Post-Collisional Mantle Processes and Magma Evolution of the El Bola Mafic–Ultramafic Intrusion, Arabian-Nubian Shield, Egypt

Abstract

The El Bola mafic–ultramafic intrusion (EBMU) in Egypt’s Northern Eastern Desert represents an example of Neoproterozoic post-collisional layered mafic–ultramafic magmatism in the Arabian–Nubian Shield (ANS). The intrusion is composed of pyroxenite, olivine gabbro, pyroxene gabbro, pyroxene–hornblende gabbro, and hornblende-gabbro, exhibiting adcumulate to heter-adcumulate textures. Mineralogical and geochemical analyses reveal a coherent trend of fractional crystallization. Compositions of whole rock and minerals indicate a parental magma of ferropicritic affinity, derived from partial melting of a hydrous, metasomatized spinel-bearing mantle source, likely modified by subduction-related fluids. Geothermobarometric calculations yield crystallization temperatures from ~1120 °C to ~800 °C and pressures from ~5.2 to ~3.1 kbar, while oxygen fugacity estimates suggest progressive oxidation (log fO₂ from −17.3 to −15.7) during differentiation. The EBMU displays Light Rare Earth element (LREE) enrichment, trace element patterns marked by Large Ion Lithophile Element (LILE) enrichment, Nb-Ta depletion and high LILE/HFSE (High Field Strength Elements) ratios, suggesting a mantle-derived source that remained largely unaffected by crustal contribution and was metasomatized by slab-derived fluids. Tectonic discrimination modeling suggests that EBMU magmatism was triggered by asthenospheric upwelling and slab break-off. Considering these findings alongside regional geologic features, we propose that the mafic–ultramafic intrusion from the ANS originated in a tectonic transition between subduction and collision (slab break-off) following the assembly of Gondwana.

1. Introduction

Neoproterozoic ultramafic–mafic intrusive complexes, dated between 610 and 550 Ma, are extensively distributed across the Arabian–Nubian Shield (ANS) (Figure 1a,b). Within the ANS, these rocks originated in various tectonic environments, including oceanic and continental environments. These complexes are associated with either supra-subduction zone ophiolitic settings or with within-plate and late- to post-collisional magmatism [1]. In the Egyptian Eastern Desert and Sinai, the Precambrian ultramafic–mafic complexes are classified into ophiolite (e.g., Abu Dahr [2] and Wadi Arais [3]), layered mafic–ultramafic intrusions (e.g., Motaghairat [4]), and concentrically zoned Alaskan-type mafic–ultramafic intrusions (e.g., Genina Gharbia [5]). The ophiolite rocks (mainly 770 to ~720 Ma [6,7,8]), that represent remnants of ancient oceanic crust, display geochemical signatures, typical of a supra-subduction zone (SSZ), later emplaced along convergent plate boundaries that occur along major shear zones, and locally attain the lower amphibolite facies [9,10,11,12]. The Alaskan-type complexes (963 ± 81 Ma for Genina Gharbia [5]) are characterized by concentric zoning, with ultramafic rocks forming the core and mafic rocks occurring along the outer margins, with both exhibiting typical arc affinity (e.g., Gabbro Akarem [13]; Genina Gharbia [5]; Abu Hamamid [14]). The layered ultramafic–mafic intrusions (e.g., Motaghairat ≈700 Ma [4]; Korab Kansi ≈ 740 Ma [15]; Rahaba ≈ <600 Ma [16]; Wadi El Dib ≈ 591.5 ± 3.5 Ma [17]), comprise unmetamorphosed, differentiated rock sequences. These sequences typically begin with ultramafic cumulates at the base, gradually transitioning upward into layered troctolite, hornblende-bearing rocks and gabbros [4].

The ANS mafic–ultramafic intrusions possess significant economic potential, with the nature and typology of these intrusions playing a crucial role in determining the associated mineralization types. Alaskan-type mafic–ultramafic intrusions, such as Gabbro Akarem [18] and Genina Gharbia [5], are known to possess sub-economic concentrations of Cu–Ni–S–PGE mineralization. In contrast, many layered intrusions, including Korab Kansi, Rahaba, and Gabal Akab El Negum, possess economically viable deposits of ilmenite-titanomagnetite, which are of significant commercial interest [15,16,19].

The ultramafic rocks of the ANS layered mafic–ultramafic intrusions include dunite, peridotite, and pyroxenite that are generally well preserved, with minor serpentinization commonly restricted to zones affected by shearing [4,15,16,17]. The mafic component constitutes a sequence of rocks arranged upward with decreasing basicity ranging from olivine gabbro to anorthosite. A distinctive feature of these intrusions is the prevalence of cumulate textures, which are seldom observed in the ANS ophiolitic sequences [12,20].

Layered mafic–ultramafic intrusions typically occur as small, steeply inclined, dyke-like or sheet-like bodies and are usually unmetamorphosed [21]. Their origin is generally attributed to Neoproterozoic mantle plumes or superplume activity during the breakup of Rodinia [22,23,24], whereas two arc-related and plume-driven models were suggested for their emplacement [23,25]. Nevertheless, it is widely acknowledged that the involvement of volatile–rich mantle plumes or the ascent of asthenospheric melts could play a conclusive role in the generation and emplacement of these mafic–ultramafic intrusions [9,19,24,26,27].

Figure 1. (a) Spatial distribution of mafic and ultramafic rocks in the Egyptian Eastern Desert [28]. (b) Inset showing the Arabian–Nubian Shield; (c) detailed geological sketch of the El Bola layered mafic–ultramafic intrusion.

Figure 1. (a) Spatial distribution of mafic and ultramafic rocks in the Egyptian Eastern Desert [28]. (b) Inset showing the Arabian–Nubian Shield; (c) detailed geological sketch of the El Bola layered mafic–ultramafic intrusion.

The El Bola mafic–ultramafic intrusion (EBMUI) is situated in the southern part of the Northeastern Desert (NED) of Egypt. Although the El Bola region has been geologically studied, research has predominantly focused on its granitic assemblages and their petrogenesis [29,30]. In contrast, the origin and development of the EBMUI remain debated. Some authors have interpreted it as part of an island-arc metagabbro–diorite suite emplaced in an arc setting [30,31], whereas others have linked it to older ophiolitic-affinity [32]. Recently, Khedr, et al. [33] investigated the Fe-Ti oxide ore endowment located in the El-Baroud mafic complex that is considered to be the northwestern extension of the El Bola area, where they proposed that the Fe–Ti oxide-rich layers appear to have formed through a combination of in situ crystallization and liquid-immiscibility processes. This work represents the first comprehensive investigation of the El Bola mafic–ultramafic intrusion, focusing on its mineralogical and petrochemical characteristics, in order to constrain the nature of the parental magma, petrogenetic evolution, and tectonic setting of this intrusion. Ultimately, this study contributes to a broader understanding of Neoproterozoic magmatic activity and crustal evolution in this portion of Gondwanaland, offering insights into the magmatic behavior and evolution of the Nubian Shield.

2. Geologic Setting

The El Bola mafic–ultramafic intrusion (EBMU) appears as a relatively small, sill-like body trending NE–SW, and it dips gently (25–35°) toward the northeast, extending approximately 20 km in length and 1–4 km in width, with a surface area of about 45 km². EBMU represents the western extension of the El-Baroud Layered Gabbros [33] and the southern extension of the Wadi Abu Hadieda post-collisional gabbroic intrusion [34] and is exposed along Wadi El-Bola (Figure 1c), from which it derives its name. The area is structurally complex, intersected by steeply dipping faults trending NE and NW, predominantly exhibiting sinistral displacement (Figure 1c), and the area is predominantly composed of the following rocks in a chronological sequence (oldest to youngest):gneisses, metavolcanic rocks, metagabbro–diorite complexes, the mafic–ultramafic intrusion, and finally, younger granite (Figure 1c).

The El Bola gneisses represent the southern portion of the El-Markh and Abu Furad Gneisses [30]. These gneisses represent the early formed rocks within the El Bola area and dominate the western portion of the study area and extend further westward beyond the mapped boundaries (Figure 1c), are exposed along the tributaries of Wadi El-Bola, and have been intruded by both metagabbro–diorite and older granites (Figure 2a).

The metavolcanic rocks occupy the southern sector of the study area, extending further southward and to the east. These metavolcanic rocks are considered as a part of the older metavolcanics (OMVs) succession [35]. They are intruded to the north by metagabbro–diorite and EBMU (Figure 2b), while to the west they are cut by older granite (Figure 1c), where intrusive contacts are clearly observed. The metavolcanites consist mainly of metabasaltic andesites, meta-andesites, and their pyroclastic equivalents, and geochemically they display a pronounced calc-alkaline affinity [36].

Metagabbro–diorites form an elongated mass, located centrally within the study area. It has intruded into a sequence of calc-alkaline metavolcanic rocks and, in turn, is cross-cut by the EBMU (Figure 2c) and as well as by both older and younger granitic phases, where their characteristic offshoots and apophyses are clearly observed.

Older granitic rocks represented by tonalite and granodiorite were emplaced approximately at 663 Ma (Figure 1c) [37,38,39]. Field evidence indicates that this granitoid pluton intrudes all surrounding rock units except for the mafic–ultramafic intrusion (Figure 2d) and the younger granite phases (Figure 2e). The intrusive contacts often exhibit features of magmatic stopping, particularly along their roofs and margins, where narrow zones (typically a few meters wide) display well-developed foliation (Figure 2a). Additionally, spheroidal xenoliths of varying sizes are frequently observed within the granitic body.

The El Bola mafic–ultramafic intrusion (EBMU), which represents the focus of the present study, forms an elongated body and intrudes into the metagabbro–diorite (Figure 2c,d), containing xenoliths of varying sizes and shapes, some of which exhibit localized thermal alterations at their contacts with surrounding country rocks. The EBMU is itself cut by younger granitic phases (Figure 2e,f). The EBMU is minimally weathered, with moderate-to-high-relief hills, with a characteristic dark greenish-grey to black color due to its high mafic mineral contents (Figure 2d). It represents a prominent example for rhythmically layered ultramafic–mafic intrusions in the Egyptian Eastern Desert. Field observations during sampling revealed defined rhythmic layering where individual layers grade gradually into one another without distinct contacts and are a few centimeters thick, displaying subtle variations in pyroxene content (Figure 2g). It comprises the following rock varieties: pyroxenites, gabbros, and subordinate anorthositic gabbros. Pyroxenites are less abundant and are exposed to the northeastern portion of the intrusion. Proceeding southwards, the pyroxenites gradually transition into melanocratic gabbros, which in turn grade into normal gabbros. Field observations reveal that pyroxenites and gabbroic rocks are interlayered, yet without distinct thermal and/or intrusive contacts (Figure 2b), and abrupt contacts between pyroxenite and gabbros were not affirmatively observed in the field. Anorthositic gabbros are present as small, scattered veins and pockets within the broader gabbroic units (Figure 2h). Sometimes the gabbroic rocks constitute centimeter-scale stringers and pockets of pegmatitic gabbros, where they are distinguishable by the abundant megacrysts of plagioclase and pyroxene (Figure 2i,j).

The younger granitoids in the El Bola area are predominantly fine-to-medium grained and include compositions from monzogranite to alkali feldspar granite [29,30,31]. These granitoids intruded all earlier rocks, including the EMBU (Figure 2f).

3. Petrographic Description of the EBMU Intrusion

The El Bola mafic–ultramafic intrusion (EBMU) is composed of fresh, unmetamorphosed mafic and ultramafic igneous rocks. Modal analysis of EBMU rocks was conducted through detailed point counting (Supplementary Table S1); five distinct rock units have been identified within the intrusion. These rocks reflect varying mineralogical compositions and textures, which provide insights into the magmatic differentiation and crystallization history of the intrusion. The EBMU rock varieties exhibit a range of cumulate textures, varying from adcumulate to heteradcumulate types. Their modal mineral compositions, along with their petrographic classifications, are summarized in Supplementary Table S1 and illustrated in Figure 3. These textural and mineralogical variations provide important constraints on the crystallization sequence and magmatic evolution of the intrusion.

The pyroxenite is a massive medium-grained rock, typically dark to black in color, and composed mainly of interlocking network of euhedral to subhedral pyroxene crystals, primarily diopside (40–50 vol.%) and enstatite (20–30 vol.%), with olivine occurring as cumulus anhedral to subhedral grains. Olivine grains often display fresh, unaltered cores, while their rims may show partial alteration to serpentine and magnetite (Figure 4a); meanwhile, some grains host inclusions of cumulus Cr-spinel (Figure 4b), indicating early crystallization within a high-temperature magmatic environment. Texturally, the pyroxenite is predominantly adcumulate, with up to 5% intercumulus phases including pyroxene and amphibole (Figure 4b). However, in certain localities, the texture transitions into a mesocumulate structure, reflecting slight variations in crystal accumulation and post-cumulus melt presence during solidification.

Gabbroic rocks dominate the EBMU, accounting for over 90% of its total volume. These include olivine gabbro, pyroxene gabbro, and pyroxene–hornblende gabbro varieties.

The olivine gabbro is a coarse-grained orthocumulate rock, composed mainly of cumulus plagioclase (40–45 vol.%), and olivine (20–35 vol.%), with minor intercumulus augite, and hornblende Orthopyroxene (Opx) occurs in trace amounts, whereas Fe-Ti oxides and apatite represent accessory minerals (Figure 4c). Olivine and plagioclase were the first crystallizing phase.

Figure 4. Petrographical features of the EBMU rocks showing (a) pyroxenite consisting mainly of granoblastic pyroxene and olivine cumulate; (b) representative BSE (back-scattered electron) image of pyroxenite exhibiting cumulus Cr-spinel within olivine which has been altered to magnetite along the rim; (c) olivine gabbro consists of olivine and plagioclase with pyroxene and amphibole; (d) BSE of olivine gabbro exhibiting orthopyroxene overgrown olivine forming a corona texture and plagioclase within hornblende forming a poikilitic texture; (e) pyroxene gabbro exhibiting subophitic texture. Clinopyroxene and orthopyroxene are the dominant mafic phases, displaying high-to-moderate interference colors with interstitial growth between the plagioclase and pyroxenes. And minor hornblende as an intercumulus or alteration product; (f) BSE of pyroxene gabbro displaying coarse-grained cumulate texture. The image shows intergrown clinopyroxene and orthopyroxene crystals, with plagioclase occurring as laths and interstitial grains. Amphibole appears as both replacement and intercumulus phases around pyroxenes and plagioclase, indicating late-magmatic or subsolidus hydration; (g) pyroxene–hornblende gabbro exhibiting euhedral to subhedral plagioclase, forming grain clots and exhibiting pericline twinning; (h) BSE of primary amphibole hosting plagioclase secondary amphibole that contains pyroxene relicts in the core; (i) hornblende gabbro consisting of cumulus plagioclase, (20–35 vol%) intercumulus hornblende, and interstitial diopside; (j) BSE of primary amphibole hosting plagioclase forming poikilitic texture. Mineral abbreviations [41]: (Ol) olivine, (Cpx) clinopyroxene, (Opx) orthopyroxene, (Mag) magnetite, (Spl) spinel, (Amp) amphibole, (Hb) hornblende, (Pl) plagioclase, (Ilm) ilmenite, and (ap) apatite.

Figure 4. Petrographical features of the EBMU rocks showing (a) pyroxenite consisting mainly of granoblastic pyroxene and olivine cumulate; (b) representative BSE (back-scattered electron) image of pyroxenite exhibiting cumulus Cr-spinel within olivine which has been altered to magnetite along the rim; (c) olivine gabbro consists of olivine and plagioclase with pyroxene and amphibole; (d) BSE of olivine gabbro exhibiting orthopyroxene overgrown olivine forming a corona texture and plagioclase within hornblende forming a poikilitic texture; (e) pyroxene gabbro exhibiting subophitic texture. Clinopyroxene and orthopyroxene are the dominant mafic phases, displaying high-to-moderate interference colors with interstitial growth between the plagioclase and pyroxenes. And minor hornblende as an intercumulus or alteration product; (f) BSE of pyroxene gabbro displaying coarse-grained cumulate texture. The image shows intergrown clinopyroxene and orthopyroxene crystals, with plagioclase occurring as laths and interstitial grains. Amphibole appears as both replacement and intercumulus phases around pyroxenes and plagioclase, indicating late-magmatic or subsolidus hydration; (g) pyroxene–hornblende gabbro exhibiting euhedral to subhedral plagioclase, forming grain clots and exhibiting pericline twinning; (h) BSE of primary amphibole hosting plagioclase secondary amphibole that contains pyroxene relicts in the core; (i) hornblende gabbro consisting of cumulus plagioclase, (20–35 vol%) intercumulus hornblende, and interstitial diopside; (j) BSE of primary amphibole hosting plagioclase forming poikilitic texture. Mineral abbreviations [41]: (Ol) olivine, (Cpx) clinopyroxene, (Opx) orthopyroxene, (Mag) magnetite, (Spl) spinel, (Amp) amphibole, (Hb) hornblende, (Pl) plagioclase, (Ilm) ilmenite, and (ap) apatite.

Plagioclase occurs as cumulus coarse prismatic laths and is sometimes enclosed in olivine, clinopyroxenes (Cpxs), and amphiboles, displaying poikilitic textures (Figure 4d). Olivine crystals are subhedral to anhedral, commonly fractured, and partially altered to chlorite and opaque minerals. Orthopyroxenes (Opxs) are typically overgrown with olivine (Figure 4d), while amphibole often encloses earlier phases and locally replaces Cpx (Figure 4d).

The pyroxene gabbro (px-gabbro) is fine-to-medium-grained, grayish green in color, and comprises plagioclase (55–60 vol.%), Cpx (10–20 vol.%), Opx (7–10 vol.%), and amphibole (3–5 vol.%). Accessory minerals include Fe-Ti oxides (e.g., magnetite and ilmenite) at up to 15 vol.% along with rare olivine and apatite (Figure 4e,f). Plagioclase displays elongated crystals displaying lamellar and pericline twinning, whereas pyroxenes occur as a subhedral intercumulus grain between the plagioclase. Cpx is greenish-brown, whereas Opx forms pale-brown crystals surrounding plagioclase, displaying a common ophitic texture. Amphibole rims around pyroxenes create a distinct corona texture in some samples (Figure 4f).

The pyroxene–hornblende gabbro (px-hbl-gabbro) is medium-to-coarse grained, and consists primarily of plagioclase (52%–66%), pyroxene (13%–20%), and brown hornblende (20%–30%) with accessory biotite, Fe-Ti oxides, and apatite. Plagioclase forms euhedral to subhedral grains that are often clustered and displaying a pericline twinning (Figure 4g). Cpx appears as subhedral, pale green to colorless crystals that either separate plagioclase from hornblende or appear as relict cores rimmed by hornblende (Figure 4h). Opx typically occurs as subhedral pale brown grains, occasionally enclosed within hornblende. Hornblende, the dominant mafic phase, displays pleochroic yellow-brown crystals that enclose inclusions of apatite and pyroxene, as well as plagioclase, exhibiting partial alteration to chlorite and iron oxide minerals. Subhedral Fe-Ti oxide crystals are scattered among the major mineral phases (Figure 4g,h) and are often enclosed within hornblende (Figure 4c,d). Apatite typically occurs as a small-to-medium inclusion within hornblende (Figure 4c and Figure 5b).

The hornblende gabbro (hbl-gabbro) is medium-to-fine-grained and consists of cumulus plagioclase (42–60 vol%) and intercumulus hornblende (20–35 vol%) with accessory phases such as diopside, apatite, and magnetite (Figure 4i). Plagioclase is typically euhedral and medium-grained, occasionally enclosed by coarse hornblende oikocrysts (Figure 4j). Hornblende appears as euhedral to subhedral interstitial grains. Diopside occurs as short prismatic interstitial grains, partly altered to uralite, and often forms corona textures, where Fe-Ti oxides are expelled along the pyroxene cleavage and surrounded by amphibole and/or biotite (Figure 4j).

4. Materials and Methods

The chemical composition of the minerals was detected by a CAMECA SX100 electron microprobe at the Deutsches GeoForschungs Zentrum (GFZ) in Potsdam, Germany. Operating conditions included an accelerating voltage of 15 kV and a beam current of 20 nA. Peak counting times were 20–30 s for major elements and 30 s for minor elements, with background counts collected for 5–15 s. Data reduction performed by the PAP correction method was implemented with the CAMECA software (5.21 of PeakSight).

Whole-rock major oxides (wt.%) and trace element concentrations (ppm) were determined by X-ray fluorescence (XRF) spectrometry on fused glass beads prepared from 0.4 g of sample powder and 4 g of lithium tetraborate flux. Analyses were performed by a Rigaku ZSX-100e XRF spectrometer at Kagoshima University, Kagoshima, Japan. Fourteen Geological Survey of Japan (GSJ) standard reference materials (JA-1, JA-2, JA-3, JB-1a, JB-2, JB-3, JF-1, JF-2, JG-1, JG-1a, JG-2, JG-3, JGb-1, and JP-1) were employed for calibration and matrix correction, following the procedure of [42]. The detection limits were 0.01 wt.% for major elements, 0.001 ppm for most trace elements, and 1 ppm for Ag, Cu, and Mo. Loss on ignition (LOI) was employed, where the powdered samples were heated at 650 °C for six hours to remove volatile components. After this ignition process, the powders were prepared for XRF analysis. Rare earth element (REE) measurements were conducted at ACME Analytical Laboratories (Bureau Veritas) in Vancouver, BC, Canada, where REEs were measured in the acid-digested sample by Inductively Coupled Plasma–Mass Spectrometry (ICP-MS). The REE data have an accuracy and precision of more than 8% (relative), as determined by monitoring and repeat analyses compared to international reference standards. For other details see the following: www.acmelab.com (accessed on 27 May 2025).

5. Results

5.1. Mineral Chemical Composition

Representative EPMA data for spinel, olivine, pyroxene, amphibole, and plagioclase are provided in Supplementary Tables S2–S6.

5.1.1. Chromite

Chromite grains analyzed from the pyroxenite are identified as igneous aluminian chromite (Figure 5a; [43]), showing no signs of alteration into ferrichromite or chromian magnetite, unlike the ophiolitic ultramafics of the ANS [11,44]. Their compositions (Figure 5a) align closely to those of the ANS layered ultramafic intrusions [4,45]. They possess relatively high concentrations of Cr (38.84–49.83 wt.%) and Fe (23.82–43.84 43 wt.%) with low Al (10.82–22.843 wt.%), Cr# values of 0.56–0.71, and Mg# values ranging from 0.20 to 0.43, which are the typical characteristics of layered intrusions rather than those of komatiites (Figure 5b).

5.1.2. Olivine

Olivine is present in both the pyroxenite and olivine gabbro. Within the pyroxenite, olivine shows forsterite (Fo) values between 78.06 and 80.77, slightly higher than those in the olivine gabbro (Fo = 76.67–77.08), indicating progressive fractional crystallization (Supplementary Table S3; Figure 5c). They possess low contents of NiO (0.11–0.14 wt.%), significantly lower than typical ophiolitic olivine (Figure 5c; [9,12,46,47]), whereas they resemble those from the ANS layered mafic–ultramafic associations Figure 5d; e.g., [4,15,45]) and are distinct from those of the Alaskan-type [5,13,48,49,50]. Their MnO contents are negatively correlated with their Fo, supporting differentiation or crystal segregation (Figure 5e). Overall, their low contents of Fo (Fo < 83) align with those of unmetamorphosed, non-ophiolitic intrusions of the ANS Fo < 83; [9,45,51,52,53]).

Figure 5. Spinel and olivine chemical composition of the EBMU rocks: (a,b) chromite compositions of EBMU pyroxenites; (b) labeled fields show comparisons to chromites in komatiites [54], to mafic–ultramafic layered intrusions [55], and to chromite from the ANS layered intrusions [4,16]; (c) NiO (wt.%) versus Fo content diagram for olivines. Compositional ranges of the ANS layered intrusions [56,57,58], Egyptian ophiolite [9,12,47], and Egyptian Alaskan intrusions [5,9,18] are employed to facilitate comparison. The field of ophiolitic olivine is adapted from [46]; (d) comparison of the forsterite (Fo) content in the studied olivines with that of the global Alaskan-type olivines. Data sources: the Alaskan-type complexes of Quetico and Samuel Lake [59]; Union Bay (Polaris) and Duke Island [60]; Blashke Island [61]; Motaghairat [4]; Abu Hamamid [5,48]; Genina Gharbia [5]; Gabbro Akarem [13], Dahanib [9]; Alaskan Complexes [60]; and Arabian Shield [62]; (e) variation diagram of MnO (wt.%) and Fo content of the olivines; fractional crystallization trend after [63]. The fields of ANS mafic–ultramafic intrusion are the same as (c).

Figure 5. Spinel and olivine chemical composition of the EBMU rocks: (a,b) chromite compositions of EBMU pyroxenites; (b) labeled fields show comparisons to chromites in komatiites [54], to mafic–ultramafic layered intrusions [55], and to chromite from the ANS layered intrusions [4,16]; (c) NiO (wt.%) versus Fo content diagram for olivines. Compositional ranges of the ANS layered intrusions [56,57,58], Egyptian ophiolite [9,12,47], and Egyptian Alaskan intrusions [5,9,18] are employed to facilitate comparison. The field of ophiolitic olivine is adapted from [46]; (d) comparison of the forsterite (Fo) content in the studied olivines with that of the global Alaskan-type olivines. Data sources: the Alaskan-type complexes of Quetico and Samuel Lake [59]; Union Bay (Polaris) and Duke Island [60]; Blashke Island [61]; Motaghairat [4]; Abu Hamamid [5,48]; Genina Gharbia [5]; Gabbro Akarem [13], Dahanib [9]; Alaskan Complexes [60]; and Arabian Shield [62]; (e) variation diagram of MnO (wt.%) and Fo content of the olivines; fractional crystallization trend after [63]. The fields of ANS mafic–ultramafic intrusion are the same as (c).

5.1.3. Pyroxenes

Pyroxenes are found in all rock types of the EBMU (Figure 4). The results of the analyzed pyroxenes from the EBMU are compiled in Supplementary Table S4. Following Morimoto, Fabries, Ferguson, Ginzburg, Ross, Seifert and Zussman [53], the pyroxene compositions were calculated assuming six oxygens in the formula unit.

Orthopyroxenes in all EBMU rocks are predominantly enstatite (Figure 6a), with generally uniform composition (Supplementary Table S4). Their compositions vary slightly by rock type: pyroxenite (Wo_1–3En_73–75Fs_24–25), olivine gabbro (Wo_2–3En_72–73Fs_24–25), pyroxene gabbro (Wo_2–3En₇₂Fs_25–26), Px-hb-gabbro (Wo_0–2En_69–71Fs_27–28), and hbl-gabbros (Wo_2–4En_65–68Fs_27–33). The Mg# content of the analyzed Opx grains decreases systematically from 0.76 in pyroxenite to 0.6 in hbl-gabbros, suggesting igneous origin (Figure 6b). Moreover, the analyzed Opx of the EBMU aligns with a low-pressure fractionation trend (Figure 6c), resembling those of the Korab Kansi layered intrusion [15] and differing from those of the Alaskan-type of Genina El-Gharbia [5] in the ANS. In addition, their chemical compositions are consistent with those of boninite (Figure 6d).

The clinopyroxenes of the EBMU show a wide compositional range between different rocks: pyroxenite (Wo_41–49En_46–52Fs_5–7), olivine gabbros (Wo_43–48En_44–49Fs_7–9), px-gabbros (Wo_39–48En_42–49Fs_-11), px-hbl-gabbros (Wo_43–49En_40–45Fs_1–12), and hbl-gabbros (Wo_44–46En_41–43Fs_12–13). Most of the Cpxs of the EBMU are plotted as diopside (Wo_45–49), whereas a few samples have more tendencies to augite (Wo_39–45; Figure 6a). TiO₂ content is highest in pyroxenite and olivine gabbro (up to 0.90 and 0.82 wt.%, respectively), while Mg# values range from 0.72 to 0.92, decreasing from pyroxenite to hornblende gabbro (Supplementary Table S4). They possess relatively low concentrations of Al₂O₃ (Supplementary Table S4; Figure 6e), supporting a subalkaline affinity which is consistent with those of many layered intrusions of the ANS such as Motaghairat, Wadi El Dib, Korab Kansai, and Rahaba [4,15,17].

5.1.4. Amphiboles

Amphiboles occur in all EBMU rocks, but they were analyzed from three rocks of the EBMU (px-gabbro, px-hbl-gabbros, and hbl-gabbros) where they occur abundantly (Figure 4). The analyzed amphiboles exhibit a broad compositional spectrum with respect to their silica contents (42.82–47.24 wt.%), whereas they exhibit limited variation with respect to TiO₂, Al₂O₃, FeO, and MgO (Supplementary Table S5).

The analyzed amphiboles from the EBMU intrusion are calcic (Figure 7a; [73]) with high silica contents (Tsi > 6.27), and Mg# is generally > 0.68 (Supplementary Table S5), indicating a magnesian nature similar to their coexisting pyroxenes (Supplementary Table S4). Moreover, the analyzed amphiboles are mainly differentiated into two subgroups (Figure 7b): group 1 which is primary amphiboles (mainly pargasite; Figure 7c) consisting of to px-gabbro and group 2 that represents secondary amphiboles (mainly magnesiohornblende, Figure 7c) that replace pyroxenes (Figure 4h) and mainly contain both px-hbl- and hbl-gabbro. Pargasite, with higher TiO₂, Al₂O₃, and Na₂O contents, reflects crystallization from mantle-derived magmas (Figure 7c–e). These amphiboles formed under low-to-moderate pressures (~5 kbar; Figure 7f).

5.1.5. Plagioclase

The data analyzed for plagioclase from different gabbroic rocks of the EBMU are given in Supplementary Table S6. It is clear that the analyzed plagioclase exhibits a wide compositional range between different rocks (Supplementary Table S6; Figure 7g). Plagioclases from both the olivine gabbros and px-gabbro are typically calcic (An_85–90 and An_90–92, respectively), whereas those from the px-hbl- and hbl-gabbros are less calcic (An_53–66 and An_42–49, respectively; Figure 7g).

5.1.6. Oxide Minerals

Ilmenite and magnetite represent the dominant oxide mineral phases in the EBMU (Figure 4f; Supplementary Tables S1 and S7). Ilmenite compositions show moderate contents of TiO₂, FeO, MnO, and V₂O₅ contents (up to 48.85, 48.47, 2.83, and 3.58 wt.%, respectively), with low MgO and Al₂O₃ (≤2.22 and 0.13 wt.%). These characteristics align with those of typical layered intrusions within the ANS (Figure 7h). According to the Stormer [74] method, the analyzed ilmenites display minimal hematite contribution (<9%mol). Their low MgO levels suggest crystallization from mafic melts [75]. High V₂O₅ levels point to an oxidized magmatic environment and a V–Ti-rich parental magma source [19,76]). On the other hand, the analyzed magnetite from different rocks of the EBMU possess relatively high contents of FeO (74.33–82.74 wt.%; Supplementary Table S7) and are characterized by slight enrichment of TiO₂ contents (1.48–5.12 wt.%) and V₂O₅ (0.14–0.68 wt.%). According to the method of [74]; the analyzed magnetite possesses less than 5.9 mol.% of ulvospinel content.

Figure 7. Amphibole chemical composition of El Bola mafic–ultramafic intrusion: (a) general classification diagrams of amphiboles [77], with the main four groups depending on the occupancy of the B site; (b) compositional variations of amphibole plotted in terms of cations (Ca+Na+K) vs. Si (apfu) diagram after [78]; (c) (Na + K + 2Ca)A vs. (Al + Fe³⁺+2Ti)A amphibole classification diagram [73] showing the composition of the analyzed amphiboles; (d) Ti-alkalis variation diagram for the analyzed amphiboles [79]; (e) correlation between TiO₂ and Al₂O₃ in primary amphiboles [80]; (f) Si vs. Al^VI [81] binary diagrams for the analyzed amphiboles; (g) feldspar classification diagram of the EBMU intrusion [82]; (h) ternary plot of cation proportions of Fe²⁺, Fe³⁺, and Ti⁴⁺ solid system diagram showing the composition and approximate equilibrium tie lines (dashed lines) between analyzed ilmenite and magnetite from the EBMU intrusion after [83,84]). Fields of ilmenite and magnetite from ANS are adopted from [15,19,33].

Figure 7. Amphibole chemical composition of El Bola mafic–ultramafic intrusion: (a) general classification diagrams of amphiboles [77], with the main four groups depending on the occupancy of the B site; (b) compositional variations of amphibole plotted in terms of cations (Ca+Na+K) vs. Si (apfu) diagram after [78]; (c) (Na + K + 2Ca)A vs. (Al + Fe³⁺+2Ti)A amphibole classification diagram [73] showing the composition of the analyzed amphiboles; (d) Ti-alkalis variation diagram for the analyzed amphiboles [79]; (e) correlation between TiO₂ and Al₂O₃ in primary amphiboles [80]; (f) Si vs. Al^VI [81] binary diagrams for the analyzed amphiboles; (g) feldspar classification diagram of the EBMU intrusion [82]; (h) ternary plot of cation proportions of Fe²⁺, Fe³⁺, and Ti⁴⁺ solid system diagram showing the composition and approximate equilibrium tie lines (dashed lines) between analyzed ilmenite and magnetite from the EBMU intrusion after [83,84]). Fields of ilmenite and magnetite from ANS are adopted from [15,19,33].

5.1.7. Whole-Rock Chemical Composition

A comprehensive geochemical dataset, including major, trace, and REE compositions for 28 selected EBMU whole-rock samples, is compiled in Supplementary Table S8. The EBMU samples are generally fresh and show weak alteration which is reflected by their low-to-moderate values of loss on ignition (LOI; 0.58–2.75 wt.%; Supplementary Table S8). The EBMU rocks display significant chemical variability, with their Mg# values ranging from 0.83 to 0.92 (Supplementary Table S8). Pyroxenite possesses the highest content of MgO with high concentration of Ni and Cr but has the lowest SiO₂ content among the other analyzed rocks. With its high concentrations of SiO₂ and Al₂O₃ contents and notably low MgO concentration (Supplementary Figure S1), Hbl-gabbro represents the most evolved composition derived from the primitive magmatic source. The Harker diagram (Supplementary Figure S1) shows an increase in FeO alongside MgO, coupled with declining SiO₂ and Al₂O₃ levels, implying the influence of fractional crystallization processes during the magmatic evolution of the EBMU. These rocks are significantly alkali poor, with TiO₂ content rising from pyroxenite to olivine gabbro, then declining toward hornblende gabbro. Olivine gabbro has the highest Mg (0.92), MgO (23.05 wt.%), Cr (up to 498 ppm), and Ni (up to 548 ppm) contents and has elevated FeO* content (up to 13.51 wt.%) but is low in SiO₂ content (41.04–42.45 wt.%) relative to other gabbroic rocks (Supplementary Table S8).

The chemical composition of the EMBU rocks closely resembles that of primary ferropicritic melts (Figure 8a; [85]), suggesting that the composition of olivine gabbro represents the original melt composition. The EMBU samples align with a differentiation trend, marked by increasing silica and alkali content (Figure 8a). The Olivine gabbro plots are within the picrite field, while other gabbroic types fall into the subalkaline gabbro field (Figure 8a). Hbl-gabbros appear to be the most evolved, displaying the lowest amounts of TiO₂ (0.46–0.76 wt.%) and MgO (5.34–6.46 wt.%) with moderate FeO (6.43–6.98 wt.%). Most of the EBMU rocks display a distinct tholeiitic affinity on the AFM diagram except for some samples of px-hbl- and hbl-gabbro which display transitional characteristics between the calc-alkaline and tholeiitic affinity resembling ANS post-collision mafic rocks (Figure 8b). The shift from tholeiitic to calc-alkaline traits in the EBMU rock samples is obviously displayed on a SiO₂ vs. FeOt/MgO plot (Supplementary Figure S2a). Moreover, the EBMU rocks exhibit low amounts of K₂O (<0.5 wt.%) comparable to those found in the ANS post- rogenic layered mafic–ultramafic intrusions (e.g., Motaghairat [4]; Korab Kansi [15]; Rahaba, [16]; Wadi El Dib, [17]), where they straddle in the field of low-K tholeiitic series, except some samples of px-hbl-gabbro have tendencies towards the calc-alkaline field (Supplementary Figure S2b), indicating the transition of the EBMU magma from primitive low-K tholeiitic magmas toward more evolved calc-alkaline magmas. Notably, most of the ANS post-orogenic layered intrusions display transitional features between tholeiitic and calc-alkaline types, with few distinctly fitting solely into either the tholeiitic or calc-alkaline categories [86].

The EMBU rocks have a homogenous distribution of trace elements reflecting their mineral compositions. These rocks display varying concentrations of compatible trace elements, including Cr (16–106 ppm), Ni (63–460 ppm), Co (36–106 ppm), and Sc (10–43 ppm). The progressive dropdown of Ni, Co, and Cr concentrations from pyroxenite to hbl-gabbros reflects a settling of the olivine and Opx-Cpx in pyroxenite (Supplementary Figure S1). Large Ion Lithophile Elements (LILEs) such as Rb (2–14 ppm), Ba (25–215 ppm), and Sr (200–875 ppm) are present in notable amounts, while Zr (13–34 ppm) generally increases from ultramafic to hbl-gabbros (Supplementary Figure S1). Additionally, EBMU rocks exhibit low concentrations of High Field Strength Elements (HFSEs), including Th (0.05–0.16 ppm), U (0.03–0.8 ppm), Nb (0.48–1.63 ppm), and Ta (0.04–0.16 ppm).

Primitive mantle-normalized incompatible trace element patterns for the EBMU intrusion show that both mafic and ultramafic rocks display similar patterns (Figure 8c,d), supporting their common magmatic origin. The ultramafic and olivine gabbro units display a distinct positive Ti anomaly, likely resulting from early crystallization of Ti-rich phases such as ilmenite. In contrast, the hbl-gabbro shows a negative Ti anomaly due to its lower Ti content. The EBMU suite is characterized by elevated concentrations of LILEs like Pb, Rb, Ba, and Sr, alongside notable depletions in Nb and Ta (Figure 8c,d). Additionally, the rocks exhibit elevated LILE/HFSE values (e.g., Sr/Th, Rb/Th, Sr/Zr, and Ba/Nb; Supplementary Table S8), matching with most of the ANS post-collisional layered mafic–ultramafic intrusions (Figure 8a; e.g., [4,17,19,33,45]). Px-hbl- and hbl-gabbro possess greater amounts of Rb, Ba, and Sr compared to the other gabbroic samples, with their predominate quantity of plagioclase (Supplementary Table S1).

EBMU rocks show distinct negative anomalies in Ta and Nb, a hallmark of arc-related magmatic sources [87]. Additionally, they display strong positive anomalies in Pb and U relative to adjacent elements (e.g., Ce, Pr, Th, and Nb), suggesting enrichment from a metasomatized mantle source [88].

The total concentrations of rare earth elements (∑REE) within the EBMU progressively increase from pyroxenite to hbl-gabbro (Figure 8e,f). Chondrite-normalized REE patterns show enrichment in light REEs, with (La/Lu)_N values ranging between 3.28 and 6.31 (Supplementary Table S8; Figure 8), aligning with those of other post-orogenic layered mafic–ultramafic intrusions of the ANS [1.03–9.21; 4,15]). The LREE/HREE ratio in the EBMU rocks increases toward hornblende gabbro, indicating progressive fractionation (Supplementary Table S8). Nearly all samples show a slightly positive Eu anomaly, with the weakest anomaly appearing in ultramafic and olivine gabbro (average Eu/Eu* = 0.98 and 1.01), whereas the strongest is observed with px-, px-hbl-, and hbl-gabbros (Eu/Eu* = 1.20, 1.30, and 1.63), suggesting increased plagioclase accumulation. The REE patterns closely resemble those of ANS layered intrusions, distinguishing them from ANS Alaskan-type intrusions (Figure 8e,f).

Figure 8. Whole-rock chemical classification: (a) total alkali vs. silica (TAS) diagram [89]; dividing line between tholeiitic and alkaline rock series after [90]; (b) AFM diagram; dividing line between tholeiitic and calc-alkaline field after [91]. Fields of cumulate and non-cumulate mafic–ultramafic rocks are from [92]; (c,d) primitive mantle-normalized trace element patterns; (e,f) chondrite-normalized REE patterns of the EBMU intrusion. Data for ANS layered mafic–ultramafic intrusion are compiled from [4,15,17,45]; primitive mantle, chondrite normalizing values, N-MORB, E-MORB, and OIB data are from [93].

Figure 8. Whole-rock chemical classification: (a) total alkali vs. silica (TAS) diagram [89]; dividing line between tholeiitic and alkaline rock series after [90]; (b) AFM diagram; dividing line between tholeiitic and calc-alkaline field after [91]. Fields of cumulate and non-cumulate mafic–ultramafic rocks are from [92]; (c,d) primitive mantle-normalized trace element patterns; (e,f) chondrite-normalized REE patterns of the EBMU intrusion. Data for ANS layered mafic–ultramafic intrusion are compiled from [4,15,17,45]; primitive mantle, chondrite normalizing values, N-MORB, E-MORB, and OIB data are from [93].

6. Discussion

6.1. Geothermobarometric Estimations of the EBMU Rocks

Multiple thermobarometric methods based on the chemical composition of minerals (pyroxenes, amphiboles, plagioclase, and oxide minerals) were employed to evaluate pressure (P), temperature (T), and oxygen fugacity (fO₂) during mineral crystallization in the EBMU intrusion (Supplementary Table S9). Using a pyroxene thermometer (Figure 6a; [64]), high crystallization temperatures were yielded, with pyroxenite and olivine gabbro exhibiting the highest temperatures (980–1120 °C and 990–1120 °C, respectively; Supplementary Table S9), then progressively decreasing to 800–940 °C in hbl-gabbro, matching well with the evolutionary trend of fractional crystallization within the intrusion. The resultant broad crystallization temperature range reflects a lengthy crystallization period that consists of various changes in the composition of magma during cooling and fractional crystallization (Figure 6; Supplementary Table S9). Moreover, pyroxene thermometer results [94] further support these trends, with olivine gabbro recording temperatures between 926 and 1026 °C and hornblende gabbro exhibiting lower crystallization temperatures between 767 and 856 °C. Clinopyroxene thermometry [95] yields similar results (975–1063 °C in pyroxenite and 816–923 °C in hornblende gabbro), confirming the consistency of different calibration methods.

Pressure estimates derived from [95] a clinopyroxene barometer suggest low-to-moderate crustal emplacement depths. Pyroxenite and olivine gabbro yielded pressures of approximately 5.24 and 5.11 kbar, respectively, while pyroxene gabbro pressures are similar (~5.12 kbar). In contrast, pyroxene–hornblende gabbro records lower pressures (~3.48 kbar), suggesting emplacement at shallower crustal levels during the later stages of magmatic differentiation. No pressure values could be determined for hornblende gabbro due to the absence of suitable pyroxene pairs (Supplementary Table S9).

In comparison with mafic–ultramafic intrusions of the ANS; the estimated temperature of the EBMU intrusion (~716–1100 °C) according to various pyroxene thermometers are consistent with those Korab Kansi layered mafic–ultramafic intrusion (700–1100 °C; [15]), but slightly higher than those of Rahaba layered mafic–ultramafic intrusion (820–1100 °C) [15].

Amphibole thermometers were applied for the primary amphiboles from px-hbl gabbro and yielded consistent results with the estimated temperatures of pyroxenes (Supplementary Table S9). Applying Putirka [96] thermometer yields crystallization temperatures of 986–1002 °C for px-gabbro, supporting the thermal evolution trend toward cooler conditions during late-stage crystallization. Plagioclase–hornblende thermometer readings [97] yielded similar temperatures of 998–1017 °C for px-gabbro. Furthermore, estimated temperatures from coexisting amphibole–plagioclase pairs (Figure 9a; [98]) for the analyzed primary amphiboles ranged between 890 °C and 1020 °C (Figure 9a).

Oxide thermometer based on coexisting magnetite–ilmenite pairs [99] implies lower equilibrium temperatures, from 730–752 °C in pyroxenite to 670–672 °C in hbl-gabbro, pointing to late-stage cooling and/or subsolidus re-equilibration. Calculated oxygen fugacity (log ƒO₂) values also vary systematically with rock type, where pyroxenite and olivine gabbro display lower average oxygen log ƒO₂ (−17.21 and −17.33, respectively), while the more evolved px-hbl-gabbro records higher values (−15.67; Supplementary Figure S2c), consistent with progressive oxidation during the differentiation stage of magmatic evolution.

Figure 9. Estimated crystallization pressure, temperature, and magma type of the EBMU intrusion: (a) plots of amphibole–plagioclase equilibrium pairs were plotted following the classification scheme of [98]; (b) plot of Al^iv vs. Na cation in B-site of the studied primary amphiboles. Calculated pressure trends after [100]; (c) plot of log fO₂ vs. temperature. All buffers are from [101]; titanite+ magnetite + quartz = ilmenite + hedenbergite + O₂ and titanite + fayalite = ilmenite + hedenbergite + O equilibria from [102]; (d) Na₂O/K₂O vs. Na₂O+K₂O diagram [103] for the studied EBMU rocks; (e) Pearce elemental ratios diagram showing the dominant phases of fractionation within the EBMU intrusion [104]; (f) Al₂O₃ against TiO₂ of Cpx [105]; (g) FeO^t + Ti − Mg − Al cation plot of gabbros [106]; (h) Al^total vs. Ti clinopyroxenes’ cations; field of Cpxs from ferropicrites after [63,85].

Figure 9. Estimated crystallization pressure, temperature, and magma type of the EBMU intrusion: (a) plots of amphibole–plagioclase equilibrium pairs were plotted following the classification scheme of [98]; (b) plot of Al^iv vs. Na cation in B-site of the studied primary amphiboles. Calculated pressure trends after [100]; (c) plot of log fO₂ vs. temperature. All buffers are from [101]; titanite+ magnetite + quartz = ilmenite + hedenbergite + O₂ and titanite + fayalite = ilmenite + hedenbergite + O equilibria from [102]; (d) Na₂O/K₂O vs. Na₂O+K₂O diagram [103] for the studied EBMU rocks; (e) Pearce elemental ratios diagram showing the dominant phases of fractionation within the EBMU intrusion [104]; (f) Al₂O₃ against TiO₂ of Cpx [105]; (g) FeO^t + Ti − Mg − Al cation plot of gabbros [106]; (h) Al^total vs. Ti clinopyroxenes’ cations; field of Cpxs from ferropicrites after [63,85].

Hbl-gabbro displayed an intermediate value (−17.00; Supplementary Table S9), suggesting minor fluid interaction or re-equilibration effects during late stages of magmatic evolution. Moreover, the estimated fO₂ values likely represent cooling-related re-equilibration displayed by decreases in log fO₂ with descending temperature (Figure 9c). The ilmenite–magnetite pairs plot between the FMQ (fayalite–magnetite–quartz) and NNO (nickel–nickel oxide) buffers displays a trend of higher oxidation states at minimized temperatures. Most samples cluster near the FMQ buffer, while some lower-temperature mafic samples align closer to the NNO buffer (Figure 9c). Notably, the EBMU’s estimated fO₂ values are comparable to those typical of arc-related batholiths [107]. Accordingly, the progressive ascending of the estimated fO2 from pyroxenite to hbl-gabbros within the EBMU supports the island-arc setting of the EBMU intrusion, where its magma has been oxidized under the influence of the slab dehydrating fluid during subduction [108].

The geothermobarometric data collectively show a clear decline in temperature and pressure as magmatic differentiation progresses, reflecting the typical evolution of layered mafic–ultramafic intrusions [1,15,55,66,109]. The ascending trend of the oxygen fugacity value toward highly evolved rocks (Hbl-gabbros) in the EBMU further implies progressive oxidizing conditions during crystallization, which are likely influencing mineral stability and phase assemblages in the EBMU rocks.

6.2. Post-Magmatic Overprints

Petrographic investigations of the EBMU rocks reveal that they are generally unaltered, exhibiting minimal signs of deformation or metamorphism. In addition, their chemical data show depleted LOI values (0.58 to 2.75 ppm; Supplementary Table S8) and lack any Ce anomalies (Figure 8), refuting oxidative or reductive alteration [110]. Moreover, the EBMU mafic rocks possesses alkali concentrations typical of unaltered gabbros (Figure 9d). They possess notably reduced Nb and Ta concentrations (Figure 8d,e) with sub-chondritic Nb/Ta values ranging from 9.65 to 14.35 (Supplementary Table S8), aligning with values expected for those of a mantle signature (10–16; [110]). In addition, their low Lu/Yb ratios (>0.15) support melts of a mantle signature [93,111]. Moreover, the average concentrations of Ba and Sr in the EBMU (285 and 633, respectively) notably exceed those of continental crust (259 and 348, respectively) [111]. Furthermore, the EBMU rocks possess depleted values of La_n/Sm_n < 1.24 which is significantly below those typical for crustal-contaminated basalt (>1.5) [112], ruling out significant crustal contamination of the EBMU. Meanwhile, the depleted Lu/Yb ratio values (0.10–0.18) demonstrate a mantle signature for the parental melt with minimal crustal input [93,111]. Additionally, the EBMU displays negative Th anomalies (Figure 8d,e) with neglected values of Th/La, Th/Ce, and Th/Nb (0.0–0.02, 0.01–0.02, and 0.07–0.16, respectively) which urge against the crustal contamination process [113].

Although crustal input can lead to enrichment in Zr along with depletion in Nb and Ta [114], this is not the case for the EBMU rocks, which consistently show clear negative anomalies in Zr, Nb, and Ta across both the ultramafic and gabbroic varieties (Figure 8d,e), refuting the possibility of crustal contamination. Instead, the Nb-Ta depletion likely reflects a subduction-influenced mantle source, where HFSEs are removed by slab-derived fluids, retained in Ti-rich phases, or are inherently lacking due to arc-related processes [15,34,115]; therefore, the displayed Th anomalies within the EBMU (Figure 8d,e) could be best explained by mantle processes rather than crustal assimilation during magma ascent [116]. The EBMU rocks exhibit smooth, consistent chondrite- and primitive mantle-normalized REE and trace element patterns, indicating minimal post-magmatic alteration or element mobility (Figure 8c,d; [110]). Consequently, the primary geochemical signature of the EBMU rocks is well preserved, where their chemical data could be reliably used to demonstrate the characterizations of their mantle source.

6.3. Fractional Crystallization/Accumulation

Layered intrusions are widely accepted to result from consolidation magma(s) that is/are suffering from a high degree of fractional crystallization [117]. Field observations include the presence of rhythmic layering that is marked by variation in the proportion of pyroxene in the successive layers and the presence of pegmatitic gabbros (Figure 2g,i), supporting the predominance of this crystallization process during magmatic evolution. Petrographic studies of the EBMU rocks exhibit a dominance of cumulate texture and a significant decrease in olivine content with increasing Opxs/Cpxs from pyroxenite to mafic rocks, implying fractional crystallization. The geochemical data from the EBMU is also indicative of fractional crystallization processes as it displays a significant negative correlation between SiO₂ and MgO, whereas MgO exhibits a positive correlation with Ni, Cr, and Co (Supplementary Figure S1), which is indicative of olivine crystallization. Meanwhile, the crystallization of ilmenite and magnetite is indicated by a positive correlation of MgO with FeO and TiO₂, whereas the restriction of negative Ti anomalies to hbl-gabbro only (Figure 8e) suggests that Fe–Ti minerals likely began to crystallize and fractionate after the gabbro had been initially emplaced. Plagioclase fractionation is indicated by the negative correlation of Al₂O₃, CaO, and Eu with MgO (Supplementary Figure S1). The (Mg + Fe)/Ti vs. Si/Ti trends (Figure 9e) suggest that the EBMU magma evolved primarily through the fractionation and accumulation of olivine, plagioclase, and clinopyroxene, along with varying contributions from orthopyroxene and hornblende. The chemical composition of Cpxs from EBMU exhibits a positive correlation between the Al₂O₃ and TiO₂ contents, matching a fractional crystallization trend (Figure 9f). The Mg# content of clinopyroxenes decreases progressively from 0.92 in pyroxenite to 0.72 in hbl-gabbro (Supplementary Table S4), reflecting continuous fractional crystallization of the parent magma [118]). The chemical compositions of the EBMU’s plagioclase indicate a fractional crystallization trend from bytownitae (An85–90) within olivine gabbro to andesine (An53–63) within hbl-gabbro (Supplementary Table S6; Figure 7g). Similarly, the EBMU amphibole analysis supports a fractional crystallization trend as revealed by the systematic decline in the Mg# values from px-gabbro through px-hbl-gabbro to hbl-gabbro (Supplementary Table S5).

6.4. Parental Magma Estimated Composition

It is widely accepted that the Fe-Mg partition coefficient between olivine and melt remains relatively stable, typically ranging from 0.3 to 0.33, ([Kd_Ol-Melt = (TFeO/MgO)_Ol/(TFeO/MgO_mag)]; [119,120,121], where olivine generally contains less Mg than the original magma, reflecting the interaction between early-formed olivine crystals and the evolving residual melt [122]. The olivine possesses the highest Fo content, which is considered representative of the initial melt, and is used to estimate the composition of the liquidus olivine. The MgO/FeO ratio of the melt in equilibrium with olivine in the EBMU was calculated to range between 1.36 and 1.45, assuming a maximum forsterite content Fo_max of about 81. This estimated ratio is notably higher than the MgO/FeOt values of the whole-rock compositions, which range from 0.8 to 1.86. This discrepancy suggests that additional olivine crystals were introduced into the melt before emplacement. Therefore, the emplaced magma likely represents an evolved melt that has already experienced significant olivine crystallization at depth. In addition, the analyzed olivine crystals from the EBMU pyroxenite show relatively low NiO concentrations (<0.14%; Supplementary Table S3), which are significantly lower than those typically found in olivine formed from primitive basaltic magma (NiO ≈ 0.4%; [105]). This depletion suggests that the EBMU magma is not primitive but rather an evolved melt derived from a parent magma that had already undergone substantial differentiation, leading to a notable depletion in NiO content. Consequently, the composition of the EBMU parental magma (

Table 1. Estimated chemical compositions of El Bola mafic–ultramafic parental and primitive magmas.

	SiO2	MgO	Al2O3	FeO	CaO	Na2O	K2O	MnO	TiO2	P2O5
EBMU parental magma	43.44	19.86	9.36	14.22	8.84	1.34	0.13	0.18	1.66	0.021
EBMU primitive magma	43.04	20.56	8.46	14.61	8.19	1.28	0.07	0.16	1.35
Composition of Ferropicritic magma	43–46	14–20	8–10	12–16	8–11	0.5–1.5	0.1–0.6	0.2–0.3	1.0–2.5	0.1–0.3

Major elements are present in wt.%. EBMU parental magma values were calculated according to the method of [123]; EBMU primitive magma values are representative of an olivine gabbro sample (R4/26); data for the composition of Ferropicritic Magma are from [85].

) was estimated according to the proposed method of Li and Ripley [123], which integrates olivine chemistry, whole-rock geochemical data, and the principle of the olivine–melt equilibrium. The olivine crystal with the highest proportion of Fo (Sample R4/15; Supplementary Table S3) was considered as the most primitive, while the average composition of five representative primitive olivine gabbro samples (R4/22, R4/25, R4/26, R4/28, and R4/29; Supplementary Table S8) was used for the calculation.

The parental magma of the EBMU is characterized by high concentrations of SiO₂ (43.44 wt.%), FeO (14.22 wt.%), TiO₂ (1.66 wt.%), and MgO (20.56 wt.%; Table 1). These values suggest that the magma was of a picritic basaltic nature, particularly given its elevated MgO content (>12%). This interpretation is further supported by Figure 8a, where most olivine gabbro samples are plotted within the picrite field and align with high-Mg tholeiitic basalts (Figure 9g). Cpxs represent the main constituent that is found in all EBMU rocks (Figure 4); where their chemical compositions (Supplementary Table S4) are comparable with those of ferropicrite (Figure 9h). Additionally, the estimated composition of the parental magma of the EBMU intrusion appears to be Fe-Ti-rich (Table 1), which obviously is supported by a strong positive correlation between FeO and TiO₂ (Supplementary Table S8), suggesting that the melt was enriched in both elements (typical of a ferropicritic composition; ([9,19,24,26,27])). Furthermore, the EBMU rock units show higher concentration levels of LREE over HREE (Figure 8f,g), implying that these ferropicritic melts originated from a low-degree partial melting of a mantle plume source at high pressure [124].

The occurrence of amphiboles (Figure 4) within all the EBMU mafic rocks implies a subsequent rise in the water content of the magma likely following the crystallization of early anhydrous phases [125]. Moreover, the analyzed plagioclase of the EBMU ranging from calcic andesine to anorthite (Figure 7g) indicate that their parental melt was hydrous and rich in Al and poor in alkalis [126]. The EBMU rocks are also notably enriched in fluid-mobile elements such as Ba, Sr, and Pb (Figure 8c,d), further reinforcing the interpretation that their ferropicritic element was influenced by water-rich metasomatic processes.

The estimated chemical composition of the EBMU parental magma (Table 1) is comparable to the composition of the studied olivine gabbro (R4/26; Supplementary Table S8) and similar to primary ferropicritic melts [85,127]. This conclusion is consistent with most of the ANS post-orogenic mafic–ultramafic layered intrusions such as Motaghairat [4]; Korab Kansi [15]; and Imleih [45], where their magma are suggested to be ferropicritic and with the interaction between the metasomatized lithospheric mantle and the ascending asthenospheric melts—or possibly a mantle diapir—resulting in hybrid magmas that crystallized to produce the observed layered assemblages. Moreover, the estimated MgO > 12% for the EBMU implies a high degree of partial melting of the mantle source [128].

6.5. Mantle Source Geochemical Characteristics

Mafic–ultramafic intrusions originate from various mantle sources, including the lithospheric mantle, asthenosphere, mantle plumes, or through interactions among these sources [129,130]).

Trace element signatures of the EBMU intrusion display pronounced Nb and Ta depletions, enrichment of LILEs, and elevated LILE/HFSE values (Figure 8d) along with mild LREE/HREE fractionation (Figure 8f). These geochemical features point to a mantle-driven origin for the EBMU [131]. The neglected Ca proportions in the EBMU olivine rocks (0.01–0.04; Supplementary Table S3) imply that their primitive magma originated from a metasomatized mantle source that was merely affected by fluids from subducted slab [132]. Moreover, it is widely accepted that the generation of magma from partial melting of the metasomatized mantle source is characterized by a LILE/HFSE ratio for which the significant increment of LILEs is typically attributed to subduction-related processes, where hydrous fluids released from the descending slab preferentially transport these elements into the overlying mantle wedge [88,133,134]. The significantly elevated concentrations of LILEs (e.g., Rb, Ba, and Sr) coupled with subdued concentrations of HFSEs (e.g., Nb, Ta and Zr) within the EBMU intrusion (Figure 8d,g) revealed that the EBMU intrusion was derived from a metasomatized mantle source and crystallized from tholeiitic basaltic magmas, which display geochemical characteristics typical of a subduction-related arc setting [133,135].

The EBMU displayed U/Th >1, implying the fractionation of U from Th, where U is preferentially mobilized relative to Th during subduction, likely through dehydration of the subducting slab, and was subsequently introduced into the overlying mantle via slab-derived hydrous fluids [133,134]. The EBMU intrusion is marked by low Th/Ta and Ta/Hf ratios, which are prevalent in magmas derived from a depleted mantle source (DMM; Figure 10a; [136]). The EBMU chemical data exhibits relative values of Dy/Dy* and Dy/Yb (Figure 10b), where the data are plotted in the MORB-type magma plot, whereas the elevated LREE level implies derivation from an E-MORB-like asthenospheric mantle source. Moreover, the EBMU is characterized by high Nb/Ta (9.65–14.35) and Zr/Hf (11.59–41.81) and high sub-chondritic Nb/U (6.86–12.68) which point to mantle source depletion by former melt extraction [137]. Furthermore, pronounced high concentrations of Pb coupled with depletion of HFSEs (e.g., Nb, Th, Zr, and Hf; Figure 8d) reflect a high degree of modification to the mantle source by slab-derived fluid metasomatism [15,16,138]. Although the EBMU rocks are HFSE-depleted, they possess HFSE concentrations that are more consistent with OIB and are notably elevated relative to those of E-MORB (Figure 8d), reflecting subsequent inputs with such elements to their magma sources. As we concluded previously, the EBMU intrusion’s parent magma was least affected by crustal contamination; thus, it may possibly be influenced by the subcontinental lithospheric mantle (SCLM) instead of continental crust. Notably, island-arc calc-alkaline basalts (IACABs) typically display reduce ratios of Ce/Pb (~3 in IACAB) [139] compared to ocean island basalts (OIBs; Ce/Pb = 25) [93]. The plots of the EBMU samples near or above the MORB-OIB array (Figure 10c) support the modification of the EBMU mantle source with a subducted crustal component [140]. The EBMU rocks have relatively high Ce/Pb (3.66–16.36) ratios that correspond to those of OIBs rather than IACABs. Accordingly, the pronounced Pb enrichment (Figure 8d) within the EBMU could have resulted from the dehydration of the subducted slab-released fluids that extracted Pb from the oceanic crust and transferred it into the mantle wedge, producing Ce and Pb fractionation and enriching Pb in the mantle and its derived magmas. [141]. The high ratios of Ba/Nb (36.08–215.35), Ta/U (1.17–2.11), and Nb/U (12.01–22.19) coupled with low ratios of Nb/La (0.06–0.24) and Ce/Pb (3.66–16.36) are indicative of the infiltration of subduction-released fluids to the mantle source of the EBMU intrusion [93,115,142,143]. Moreover, the high La/Nb (4.11–16.90) and Ba/La (3.01–23.62) ratios confirm subduction contributions [144].

The EBMU intrusion generally possesses high values of U/Th ratios indicating that the metasomatism of the upper mantle was primarily driven by an increased input of hydrous fluids rather than melts from the subducted slab or associated sediments [133]. Moreover, Rb and U are mobile in the fluid component, whereas Zr, Nb, and Y are more stable [25]. Moreover, rocks derived from subduction zones typically exhibit elevated Rb/Y, Rb/Nb, and Ba/Nb ratios due to the enrichment of Rb from slab-derived fluids, while within-plate basalts often show higher Nb/Y and Th/Nb ratios, reflecting a mantle source less influenced by subduction processes [145].

Figure 10. Mantle source and magmatic evolution of the EBMU intrusion: (a) Ta/Hf vs. Th/Ta diagram [136] for the studied EBMU rocks. Depleted MOR Mantle (DMM) data from [146], Upper Crust (UC) from [111], and ocean island basalt (OIB) from [93]; (b) plots of the EBMU rock chemical data on Dy/Dy* versus Dy/Yb diagram from [147]; (c) plots of EBMU rock chemical data on Nb/Yb vs. Th/Yb diagram from [140]. Data of ANS post-orogenic layered mafic–ultramafic intrusions compiled from [4,16,19,33,45]. (d–f) plots of the EBMU chemical data on (d) Nb/Y versus Rb/Y diagram [145]; (e) Rb/Nb versus Th/Nb diagram [148] and (f) Ba/Nb versus Th/Nb diagram [143]; (g) Gd/Yb versus Dy/Yb [130]; (h) La/Yb versus Sm/Yb [149].

Figure 10. Mantle source and magmatic evolution of the EBMU intrusion: (a) Ta/Hf vs. Th/Ta diagram [136] for the studied EBMU rocks. Depleted MOR Mantle (DMM) data from [146], Upper Crust (UC) from [111], and ocean island basalt (OIB) from [93]; (b) plots of the EBMU rock chemical data on Dy/Dy* versus Dy/Yb diagram from [147]; (c) plots of EBMU rock chemical data on Nb/Yb vs. Th/Yb diagram from [140]. Data of ANS post-orogenic layered mafic–ultramafic intrusions compiled from [4,16,19,33,45]. (d–f) plots of the EBMU chemical data on (d) Nb/Y versus Rb/Y diagram [145]; (e) Rb/Nb versus Th/Nb diagram [148] and (f) Ba/Nb versus Th/Nb diagram [143]; (g) Gd/Yb versus Dy/Yb [130]; (h) La/Yb versus Sm/Yb [149].

The geochemical signatures of the EBMU intrusion, with particularly elevated Rb/Y, Rb/Nb, and Ba/Nb ratios compared to Th/Nb and Nb/Y, demonstrate enrichment controlled primarily by hydrous fluid input, as opposed to melts from the subducted slab or associated sediments (Figure 10d–f). Furthermore, the EBMU shows Nb/Zr > Th/Zr, supporting derivation from a depleted mantle source [150]. Accordingly, the formation of the primitive magmas in the EBMU intrusion was strongly influenced by contributions from the upper mantle, which was metasomatized by hydrous fluids released during slab dehydration. This interaction imparted arc-like geochemical signatures to the magmas, reflecting the critical role of both mantle sources and slab-derived fluids in their genesis. The mantle source shows distinct negative Nb and Ta anomalies, indicating it was likely metasomatized by slab-derived fluids during dehydration processes in a subduction zone environment.

REEs’ abundance and their ratios are valuable indicators for estimating the melting depths of the EBMU rock sources. Typically, partial melting of a garnet-bearing lherzolite source yields melts with reduced Yb concentrations and elevated ratios of Sm/Yb, in contrast to melts derived from garnet-free mantle sources [151,152].

The analyzed samples from the EBMU intrusion have low Dy/Yb and Gd/Yb ratios that correspond to the spinel peridotite melting curve in Figure 10g,h. This conclusion implies that the partial melt occurs at depths from 60 to 80 Km and extends to the spinel-garnet transition zone [153], which could be attributed to thinning of the lithospheric mantle and/or slab-break off enabling the hot asthenosphere to flow upward through the slab window. This causes a thermal anomaly that triggers decompression melting of the asthenospheric mantle along with melting of the metasomatized mantle layer above [154,155].

6.6. Tectonic Setting

The ANS mafic–ultramafic rocks are typically associated with two main settings: older, metamorphosed ophiolitic complexes that exhibit tectonic boundaries with surrounding rocks, and younger, relatively unmetamorphosed intrusions that often resemble Alaskan-type or layered intrusions. The EMBU rocks are not related to ophiolitic complexes as they exhibit a nearly unmetamorphosed nature and lack tectonic contact with their country rocks (Figure 1); moreover, the highest value of Fo in olivine from the EBMU is about 80 (Supplementary Table S3) which is much lower than the values from ANS ophiolite (>88; [10,20,44,49,156]). Meanwhile, the distribution of the EBMU rocks lacks concentric zonation which is a significant feature of many ANS Alaskan-type intrusions e.g., [14,62]). In addition, chemical data of the EBMU minerals such as olivine, chromite, and pyroxenes from the EBMU intrusion are more consistent with layered mafic–ultramafic intrusion rather than Alaskan-type intrusion (Figure 5 and Figure 6). Furthermore, the Cpxs compositions of the EBMU mafic rocks closely align with those typically found in orogenic and island-arc tholeiitic (IAT) intrusions (Figure 11a,b). They also plot within the volcanic arc or island-arc basalt fields (Figure 11c) and exhibit a geochemical trend consistent with arc-related cumulates (Figure 6e). Meanwhile, the analyzed Cpxs from the EBMU rocks exhibit both calc-alkaline and tholeiitic affinities (Figure 9f), indicating an island-arc setting [157]. Furthermore, the Mg#s of the analysed Cpxs from the EBMU intrusion ranging from 0.75 to 0.92 (Supplementary Table S4), which are analogous to those from Motaghairat layered intrusion (0.71–0.29), which developed in a subduction environment and are akin to the post-collisional extension regime [4], whereas the compositions of Opxs from the EBMU are comparable to those from Korab Kansi layered mafic–ultramafic rocks (Figure 6; [15]) that were originated from hydrous ferropicritic melts of a tholeiitic nature that were modified by fluids related to subduction into calc-alkaline magmas that evolved through experienced fractional crystallization during the maturation of an island-arc system.

Although some of the EBMU rocks have transitional characteristics between the calc-alkaline and tholeiitic affinities, they all fall within the compositional field of arc-related mafic and ultramafic cumulates. EBMU mafic rocks, in particular, show similarities to ANS post-collisional gabbroic rocks, which are believed to have formed from arc-like tholeiitic magmas in an extensional tectonic setting Figure 8b; e.g., [15,16,33,34]. Moreover, the EBMU rocks display high concentration levels of LILEs (e.g., Rb, Ba, Sr, and Pb) with significantly subdued amounts of Nb and Ta (Figure 8e) and significantly elevated LILE/HFSE ratios (e.g., Sr/Zr, Ba/Nb, andRb/Th; Supplementary Table S8), which are the typical features of post-collisional magmatism, and they exhibit subduction-like geochemical signatures inherited from earlier subduction events through the development of the ANS [1,4,162,163]. Moreover, the studied olivine gabbro’s chemical composition approximates the EBMU parental magma which is generally enriched with Mg, Fe, Ti, and V. These characteristics are comparable to parental melts of the ANS layered associations that are believed to have originated from high-Mg-Ti tholeiitic magmas that experienced extensive fractional crystallization [4,15,19].

Many studies have connected the formation of mafic–ultramafic rocks to slab break-off events occurring during or after continental collisions [26,129,130,135,164]. Additionally, some researchers have suggested a mantle plume model, proposing that these rocks formed in a continental rift environment driven by upwelling mantle plumes [165]. The mantle plume mechanism was suggested as a major activity during ANS formation [48,156]. The EBMU possess trace element concentrations that correspond to those of OIBs (Figure 8). Hofmann [115] concluded that magmatic rocks derived from mantle plumes typically exhibit elemental compositions resembling those of OIBs. Therefore, we acknowledge the mantle plume origin for the EBMU intrusion, for which, subsequent to the collision of accreted island arcs with the Saharan Metacraton, slab break-off and lithospheric delamination led to post-collisional extension during which these rocks were emplaced; this model (Figure 12) was assigned for most of the ANS layered mafic–ultramafic intrusions [4,15,16].

The petrogenetic model for the ANS post-orogenic layered mafic–ultramafic intrusion was introduced by [16], which could be summarized into three stages as follow: (1) The subduction stages, where the accreted island arcs and the Saharan Metacraton collided before 610 Ma [166], after which post-collisional extensional magmatism ensued [167,168]. Moreover, U–Pb zircon dating across the ANS reveals a shift from subduction-related (I-type) magmatism (c. 780–630 Ma) to post-collisional, extensional/hot asthenosphere-influenced magmatism (c. 635–590 Ma) [86]. In addition, the slab detachment took place 10 to 20 Ma after continental collision (~630 Ma), introducing heat from the asthenospheric mantle and melting the previously metasomatized lithosphere and generating slab-break-off magmatism c. 630–600 Ma [169]. Consequently, as the subduction process proceeded, the SCLM was altered through metasomatism by fluids released from the subducted oceanic slabs (Figure 12a). (2) Through slab break-off and asthenosphere upwelling, a portion of the slab break-offs took place after subduction and sunk into the asthenosphere and were destabilized, and active upwelling occurred through the slab window (Figure 12b), which may haven been followed by lithospheric delamination from thicker, dense lithosphere where the asthenosphere rose (mantle plume; [170]). (3) Emplacement of post-orogenic mafic–ultramafic intrusions: the plumed mantle interacted with the base of the metasomatized SCLM, producing thermal degradation and partial melting of the upper mantle (Figure 12c), leading to mantle-derived melts. Mantle plumes underwent adiabatic decompression melting, which produces primary melts enriched in Mg, Fe, V, and Ti [171]. These initial mantle-derived melts contributed small volumes of magma to the base of the lithosphere. These melts, which represent the parental magma of the EBMU high-Mg and Ti-rich gabbros, ascended into the lower crust and triggered post-collisional mafic magmatism.

7. Conclusions

The El Bola mafic–ultramafic intrusion (EBMU) represents a well-defined rhythmic layering intrusion composed of pyroxenite and gabbroic rocks varying from olivine to hornblende gabbros with subordinate anorthosite. Geochemical and mineralogical data revealed that the EBMU represents a complex magmatic evolution, primarily governed by extensive fractional crystallization of a high-Mg and Fe- and Ti-enriched ferropicritic parental magma. These data strongly support the limited role of crustal contamination during the magma’s ascent and emplacement. Thermobarometric estimations indicate a progressive decrease in crystallization temperatures and pressures, consistent with magmatic differentiation within a layered intrusion environment. Redox conditions, as inferred from the chemical composition of minerals, show a trend of increasing oxygen fugacity during the crystallization sequence. This suggests the involvement of oxidizing fluids, most likely derived from a subducted slab, which metasomatized the mantle source prior to melt generation. The parental magma appears to have originated from a hydrous, metasomatized spinel lherzolite mantle source, modified by subduction-related components. Trace element characteristics support this interpretation, showing signatures typical of fluids from a subduction-modified mantle wedge. The geodynamic setting of the EBMU intrusion aligns with a post-collisional extensional regime, most likely triggered by asthenospheric upwelling and slab break-off events. These processes facilitated the ascent of primitive, mantle-derived melts, which crystallized to form the observed mafic and ultramafic rocks. Collectively, the data suggest that the EBMU intrusion represents a magmatic response to late-stage tectonic processes associated with the post-collisional evolution of the Arabian–Nubian Shield.