13.2	21.9
Ce	49.5	43.2	40.8	43.2	38.7	62.2	42.9	49.2	28.5	47.6
Pr	6.4	5.5	5.5	5.6	4.8	8	5.3	6.5	3.9	6.2
Nd	28.1	23.8	24.1	23.8	19	35.1	21.4	27.9	16.9	26.7
Sm	7.6	6.6	6.8	6.7	5.3	9.1	5.9	7.3	4.9	7.2
Eu	2.5	2.4	2.6	2.4	1.9	2.7	1.5	2.9	1.7	2.4
Gd	6.9	6.2	6.6	6.1	5	7.4	5.1	6.7	4.9	7
Tb	1	0.9	1	0.9	0.7	1.1	0.8	1	0.8	1.1
Dy	5.4	4.8	5.2	4.7	4	5.8	4.3	5.2	4.1	5.9
Ho	1	0.9	0.9	0.9	0.7	1.1	0.8	0.9	0.8	1.1
Er	2.4	2.2	2.3	2.1	1.8	2.8	2.1	2.3	1.9	2.8
Tm	0.4	0.3	0.3	0.3	0.3	0.4	0.3	0.3	0.3	0.4
Yb	2	1.8	1.9	1.8	1.8	2.7	2.3	2	1.7	2.6
Lu	0.3	0.3	0.3	0.3	0.3	0.4	0.3	0.3	0.3	0.4
Be	1.2	1.1	1	1.1	1.2	1.1	1.3	1.1	1	1.9
Cr	112	146	75	116	96	26	116	80	98	35
Rb	34	30	25	29	33	25	48	26	21	49
Sr	525	539	523	591	504	404	243	409	429	496
Zr	193	169	159	173	281	194	209	193	114	182
Nb	25	23	19	22	47	26	12	18	15	28
Ba	569	505	518	558	683	663	467	886	327	524
Hf	5.9	5.1	4.9	5.1	9.1	6.1	6.8	5.7	3.7	5.9
Ta	1.7	1.5	1.3	1.4	2.8	2	1	1.2	0.9	1.9
Th	3.3	2.6	2.3	2.4	2.4	4	5.7	2.8	1.9	4.5
U	0.6	0.5	0.4	0.5	0.7	0.7	1.2	0.5	0.4	0.9
Pb	5	2.6	3.8	2.4	5.8	4.2	5.9	3.5	2.2	4.6
Cs	0.5	0.4	0.3	0.2	1.4	0.8	1	0.4	0.3	0.9
Sn	17.7	1.8	1.5	1.5	9.2	2.4	2	2.8	1.2	2.3
ΣREE	138	119	117	120	103	168	114	135	84	133
ΣLREE	118	102	98	102	88	146	98	117	69	112
ΣHREE	19	17	19	17	15	22	16	19	15	21
L/H	6.1	5.9	5.3	6	6	6.8	6.1	6.2	4.7	5.3
LaN/YbN	8.4	8.3	6.8	8.3	7.1	7.8	6.7	8.2	5.5	6.2
LaN/SmN	2	2	1.8	2	2.2	2.1	2.3	2	1.7	2
TbN/YbN	2.3	2.4	2.3	2.3	1.9	1.8	1.6	2.3	2	1.9
δEu	1.1	1.1	1.2	1.1	1.1	1	0.8	1.2	1.1	1
δCe	0.96	0.98	0.98	0.97	1.00	0.98	0.97	0.98	0.97	0.99
Hf/Hf*	1	1	1	1	2.3	0.9	1.5	1	1	1.1
Ti/Ti*	0.9	1.1	1	0.9	1.3	0.9	0.7	0.8	1	0.9
Sample	CZK06-40	CZK06-44	CZK06-46	CZK06-49	CZK06-50	CZK06-53	CZK06-55	CZK06-56	CZK06-60
Phase	II	II	II	I	I	I	I	I	I
Y	30.4	23.7	28.1	29.2	30.7	29.5	26.9	21.4	24
La	23	21.4	23.8	25.3	25.2	21.2	22.2	23.6	23.3
Ce	49.7	47.8	52.9	55.6	56.3	47.8	49.3	52.2	51.2
Pr	6.5	6.3	7.1	7.5	7.5	6.6	6.7	7.2	7
Nd	28.5	26.7	28.3	30.8	32.1	29.7	30.3	30.9	32.4
Sm	7.5	6.9	8.5	8.8	8.8	8.5	8.7	8.9	9
Eu	2.5	2.7	3.5	2.8	2.8	2.7	2.6	3.8	3.9
Gd	7.4	6.7	8.2	8.2	8.3	8.1	7.7	7.5	7.8
Tb	1.2	1	1.2	1.2	1.3	1.2	1.2	1.1	1.1
Dy	6.3	5.1	6.3	6.4	6.5	6.6	6.4	5.5	5.6
Ho	1.2	1	1.2	1.2	1.2	1.2	1.2	1	1
Er	3	2.4	2.9	3	3	3	3	2.4	2.4
Tm	0.4	0.3	0.4	0.4	0.4	0.4	0.4	0.3	0.3
Yb	2.7	2.1	2.6	2.7	2.8	2.8	3	2.3	2.2
Lu	0.4	0.3	0.4	0.4	0.4	0.4	0.4	0.3	0.3
Be	2.2	1	1.1	1.7	1.7	1.4	1.5	0.9	0.9
Cr	44	45	43	110	119	77	71	77	99
Rb	61	28	29	42	42	35	32	15	18
Sr	498	546	446	419	427	334	287	422	485
Zr	190	197	214	269	272	239	235	150	153
Nb	31	21	20	23	23	19	19	20	19
Ba	535	683	932	683	682	550	543	859	851
Hf	6.2	5.7	6.6	7.9	8.1	7.3	7.2	4.6	4.7
Ta	2.1	1.3	1.2	1.6	1.4	1.2	1.4	1.3	1.2
Th	5.4	2.1	2.6	3.3	3.4	2.7	2.9	2.6	2.5
U	1.1	0.4	0.5	0.7	0.7	0.5	0.6	0.4	0.4
Pb	5.3	2.5	4.8	4.7	5.3	3.6	4.3	2.4	2.9
Cs	1.2	0.3	0.3	0.5	0.5	0.5	0.3	0.3	0.3
Sn	2.8	1.3	1	2.8	2.9	2.5	4.3	1.5	26.4
ΣREE	140	131	147	154	156	140	143	147	148
ΣLREE	118	112	124	131	133	117	120	127	127
ΣHREE	23	19	23	23	24	24	23	20	21
L/H	5.2	5.9	5.4	5.6	5.6	4.9	5.1	6.2	6.1
LaN/YbN	6.1	7.4	6.6	6.7	6.6	5.4	5.3	7.5	7.7
LaN/SmN	2	2	1.8	1.9	1.9	1.6	1.6	1.7	1.7
TbN/YbN	2	2.1	2.2	2.1	2.1	2	1.8	2.2	2.3
δEu	1	1.2	1.3	1	1	1	0.9	1.4	1.4
δCe	0.98	0.99	0.99	0.98	0.99	0.99	0.98	0.97	0.97
Hf/Hf*	1.1	1	1.1	1.2	1.2	1.2	1.1	0.7	0.7
Ti/Ti*	0.9	1	0.9	0.6	0.6	0.7	0.7	0.8	0.7

Note: ΣREE is the total content of all rare earth elements (REE), including lanthanides and yttrium. ΣLREE refers to the sum of the contents of REE with lighter atomic weights, which usually includes La, Ce, Pr, Nd, Sm, and Eu. ΣHREE is the total content of REE with heavier atomic weights, typically including Gd, Tb, Dy, Ho, Er, Tm, Yb, and Lu. L/H = ΣLREE/ΣHREE. Ti/Ti* = Ti_N/(Nd_N^−0.055 × Sm_N^0.333 × Gd_N^0.722), Hf/Hf* = Hf_N/(Sm_N × Nd_N)^0.5.

, respectively. Notably, the LOI values are negative in samples CZK06-18, CZK06-21, and CZK06-24, which may be attributed to the mass increase caused by the oxidation of Fe at 900 °C.

Mg isotopic analysis was performed at Createch Testing Technology Co., Ltd. (Tianjin, China). Sample dissolution, column chemistry, and instrumental analysis procedures strictly followed the methodology outlined by [15]. Initially, powder samples were dissolved using a HF-HCl-HNO₃ mixture in Savillex screw-cap beakers (Savillex Corporation, Middleton, WI, USA). Magnesium separation was subsequently achieved via cation exchange chromatography, utilizing pre-cleaned Bio-Rad AG50W-X8 resin (Bio-Rad Laboratories, Inc., Hercules, CA, USA) (200–400 mesh) in 1 N HNO₃ medium. The purified Mg solution for mass spectrometry was obtained after two sequential separation cycles. Prior to instrumental analysis, the eluted solution was evaporated to dryness and redissolved in a 3% HNO₃ solution. USGS standard materials (BHVO-2 and BCR-2), along with a procedural blank, were prepared using identical protocols to evaluate analytical precision, accuracy, and blank contributions [16]. Mg isotopic compositions were measured via a Neptune Plus MC-ICP-MS, with ²⁴Mg signal intensities averaging 4–5 V/ppm. Each sample was analyzed at least four times. Results are reported in δ notation as: δ*Mg = [(*Mg/²⁴Mg)_sample/(*Mg/²⁴Mg)_DSM3 − 1] × 1000, where * = 25 or 26, and DSM3 is a pure Mg-based standard solution [17]. Instrumental mass fractionation was corrected using the standard-sample bracketing method. Detailed analytical protocols are provided in [18]. Mg isotopic compositions of the samples and reference materials are presented in

Table 3. Mg isotopic compositions for the tholeiitic basalt samples and various standard materials.

Sample No.	Phase	δ26Mg	±2σ	δ25Mg	±2σ
CZK06-16	III	−0.267	0.046	−0.138	0.037
CZK06-26	III	−0.150	0.025	−0.078	0.029
CZK06-32	II	−0.243	0.038	−0.127	0.036
CZK06-38	II	−0.420	0.012	−0.216	0.019
CZK06-46	II	−0.344	0.010	−0.175	0.039
CZK06-55	I	−0.223	0.034	−0.116	0.047
CZK06-60	I	−0.212	0.003	−0.110	0.009
Replicate	I	−0.173	0.017	−0.091	0.050
BCR-2	USGS standard materials	−0.154	0.019	−0.080	0.030
BHVO-2		−0.230	0.019	−0.118	0.010
BS Mg	Quality control sample during testing	δ25MgDSM3‰ = −1.069 ± 0.027 (2SD, n = 10)
		δ26MgDSM3‰ = −2.074 ± 0.057 (2SD, n = 10)

δ*Mg = [(*Mg/²⁴Mg)_sample/(*Mg/²⁴Mg)_DSM3 − 1] × 1000, where * = 25 or 26, DSM3 is a solution made from pure Mg. 2SD indicates twice the standard deviation from four repeated measurements of one sample solution.

.

5. Results and Interpretation

5.1. Volcanic Petrography

Notable petrographic differences have been identified between basalts from the TCV formed prior to ~1 Ma and those erupted later [4]. Specifically, phenocrysts in the tholeiitic basalts from drill core CZK06 are predominantly plagioclase, with a relatively low abundance (typically < 10%), consistent with the petrographic characteristics of basalts older than 1 Ma. These tholeiitic basalts exhibit (dark) gray coloration, with massive, vesicular, or amygdaloidal structures, and porphyritic to few-porphyritic textures. Detailed petrographic features are presented in

Table 4. Petrographic characteristics of TCV tholeiitic basalts.

Sample	Texture and Structure	Matrix	Phenocryst	Episode
CZK06-16	porphyritic, vesicular (8%–10%)	intersertal, Pl (0.1–0.35 mm, major) + Px (microcrystal, few) + Ol (microcrystal, few) + Vit (few)	Pl (0.4–0.6 mm, 10%), Ol (0.3–1.2 mm, 3%–4%), Cpx (0.4–0.8 mm, 2%)	Phase III
CZK06-18	porphyritic, vesicular (6%–8%)	pilotaxitic-intergranular, Pl (0.3–0.7 mm, An = 43, major) + Cpx (0.03–0.45 mm, 30%–35%) + Ol (0.05–0.12 mm, 2%–3%) + Mm	Pl (0.5–6.0 mm, 10%), Ol (0.4–1.0 mm, 3%–4%)
CZK06-21	porphyritic, massive	intergranular, Pl (0.3–0.7 mm, major) + Cpx (0.3–1.0 mm, 25%–30%) + Ol (0.03–0.1 mm, 5%)	Pl (1.0–5.0 mm, 20%), Ol (0.6–1.2 mm, 2%–3%)
CZK06-24	porphyritic, vesicular (2%–3%)	intergranular, Pl (0.3–0.8 mm, major) + Cpx (0.5–1.8 mm, 20%) + Ol (0.1–0.35 mm, 7%–8%) + Mm (few)	Pl (1.0–4.6 mm, 10%–15%), Ol (<1%)
CZK06-26	porphyritic, massive	intergranular, Pl (0.4–1.2 mm, major) + Cpx (0.3–0.6 mm, 10%) + Ol (0.1–0.3 mm, 5%) + Mm (0.1–0.3 mm, 3%–4%)	Pl (1.0–4.0 mm, An = 62, 10%–12%), Ol (<1%)
CZK06-27	porphyritic, vesicular and amygdaloidal (2%–3%)	intergranular, Pl (0.08–0.15 mm, major) + Dm + Cry + Vit	Pl (0.6–3.6 mm, 7%–8%), Cpx (<1%)	Phase II
CZK06-32	few-porphyritic, vesicular and amygdaloidal (3%–4%)	intersertal, Pl (0.04–0.18 mm, major) + Px (microcrystal, 15%) + Vit	Pl (<1%), Cpx (0.4–2.0 mm, 2%–3%)
CZK06-35	porphyritic, vesicular and amygdaloidal (5%–6%)	intersertal, Pl (0.1–0.3 mm, major) + Px (microcrystal, 15%) + Vit	Pl (0.6–0.8 mm, 3%–5%), Cpx (<1%)
CZK06-38	porphyritic, vesicular and amygdaloidal (3%–4%)	intergranular, Pl (0.08–0.35 mm, major) + Px (<0.15 mm, 15%) + Ol (0.08–0.15 mm, 2%)	Pl (0.4–1.2 mm, 4%–5%), Cpx (0.5–1.2 mm, 1%–2%)
CZK06-39	porphyritic, amygdaloidal (3%–4%)	intergranular, Pl (0.1–0.45 mm, major) + Px (<0.1 mm, 4%) + Ol (<0.1 mm, 6%) + Mm (few)	Pl (0.4–4.4 mm, 8%–10%)
CZK06-40	porphyritic, massive	pilotaxitic-intergranular Pl (0.1–0.2 mm, major) + Px (<0.1 mm, 8%) + Ol (few) + Mm (few)	Pl (0.4–10.0 mm, 10%), Cpx (0.4–3.0 mm, 2%)
CZK06-44	porphyritic, vesicular and amygdaloidal (4%–5%)	pilotaxitic-intergranular, Pl (0.1–0.5 mm, major) + Ol (0.05–0.2 mm, 2%–3%) + Vit	Pl (0.6–1.4 mm, 3%–4%), Cpx (1.0–1.2 mm, 2%)
CZK06-46	few-porphyritic, vesicular (2%–3%)	intersertal, Pl (0.05–0.15 mm, major) +Dm + Mm + Vit	Pl (0.6–1.6 mm, 2%–3%)
CZK06-49	few-porphyritic, amygdaloidal (4%–5%)	intersertal, Pl (0.05–0.12 mm, major) + Px (microcrystal) + Mm + Vit	Pl (0.6–1.0 mm, 2%–3%), Cpx (0.6–1.4 mm, 2%–3%)	Phase I
CZK06-50	few-porphyritic, massive	intersertal, Pl (0.05–0.25 mm, major) + Dm + Mm + Vit	Pl (0.6–1.8 mm, 3%), Cpx (0.3–0.8 mm, 2%–3%)
CZK06-53	porphyritic, vesicular (3%–4%)	intersertal, Pl (0.1–0.3 mm, major) + Px (microcrystal) + Mm + Vit	Pl (0.6–3.8 mm, 10%–15%), Cpx (0.4–1.2 mm, 3%)
CZK06-55	porphyritic, massive	intersertal, Pl (0.08–0.35 mm, major) + Cpx (0.1–0.3 mm, 10%) + Mm + Vit (few)	Pl (0.6–3.2 mm, 8%–10%)
CZK06-56	few-porphyritic, vesicular (8%–10%)	intersertal, Pl (0.1–0.4 mm, major) + Cpx (0.1–0.3 mm, 10%) + Ol (0.1–0.3 mm, 5%) + Mm (ilmenite, 0.05–0.45 mm, 2%–3%) + Vit	Pl (0.6–1.8 mm, 5%)
CZK06-60	few-porphyritic, massive	intersertal, Pl (0.15–0.65 mm, major) + Cpx (0.05–0.2 mm, 10%) + Ol (0.1–0.2 mm, 3%–4%) + Mm (ilmenite, 0.1–0.35 mm, 2%–3%) + Vit	Pl (0.8–3.2 mm, 5%)

Whole-rock major and trace element analyses were performed on all above samples. The 7 gray-shaded samples were selected for Mg isotope analysis. Abbreviation: Ol, olivine; Px, Pyroxene; Cpx, clinopyroxene; Pl, plagioclase; Dm, dark minerals (mainly microcrystalline aggregates of Ol and/or Px); Mm, metallic minerals (mainly ilmenite); Vit, vitreous materials; Cry, cryptocrystalline.

and Figure 2.

The matrix of the TCV tholeiitic basalt samples is characterized by intergranular or pilotaxitic-intergranular structures. Their mineral compositions are relatively similar: Phase I consists of plagioclase (major) + clinopyroxene ± olivine + opaque mineral (ilmenite) (Figure 2a), Phase II consists of plagioclase (major) ± pyroxene ± olivine ± opaque mineral (ilmenite) (Figure 2b), and Phase III consists of plagioclase (major) + clinopyroxene + olivine ± opaque mineral (ilmenite) (Figure 2c). The plagioclase microlites display semi-idiomorphic to idiomorphic crystal forms, with prismatic and tabular-prismatic habits. They are characterized by broad polysynthetic twin lamellae, distinct zoning patterns, and pristine, unaltered surfaces. In the matrix of Phase III basalt, some plagioclase grains are embedded within pyroxene crystals. Additionally, phenocrysts in the tholeiitic basalts from different stages are characterized by plagioclase, with minor amounts of clinopyroxene and olivine. These phenocrysts exhibit similar morphological features in different stages, but their abundances show an increasing trend from Phase I to III, likely resulting from the intensification of magmatic differentiation. The phenocryst content in Phase I basalts is typically ≤ 5%; that in Phase II basalts ranges from 5% to 12%; and that in Phase III basalts increases to 10%–20%. For Phase I and II basalts, the phenocryst constituents are plagioclase ± clinopyroxene, whereas those in Phase III consist of plagioclase + olivine ± clinopyroxene. Plagioclase phenocrysts are semi-idiomorphic to idiomorphic, exhibiting embayed or rounded etched edges with some surfaces displaying pockmarks. Clinopyroxene occurs as short, wedge-shaped prismatic, or irregular quadrilateral granular crystals, characterized by one set of perfect prismatic cleavage, inclined extinction, and developed fractures. Olivine develops cracks and shows varying degrees of iddingsitization.

5.2. Whole-Rock Major Elements

The basalt samples display major element contents ranging from 98.5 wt. % to 99.8 wt. %, characterized by a low loss on ignition (LOI) of <3.0 wt. %. These observations suggest their relatively unaltered state, with minimal influence from subsequent modifications. Their SiO₂ concentrations show moderate variations (48.9–58.5 wt. %). Na₂O + K₂O contents are relatively low (4.4–5.7 wt. %). In the total alkali-silica (TAS) diagram (Figure 3a), all basalts belong to the subalkaline series. Phase I basalts plot within the basalt and basaltic andesite fields; Phase II samples occupy fields of basalt, basaltic trachyandesite, basaltic andesite, and andesite; Phase III samples occur in the intersection of basalt, basaltic trachyandesite, and basaltic andesite fields. CIPW norm calculations reveal that the samples contain 2%–11% quartz, with no nepheline detected (Table 1). Except for Phase II sample CZK06-32, which is affected by crustal contamination and has a lower FeO^T/MgO ratio, the other samples are classified in the tholeiitic field (Figure 3b). K₂O/Na₂O ratios are uniformly below 0.6, confirming their sodium-series affinity (Figure 3c). Additionally, their CaO (5.7–8.7 wt. %), MgO (3.8–5.8 wt. %), FeO^T (7.5–11.8 wt. %), and TiO₂ (1.5–3.0 wt. %) all exhibit moderate compositional variability. Mg^# values (37.3–54.0) and differentiation index (DI, 36.7–52.4) present relatively low values but display notable variations, indicating a certain degree of differentiation. Al₂O₃ contents are elevated (14.3–16.6 wt. %) and show an increasing trend from Phase I to Phase III.

5.3. Whole-Rock Trace Elements

The rare earth element (REE) and trace element partitioning curves (Figure 4) exhibit marked similarities to those of other Changbai Mountain basalt (CMB) samples. They display comparable element partitioning behavior to ocean island basalts (OIBs) in terms of REE, large ion lithophile elements (LILEs), and high field strength elements (HFSEs). Although, their incompatible element abundances are marginally lower than those of typical OIB, characterized by weak depletion of Th, U, Nb, Ta, and Sr, alongside enrichment of Ba, K, and Pb. The geochemical signatures may be attributed to crustal contamination, source region characteristics, and/or the degree of partial melting. Notably, certain samples display anomalous characteristics in their trace element abundances. First, Phase II sample CZK06-32 displays a significant continental crustal characteristic, characterized by pronounced negative anomalies in Nb, Ta, Sr, P, and Eu, alongside weak positive anomalies in Zr and Hf [21]. Additionally, Phase I samples CZK06-56 and CZK06-60 exhibit positive P and Eu anomalies, along with negative Zr and Hf anomalies. Phase III sample CZK06-26 exhibits pronounced positive anomalies in Nb, Ta, Zr, and Hf. These features may be attributed to source heterogeneity.

The total rare earth element (ΣREE) contents of the samples range from 84 ppm to 168 ppm, with ΣLREE/ΣHREE ratios varying between 4.7 and 6.8, La_N/Yb_N ratios ranging from 5.3 to 8.4, La_N/Sm_N ratios spanning 1.6–2.3, and Tb_N/Yb_N ratios fluctuating within 1.6–2.4. In terms of δEu, apart from Phase I sample CZK06-55 and Phase II sample CZK06-32, which display weak negative anomalies, all other samples exhibit either weak positive anomalies or no anomalies. Their δCe values, calculated as δCe = 2 × Ce_N/(La_N + Pr_N), range from 0.96 to 1.00, indicating that the Ce anomalies are absent or extremely weak.

5.4. Whole-Rock Mg Isotopes

In recent years, numerous researchers have reported the Mg isotope compositions of the CMB [10,25]. Their δ²⁶Mg values are −0.29 ± 0.27‰, primarily spanning a range of −0.56‰ to 0.23‰, which are lower than the average mantle value (N-MORB, −0.25 ± 0.04‰). Notably, some tholeiitic basalts display relatively higher values (−0.07‰ to −0.02‰). These results indicate a complex source composition [10]. In this study, TCV tholeiitic basalt samples from the CZK06 drill core plot along the terrestrial equilibrium mass fractionation line in the δ²⁵Mg vs. δ²⁶Mg diagram (Figure 5a), with Δ²⁵Mg’ values of 0.0012 ± 0.0005‰ (Δ²⁵Mg’ = δ²⁵Mg’ − 0.521 × δ²⁶Mg’, where δ^xMg’ = 1000 × [(δ^xMg + 1000)/1000] and x denotes mass 25 or 26 [26]). The δ²⁶Mg values of these samples range from −0.150 ± 0.025‰ to −0.420 ± 0.012‰. Compared to N-MORB and tholeiitic series OIB (Figure 5b), these data display significantly broader ranges, suggesting that the source compositions of the tholeiitic basalts are relatively complex.

6. Discussion

6.1. Fractional Crystallization and Crustal Contamination

6.1.1. Fractional Crystallization

Compared to high-magnesium basalts (Mg^# = 75–78) derived from a peridotite mantle source, TCV tholeiitic basalts display lower Mg^# values (37.3–54.0). These basalts exhibit OIB-like REE and trace element partition patterns (Figure 4), though their elemental enrichment degree is slightly lower, with exceptions for Ba, K, and Pb. Notably, their Rb/Sr ratios (0.04–0.20, mean = 0.08), Ba/Rb ratios (8.77–57.27, mean = 22.41), and Th/U ratios (3.43–6.50, mean = 5.18) are marginally elevated relative to typical OIB values (Rb/Sr = 0.03–0.05; Ba/Rb = 11.3; Th/U = 3.3–3.9), suggesting that these basalts have undergone a certain degree of differentiation. Geochemical studies on magma formation and evolution reveal that the enrichment of incompatible elements (A) can reflect source region characteristics, the extent of partial melting, and magma differentiation processes. However, as noted by [29], fractional crystallization cannot alter the ratio of incompatible elements (A/B); thus, the A/B-A diagram can be used to distinguish magma crystallization differentiation. In the diagrams (Figure 6), although the concentrations and ratios of incompatible elements in these basalts show minimal variation, their distribution trends remain distinct. Specifically, the distribution of data points exhibits distinctly increasing trends for ratios such as Nb/Zr, La/Sm, and (La/Yb)_N. Meanwhile, as indicated by the horizontal gray arrows, some ratios also display stable characteristics. These observations suggest that the compositions of these tholeiitic basalts are primarily controlled by source region properties and partial melting degree, with additional minor influence from fractional crystallization.

For CMB, crystallization differentiation of olivine, clinopyroxene, and plagioclase occurs prominently during the basaltic magma evolution when their MgO content is below 5 wt. % [10]. The MgO contents of the TCV tholeiitic basalts show a narrow distribution range (3.8–5.8 wt. %). Even within the restricted interval, the correlation patterns between major or trace element contents and MgO can still be clearly identified in Harker diagrams (Figure 7). With decreasing MgO content, Mg^#, CaO/Al₂O₃, FeO^T, Cr, and La exhibit linear correlation trends. Additionally, from Phase I to III, the average La_N/Yb_N and La_N/Sm_N ratios exhibit a slight increasing tendency. These characteristics indicate that these tholeiitic basalts have undergone magmatic differentiation, with this process showing an intensifying trend from Phase I to III. Further, CaO, Al₂O₃, TiO₂, and Sr exhibit a trend of initial increase followed by decrease, with their inflection points corresponding to an MgO content of ~5 wt. %. With respect to Na₂O, K₂O, and P₂O₅, their components generally show a linear increasing trend; however, when MgO content is less than 5 wt. %, their concentrations also be relatively stable in some samples. Combined with petrographic features, these trends can be attributed to the fractional crystallization of Ti-rich clinopyroxene, P-rich apatite, and plagioclase, particularly during the evolutionary stage when MgO content < 5 wt. %.

6.1.2. Crustal Contamination

Crustal contamination is inevitable when basaltic magma in the CMVF ascends through the thick continental crust. Although no crustal xenoliths have been identified within CMB, extensive prior geochemical evidence has confirmed the occurrence of crustal contamination during the evolution of basaltic magma. These lines of evidence primarily include: comparative analysis of Sr-Nd-Pb isotopic characteristics between the CMB and regional Mesozoic igneous rocks; assimilation fractional crystallization process modeling; and observations of Nb depletion features and isotopic system deviation towards the EM II end-member in individual samples [5,6]. TCV tholeiitic basalts from the CZK06 drill core exhibit low (Nb/Th)_N, (Ce/Pb)_N, and (Ti/Sm)_N ratios (

Table 5. Primitive mantle normalized ratios of characteristic trace elements.

Episode	(Nb/Th)N	(Ta/U)N	(Ce/Pb)N	(P/Nd)N	(Ti/Sm)N
Phase III	0.89–1.09 CZK06-26 is 2.36	1.40–1.59 CZK06-26 is 1.93	0.40–0.72 CZK06-26 is 0.27	0.93–1.26 CZK06-26 is 1.57	0.66–0.82 CZK06-26 is 0.99
Phase II	0.68–1.18 CZK06-32 is 0.26	0.95–1.56 CZK06-32 is 0.45	0.37–0.77 CZK06-32 is 0.29	0.98–1.47 CZK06-32 is 0.46	0.58–0.80 CZK06-32 is 0.51
Phase I	0.80–0.93	1.06–1.59	0.42–0.88	0.93–1.77	0.48–0.54
Total crust	0.17	0.28	0.16	0.40	0.38
N-MORB	2.31	1.44	1.00	1.00	0.99
E-MORB	1.65	1.34	1.00	0.98	0.79
OIB	1.43	1.36	1.00	1.00	0.59

Total crust values are sourced from [33]. N-MORB, E-MORB, and OIBs are sourced from [22].

), along with significant positive Pb anomalies. These characteristics are consistent with crustal contamination. The plots in the diagrams based on characteristic trace element ratios further confirm the crustal contamination and suggest that this process primarily occurred in the lower crust (Figure 8). Notably, sample CZK06-32 exhibits pronounced Nb and Ta depletion, accompanied by positive K and Pb anomalies (Figure 4), as well as geochemical features comparable to the East China Continental Crust [31], such as lower (Nb/Th)_N, (Ta/U)_N, (Ce/Pb)_N, and (P/Nd)_N ratios (Table 5), elevated K₂O content, and reduced concentrations of FeO^T, CaO, TiO₂, and P₂O₅ (Figure 7). This signature has also been documented to varying degrees in individual shield-forming basalts from previous studies [2,32]. Combined with the high ⁸⁷Sr/⁸⁶Sr isotopic characteristic identified in some TVC basalts by prior research [5,32], these findings suggest that the influence of crustal contamination on some samples should not be ignored.

This set of geochemical features indicates that TCV tholeiitic basalts experienced crustal contamination during magma evolution; however, these features, together with Sr-Nd-Pb isotopic characteristics, can alternatively be interpreted as reflecting the involvement of sediments in the mantle source region [10]. Given this alternative interpretation, some scholars posit that the magmatic evolution of CMB occurs within a predominantly closed system [4,10]. Except for sample CZK06-32, these samples all exhibit (Ta/U)_N, (P/Nd)_N, and (Ti/Sm)_N ratios comparable to those of OIB (Table 5). In addition, their Hf/Nd ratios (0.15–0.48, mean = 0.23), Zr/Sm ratios (16.9–53.0, mean = 26.6), La/Sm ratios (2.49–3.42, mean = 2.92), Nb/Ta (13.0–16.8, mean = 15.31), Nb/U (28.2–67.1, mean = 41.2), and Nb/La (0.79–2.60, mean = 1.07) show limited variation ranges, and are all slightly lower than those observed in OIB (Hf/Nd = 0.20, Zr/Sm = 28.0, La/Sm = 3.7, Nb/Ta = 17.8, Nb/U = 47.1, Nb/La = 1.3). Combined with characteristics typical of continental crust (e.g., depletion of Th, U, Nb, and Ta, and enrichment in Pb [21]), all these features indicate that the influence of crustal contamination on the geochemical composition of these tholeiitic basalts is minor, while also suggesting the involvement of crustal materials in their mantle source region.

6.2. Mantle Source Property and Origin of δ²⁶Mg Values

6.2.1. Mantle Source Characteristics

Based on isotope tracing techniques, previous studies have conducted extensive research on the characteristics of the mantle source region of CMB. A series of models have been proposed successively, including the DM-EM1 mixed mantle source model [8] and the PSSJ (Pacific Subducting Sediments along the Japan trench) supplementary model [4]. However, the specific composition of the mantle source region remains poorly understood. Through Sr-Nd-Pb-Hf-Mg isotope cycle studies, it is suggested that the CMB mantle source includes at least two components: Paleoproterozoic continental margin silicate sediments with EM1-like Sr-Nd isotopic characteristics, and carbonated eclogite (exhibiting MORB-like isotopic compositions and low δ²⁶Mg values) [10]. Notably, in the Pb isotope source discrimination diagram for basalts [34,35], previously reported CMB samples (data from [5,7,32]) plot in the transitional field between the end-members of recent sediments (PSSJ), ancient sediments, and depleted mantle (DM). The TCV basalts also exhibit similar projection area [5]. These indicate the geochemical compositional complexity and multi-source nature of their mantle sources.

Compared to continental arc basalts (CAB) in NE Asia [23], TCV tholeiitic basalts exhibit significant enrichment in trace elements except for HREEs, most notably lacking depletion in Nb, Ta, and P (Figure 4). This signature implies a marked divergence from common intermediate-basic volcanic rocks typically associated with subduction zones. However, aside from positive anomalies in Ba, K, and Pb, the REE and trace element distribution patterns of these samples show greater resemblance to those of typical OIBs, thus suggesting OIB-like geochemical affinities in TCV tholeiitic basalts. Notably, studies on Pb, Sr, and Nd isotopes of Cenozoic intraplate basalts in eastern China indicate that these basalts have an asthenospheric mantle origin, with those from the North China Craton generally exhibiting EM1-type mantle isotopic signatures [36]. Although TCV tholeiitic basalts exhibit more enrichment in REE and trace elements, the EM1 signature, as indicated by their Sr-Nd isotopes, is still pronounced [4,6]. This indicates that TCV tholeiitic basalts possess EM1-type OIB-like geochemical affinities. Regarding source characteristics, their elevated FC3MS values exhibit a broad distribution range (0.89–1.35), placing the samples within the pyroxenite melting field of OIBs alongside CMB (Figure 9a). Compared to spinel, garnet has a stronger affinity for heavy rare earth elements (HREE); thus, the MREE/HREE ratio serves as an effective parameter for evaluating the mineral composition of basaltic magma sources [10]. In the Gd/Yb vs. Yb diagram (Figure 9b), all sampling points fall within the eclogite region defined by varying proportions of pyroxene and garnet. Collectively, it can be inferred that TCV tholeiitic basalts, characterized by geochemical affinity with EM1-type OIB, are primarily derived from partial melting of pyroxenite or eclogite.

Pb isotope studies indicate that recent sediments (PSSJ) may be an important component of the CMB mantle source region [4]. TCV tholeiitic basalt samples exhibit extremely weak δCe anomalies (0.96–1.00), and remarkably poor correlation between δCe and LOI. This characteristic suggests the potential involvement of oceanic sediments in their mantle source region. On the other hand, oceanic sediments are known to have entered the mantle through deep subduction, and their recycled material likely contributes to basaltic volcanism. Such basalts are typically characterized by crust-like features, including depletion in Nb, Ta, Ti, and Eu, and enrichment in Pb. The current tholeiitic basalt samples exhibit weak depletion in Nb and Ta, and enrichment in Pb, suggesting that sediments carried by the deeply subducted Pacific plate have contributed to the basalt source region. Moreover, in the Ce/Pb vs. δCe diagram (Figure 9c), the Ce/Pb values of the samples vary significantly, falling between oceanic basalt and sediment from the Japan Trench (J). In conclusion, the source region of the tholeiitic basalts has been influenced by recycled recent oceanic sediment carried by the deeply subducted Pacific plate.

6.2.2. Origin of δ²⁶Mg Values

The asthenospheric mantle has a chondrite-like Mg isotopic composition (δ²⁶Mg = −0.25 ± 0.04‰, 2SD [28]), which remains stable during partial melting and high-temperature magmatic differentiation [38]. Silicate weathering induces significant Mg isotope fractionation, resulting in highly variable, light Mg isotopes in terrigenous clastic sediments [25]. Conversely, altered oceanic basalts show heavier Mg isotopes than N-MORB [28,39,40]. Altered oceanic crust (AOC) δ²⁶Mg values vary widely (−1.7‰ to +0.21‰) due to compositional differences, with clay-rich AOC having higher values and carbonate-rich AOC lower values [10], making Mg isotopes a selectable tracer for mantle recycling of oceanic crust materials [25,28]. TCV tholeiitic basalts primarily cluster within the N-MORB δ²⁶Mg range, with sample CZK06-26 (−0.150 ± 0.025‰) exceeding this range, whereas CZK06-46 (−0.420 ± 0.012‰) and CZK06-38 (−0.344 ± 0.010‰) fall below it. Their δ²⁶Mg distribution and negative correlation with TiO₂ (Figure 10a) initially suggest that this Mg isotopic signature may be linked to the accumulation of isotopically light ilmenite within the mantle source of high-Ti basalts [41]. But the lack of correlation was observed between TiO₂ concentrations and Nb/Ta ratios. Thus, it can be inferred that the observed Mg isotopic signature is attributable to mantle source characteristics and its partial melting.

The typical OIB studies (e.g., Hawaii, Louisville) show δ²⁶Mg heterogeneity is controlled by source composition and partial melting degree, with δ²⁶Mg negatively correlating with melting-sensitive element ratios [28]. Furthermore, OIB melts sourced from pyroxenite-dominated versus peridotite-dominated mantles exhibit distinct negative correlation trends [28]. TCV tholeiitic basalt samples within the N-MORB δ²⁶Mg range display a negative correlation with melting-sensitive element ratios (e.g., La/Sm, Nb/Sm, and Nb/Zr) (Figure 10b,c). Notably, these samples plot linearly along the trend defined by Hawaii OIB, suggesting their Mg isotope characteristics are consistent with a pyroxenite mantle source. The variations in δ²⁶Mg values are likely attributed to its partial melting degree within the source region. It is widely recognized that (altered) oceanic crust transforms into garnet pyroxenite after subduction into the mantle, forming a major component of typical OIB source regions (e.g., Hawaii OIB) [42,43]. The above-mentioned samples exhibit prominent HREE depletion, and their (Nb/La)_N values (0.56–1.00) are comparable to the normal subducted oceanic crust (<0.85). This characteristic further confirms that the pyroxenite mantle rocks feeding TCV tholeiitic basalt magma are primarily derived from subducted AOC. Notably, sample CZK06-26 displays a higher δ²⁶Mg value and positive anomalies in Nb, Ta, Zr, and Hf, accompanied by elevated Ti/Ti* and Hf/Hf* ratios, suggesting the clay-rich nature in the subducted AOC (Figure 10). The previous Pb isotope modeling indicates that the mantle source region of CMB contains recycled subducted ancient (2.2–1.6 Ga) terrigenous silicate sediments, characterized by relatively high δ²⁶Mg values [10]. By analogy, we speculate that the pyroxenite within the mantle source region may have originated from an ancient clay-rich AOC.

Late Cenozoic basalts in East China display low δ²⁶Mg values, the characteristic attributed to recycled sedimentary carbonates transported by the subducted Paleo-Pacific Plate [15,25,44,45]. Similarly, the TCV samples CZK06-46 and CZK06-38 display low δ²⁶Mg values, which may also be derived from recycled carbonate-rich subducted materials. Generally, carbonate melts are characterized by elevated Ca and REE contents, along with negative Ti and Hf anomalies. The two samples with low δ²⁶Mg values display analogous characteristics (Figure 10d–f), also confirming that the light Mg isotope composition is associated with melts derived from deeply subducted carbonate sediments [10,28,44]. The studies on the petrogenesis of Cenozoic nephelinites in eastern North China have revealed that the low δ²⁶Mg values, similar to those of marine carbonates, originate from the partial melting of carbonate-bearing sediments within the MTZ (410–660 km) [46]. Research on olivine phenocrysts in the nephelinites indicates that the MTZ exhibits an extremely high degree of carbon enrichment [47]. In addition, according to the slab melting model proposed by [48], a barrier forms at depths of 300–700 km, preventing carbonate sediments from entering the deep lower mantle. Given the stagnant slabs imaged in the MTZ via seismic tomography [3], it can be inferred that subducted carbonate sediments were entrained within these slabs. Partial melting of the carbonate sediments generates initial carbonate-silicate melts, which can provide heterogeneous low ²⁶Mg components through metasomatism of mantle rocks such as pyroxenite or eclogite.

6.2.3. Implications for Carbonatization in the Mantle Source Region

Studies on carbonate melts reveal that they are characterized by high concentrations of Ca, REE, and Ca/Al ratios, alongside notably low Ti/Ti* and Hf/Hf* ratios. Consequently, scholars commonly utilize lower Ti/Ti* and Hf/Hf* ratios, coupled with higher Ca/Al (molar) and Zr/Hf ratios, to identify carbonatized metasomatism in mantle source regions [9]. For TCV tholeiitic basalt samples, their relatively low Ca/Al (molar) ratios (0.36–0.54) and Zr/Hf ratios (31–35) seemingly deviate from the typical characteristics of carbonatization. But these low ratios potentially arise from crustal contamination. Notably, their Ti/Ti* ratios range from 0.69 to 2.28, with an average (excluding outliers) of 0.97, while Hf/Hf* ratios span 0.64 to 1.31, averaging 0.84. Compared to EM (Ti/Ti* = 1.05, Hf/Hf* = 0.89) or OIB (Ti/Ti* = 1.00, Hf/Hf* = 0.90), the tholeiitic basalts exhibit a broader range of these ratios and slightly lower average values, suggesting that the mantle source region has undergone a certain degree of carbonatized metasomatism. Additionally, researchers have utilized the trends of K₂O + Na₂O vs. TiO₂ data for basalts to constrain source characteristics and determine whether carbonatized metasomatism has occurred [25,44,49]. In the diagram (Figure 11a), the tholeiitic basalts overlap with the field of Eastern China basalts and lie adjacent to the carbonated peridotite melt trend or between that trend and carbonated pyroxenite melt trend. This observation indicates that carbonatized mantle rocks constitute an important component of the source region.

Carbonate-bearing sediments exhibit the lowest melting temperature at pressures exceeding 5 GPa [35]. At low pressures (<8 GPa), the partition coefficients of large ion lithophile elements (LILEs), such as Rb, Sr, Ba, Th, U, and La, are significantly lower than those observed at high pressures (>8 GPa) [50]. Calculations further demonstrate that low-pressure melts are characterized by higher K/U and Ba/Th ratios, whereas high-pressure melts display lower values [46]. The K/U × 10⁻³ vs. Ba/Th diagram (Figure 11b) illustrates the compositional fields of melts generated under varying pressure conditions with distinct residual mineral assemblages [35]. In the diagram, TCV tholeiitic basalts plot adjacent to the field of Hainandao basalts. Additionally, both the basalts exhibit low δ²⁶Mg values and share similar REE and trace element distribution patterns, indicating their common genetic origin. These lines of evidence collectively indicate that the carbonate melts, supplying for the carbonatization in the mantle source region, formed under low-pressure conditions.

6.3. Constraint on the Mantle Magma System

The significant velocity anomalies across cross-sections in the CMVF indicate that thermally buoyant material sourced from depth ascends to approximately 180 km before flowing northward and westward [51]. This geophysical signature supports the existence of a mantle magmatic system beneath the TCV. Mantle melting is hypothesized to initiate at the depth where the solidus intersects the liquidus and terminates at the lithosphere-asthenosphere boundary. The depth of mantle melting and lithospheric thickness can be quantitatively constrained through fitting Na/Ti vs. Sm/Yb ratios in basalts [52,53]. Applying this method, the formation depth of TCV tholeiitic magma is estimated at ~180 km (Figure 12), which closely matches depths determined by geophysical investigations [51].

Niu (2010) proposed that a low-velocity zone (LVZ) at depths of 160–180 km beneath the North China Craton lithosphere hosts significant concentrations of volatile-rich melts, primarily composed of H₂O and CO₂ [21]. Deep-seated “fertile” and readily fusible mantle material is metasomatized by these volatile-rich melts, forming basaltic magma with OIB-like characteristics. Geochemical analysis indicates that TCV tholeiitic basalts originate from an EM1-type OIB-like source. Their characteristic elemental ratios, such as Nb/Th and Ce/Pb, predominantly fall between those of LCC and E-MORB or OIB (Figure 8). From this, it is inferred that tholeiitic basaltic magma formed beneath the lithospheric mantle at depths of 160–180 km, corresponds to the Pliocene-Early Pleistocene LVZ.

7. Conclusions

This study presents elemental geochemical and magnesium isotopic analyses of tholeiitic basalts from the CZK06 drill core on the northern slope of the TCV, providing insights into their magma evolution, petrogenesis, and mantle source properties.

(1)

The tholeiitic basalts from the CZK06 drill core formed during the Pliocene-Early Pleistocene shield-forming stage, recording three phases of basaltic volcanism (Phases I to III). Geochemically classified as sodium-series volcanic rocks, these basalts exhibit clear affinities with EM1-type OIBs. Notably, their δ²⁶Mg values (−0.420‰ to −0.150‰) span a substantially wider range relative to the N-MORB.

(2)

The compositions of these tholeiitic basalts are primarily controlled by source region characteristics and partial melting degree, with minor additional influences from fractional crystallization and crustal contamination. During magmatic differentiation, fractional crystallization of Ti-rich clinopyroxene, P-rich apatite, and plagioclase is notably prominent, showing a progressively intensifying trend from Phase I to III.

(3)

TCV tholeiitic basalts are primarily derived from the partial melting of pyroxenite with carbonatization. Integrated with Mg isotopic data, our results suggest that the pyroxenite originated from subducted ancient clay-rich AOC. The carbonate melts fueling the carbonatization were generated by low-pressure melting of recent oceanic sediments, which were transported by the deeply subducted carbonate-rich Pacific Plate within the MTZ. The tholeiitic magma formed in the LVZ at depths of 160–180 km beneath the lithospheric mantle.