Isotopic Disequilibrium Between Migmatites and Protolith: Insights from a Variscan Anatectic Complex (NW of Iberian Variscan Belt, Portugal)

Abstract

Isotopic disequilibrium during the formation of high-temperature (HT) metamorphic complexes by anatexis during continental collision is a process that deserves intense discussion since it is fundamental to understand the evolution of continental crust. The axial sector of the Iberian Variscan Belt (IVB) is known by the profusion of synorogenic granites that are sometimes clearly associated with the migmatites composing the HT metamorphic complexes. The Pedregal Migmatitic Complex is located in the autochthonous domain of the IVB and is composed of metatexites and diatexites associated to syntectonic two-mica granites. The anatectic process occurred by dehydration melting of muscovite and biotite with the growth of peritectic minerals such as garnet, K-feldspar, and sillimanite in metatexites; and K-feldspar, sillimanite, and hercynite in diatexites reaching the metamorphic peak at 313.5 ± 0.5 Ma. A process of residuum-melt separation during crustal melting is attested by the Pedregal migmatites, giving origin to metatexites and residual diatexites as indicated by field evidence and their geochemical signature. Zircon oxygen isotopes and inherited zircon ages point to the Douro-Beiras Supergroup metasedimentary sequence (Beiras group) as a possible protolith of the Pedregal diatexites. Conversely, the isotopic composition of the diatexites suggests isotopic disequilibrium caused by residual mineral phases (biotite, monazite and garnet).

1. Introduction

Continental collision during an orogenic cycle leads to crustal thickening which repeatedly suffers episodes of intense deformation, high-temperature metamorphism, and crustal anatexis (e.g., Variscan and Alpine orogenies). Therefore, orogenesis and crustal anatexis are closely associated conducting to the arise of high-grade areas of many metamorphic belts and metamorphic core complexes in large areas of exposed deep continental crust, but also in low-pressure–high-temperature (LP-HT) metamorphic areas such as the inner parts of contact metamorphic aureoles [1]. In many cases, high-temperature metamorphic complexes are petrogenetically related to granites generated from large volumes of anatectic magma produced during the late stages of orogenic evolution [2,3].

Anatexis involves partial melting of crustal rocks and produces high-temperature metamorphic rocks known as migmatites. Partial melting during high-grade metamorphism typically begins in metapelitic or metapsammitic protoliths at temperatures of ~650–700 °C in the presence of H₂O (fluid-present melting reactions). With increasing temperature during prograde metamorphism, free H₂O becomes limited, and partial melting proceeds through incongruent fluid-absent reactions involving dehydration melting of hydrous minerals such as muscovite, biotite, and hornblende [4,5]. Migmatites are commonly classified as metatexites or diatexites based on morphology, temperature conditions, degree of partial melting, and melt fraction. Metatexites represent lower degrees of partial melting and are characterized by the dominance of paleosome and mesosome (unmelted material) relative to neosome (newly formed material), resulting in compositions close to the protolith. In contrast, diatexites form at higher temperatures and melt fractions, where neosome dominates and the rock develops a more homogeneous, igneous-like texture [5,6].

During the late stage of the Variscan orogenic cycle, several HT terranes exhumed along the Variscan Belt in different places such as the Appalachian terrains [4], the Armorican massif [5], the French Massif Central [6,7], the Bohemian massif [8], and in the Iberian massif [9,10,11,12,13,14,15]. In the Iberian Variscan Belt (IVB), particularly within the northern autochthonous domain, several anatectic complexes are characterized by the association of gneiss–migmatite rocks and Variscan S-type granites. These complexes are interpreted to result from partial melting of the autochthonous metasedimentary sequence that hosts them (i.e., the Douro-Beiras Supergroup), from which they inherit their radiogenic isotopic composition. Consequently, anatectic rocks are expected to display isotopic compositions similar to those of their protoliths [14,16,17]. However, isotopic variability can be observed among the rocks of an anatectic complex and their protoliths and this may reflect melting disequilibrium related to the mineral phases involved in melting reactions [18,19,20,21]. In the IVB, Barbero et al. (1995) [22] first proposed isotopic disequilibrium during crustal melting via fluid–absent reactions for the anatectic granitoids of the Variscan Anatectic Complex of Toledo (Spain).

The Pedregal Migmatitic Complex is one of the anatectic complexes within the high-temperature metamorphic domains of the IVB. The main objective of this study is to characterize this complex using integrated petrological (field relationships and macro-/microscopic features), geochemical (whole-rock and rutile geochemistry; oxygen isotopes in zircon), and geochronological (U–Pb zircon ages) data. These results aim to constrain the petrogenesis of the complex within the IVB framework. Additionally, this study investigates whether isotopic disequilibrium during partial melting explains the Sr–Nd isotopic variability observed between the Pedregal migmatites and their protoliths and evaluates the role of melting reactions and mineral phases on generating this variability.

2. Geological Setting, Lithological Descriptions, and Field Relationships

The Iberian massif is the westernmost European terrain that exposes the Variscan Orogen. In the Central lberian Zone (CIZ), three main phases of deformation were recognized (D₁, D₂ and D₃). The early-D₃ phase in the CIZ is responsible for the beginning of the migmatization and LP-HT metamorphism, which continued during the D₃ phase and culminated with the emplacement of S-type syntectonic granites [23,24,25]. The innermost part of the Iberian massif, the CIZ (Figure 1a), exposes the Douro–Beiras Supergroup, which is a metapelitic sequence that represents a major part of the autochthonous domain with an Ediacarian–Cambrian age [26]. The Douro–Beiras Supergroup is divided into the Douro and Beiras groups, which occupy the northern and southern parts of the CIZ, respectively. Both groups comprise stratigraphic sequences with an alternation of metapelites and greywackes, but the presence of calc–silicate rocks in the Douro Group is more frequent than in the Beiras Group [27]).

The Porto–Viseu metamorphic belt is a Variscan structure located in the CIZ, limited to the NNE and SSW by the Douro–Beiras and the Porto–Tomar shear zones, respectively [28]. This metamorphic belt is characterized by the presence of syntectonic granites in the core associated with the LP-HT migmatites, and a condensed metamorphic zoning with a syn-D3 thermal peak showing an orogenic metamorphism with a strong thermal anomaly [29].

At NW of the Porto–Viseu belt, specifically in the eastern border of the Porto metropolitan area, occurs the Pedregal Migmatitic Complex (PMC) (Figure 1b). The PMC comprises a small, elongated diatexitic body (appoximately 3 km²) [13] and metatexitic rocks hosted by staurolite micaschists belonging to the Douro–Beiras Supergroup [30]. Synorogenic granites are spatially associated with the migmatitic complex. These include the Porto granite, a medium-grained two-mica granite with a zircon/monazite age of 306 ± 7 Ma [31]; the Gondomar granite, a coarse-grained tourmaline-rich granite; and the Fânzeres granite, a foliated garnet-rich granite with a Rb-Sr age of 332 ± 11 Ma [32].

The Pedregal diatexites have a homogeneous, igneous appearance and a fine-to-medium-grained texture. They contain small biotitic nodules/schlieren structures (1–2 cm), which exhibit internal foliation ranging in direction from NE–SW to E–W (Figure 2a).

This trend is opposite to regional NW-SE metamorphic structure of the staurolitic micaschists and to the elongation of the main diatexitic body. The diatexites also expose inclusions of metasedimentary enclaves.

In the border of the diatexitic outcrops appear banded gneiss–migmatite rocks with metatexitic aspect. These banded rocks show a stromatic texture predominantly defined by unfoliated quartz–feldspathic leucosomes alternating with pre-partial melting structures composed by micaceous paleosomes (Figure 2b). The orientation of this compositional layering is N100° to N130°, subvertical in concordance with diatexitic elongation and the foliation of the host micaschists. The metatexites show crystals of garnet intergrowth with quartz in the quartz–feldspathic leucosomes (Figure 2c).

The contact between the two types of migmatite is sharp (Figure 2d), as the transition from metatexite to diatexite is abrupt. Locally, this contact is marked by breccia-like structures containing clasts of metatexite within a diatexitic matrix. These structures are also known as agmatite (Figure 2e).

3. Petrography

3.1. Diatexite

The Pedregal diatexite exhibits a granoblastic texture (Figure 3a) where the large crystals of muscovite stand out. The mineral assemblage of the diatexite is quartz + plagioclase + K-feldspar + biotite + muscovite + zircon + apatite + monazite + rutile ± sillimanite ± allanite ± Zn-rich hercynite.

The elongated and corroded biotite crystals define a preferential alignment, as evidenced by their shape anisotropy and homogeneous distribution (Figure 3b). The biotite–quartz and biotite–plagioclase intergranular boundaries indicate textural disequilibrium, suggesting that biotite may be a residual phase. The subhedral biotite crystals are often chloritized and contain small zircon and rutile needle-like inclusions. Small grains of unaltered biotite are included in subhedral muscovite. In the small nodules, the biotite is subhedral and less corroded than the elongated crystals. These nodules exhibit compositional banding characterised by the presence of biotite and secondary muscovite.

The quartz is subhedral to anhedral, and when included in other minerals such as muscovite and plagioclase, it takes the form of droplets. Quartz crystals frequently show inclusions of sillimanite and rutile needles. Besides the rutile needles, rutile also occurs in the matrix as isolated grains with euhedral to subhedral shape.

Regarding the feldspars, K-feldspar has greater dimensions, but plagioclase is more abundant. Plagioclase crystals are subhedral, and their polysynthetic twinning is usually obliterated by muscovite (muscovitization).

The muscovite has a secondary/retrograde character (Figure 3c), distinguishing two types: a subhedral muscovite, which shows irregular borders and quartz droplet inclusions, and a euhedral muscovite, with anomalous birefringence. Both types of muscovite have zircon, sillimanite, and hercynite inclusions (Figure 3d).

3.2. Metatexite

Metatexitic rocks show a granolepidoblastic and heterogranular medium-grained texture (Figure 4a). The mineral assemblage consists of quartz + plagioclase + K-feldspar + biotite + muscovite + garnet. The biotite and muscovite define an alignment and a preferential orientation. In these rocks are also visible a compositional banding defined by the scarce micaceous bands (melanosome) and the predominant quartz–feldspathic bands, where the quartz–K-feldspar–plagioclase association is observed (leucosome).

Two generations of subhedral biotite are founded in the metatexites (Figure 4b), a first generation that shows a light brown colour and a second generation with reddish brown colour, typically an igneous biotite crystallized from residual melt. Sometimes, the biotite crystals show inclusions of rutile needles and a slight chloritization.

The subhedral muscovite grow occurs conditioned by the other minerals showing its secondary/retrograde character and frequently having inclusions of sillimanite (Figure 4c).

In general, the quartz grains are subhedral, but when they are included in the plagioclase and K-feldspar, they show a rounded shape. The garnet occurs occasionally in large crystals intergrowing with quartz (Figure 4d).

4. Analythical Methods

4.1. Whole-Rock Geochemistry

The whole-rock geochemistry analyses were conducted in the Activation Lab (Activation Laboratories Ltd., Ancaster, ON, Canada) in order to determine the major and trace element composition. The selected analytical package was 4Litho (11+) Major Elements Fusion ICP (WRA)/Trace Elements Fusion ICP/MS (WRA4B2). This package include the following trace elements analyses: Sc, Be, V, Ba, Sr, Y, Zr, Cr, Co, Ni, Cu, Zn, Ga, Ge, As, Rb, Nb, Mo, Ag, In, Sn, Sb, Cs, La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, Hf, Ta, W, Tl, Pb, Bi, Th, U (

Table 1. Whole-rock chemical composition of the Pedregal Migmatitic Complex (DTX—diatexite; MTX—metatexite). Major and minor elements in weight percentage (wt.%) and the trace elements expressed in parts per million (ppm).

Sample	P1	P2	P3	P6	P7	P8	P9	P10
Lithology	DTX	DTX	DTX	DTX	DTX	MTX	MTX	MTX
Longitude	−8°32′37.79″	−8°32′44.53″	−8°32′16.44″	−8°31′51.60″	−8°31′33.98″	−8°32′11.34″	−8°32′11.34″	−8°31′40.77″
Latitude	41°7′10.55″	41°7′18.27″	41°7′23.53″	41°7′5.62″	41°6′56.22″	41°6′33.78″	41°6′33.78″	41°6′12.33″
SiO2	65.28	68.57	69.18	65.30	66.38	72.27	73.79	74.25
Al2O3	15.71	16.12	15.98	16.68	16.46	14.43	14.62	14.57
Fe2O3t	3.68	2.93	2.50	2.59	2.28	1.12	1.21	1.08
MnO	0.04	0.02	0.01	0.02	0.01	0.02	0.02	0.02
MgO	1.00	0.81	0.73	0.80	0.71	0.25	0.25	0.20
CaO	1.58	0.70	0.52	0.65	0.33	0.62	0.62	0.50
FeO	3.35	2.67	2.28	2.36	2.07	1.02	1.10	0.98
Na2O	2.89	2.21	2.01	2.50	2.03	3.25	3.46	3.32
K2O	5.21	5.96	5.69	5.63	5.72	4.46	4.24	4.11
TiO2	0.89	0.75	0.69	0.70	0.68	0.12	0.12	0.10
P2O5	0.43	0.50	0.52	0.55	0.42	0.30	0.31	0.29
LOI	1.74	2.30	2.55	2.79	3.30	1.29	1.50	1.40
Total	98.44	100.90	100.40	98.21	98.33	98.12	100.10	99.84
Ba	1005	677	533	527	527	322	293	258
Sr	228	113	71	94	84	62	56	40
Y	14	12	11	12	9	10	9	8
Zr	435	416	388	398	389	56	57	51
Cr	50	30	60	50	40	<20	<20	<20
Co	6	4	2	3	3	1	1	1
Ni	<20	<20	<20	<20	<20	<20	<20	<20
Cu	<10	<10	<10	<10	<10	<10	<10	<10
Zn	50	160	120	140	110	50	60	40
Ga	26	29	29	28	28	17	17	17
Ge	1	2	1	2	1	2	2	2
As	<5	<5	8	53	<5	<5	<5	<5
Rb	266	344	333	405	385	174	164	176
Nb	9	6	5	5	5	7	7	7
Mo	<2	<2	<2	<2	<2	<2	<2	<2
Ag	3.6	3.1	3.3	3.8	3.4	<0.5	0.6	<0.5
In	<0.2	<0.2	<0.2	<0.2	<0.2	<0.2	<0.2	<0.2
Sn	<1	2	3	5	4	6	6	10
Sb	<0.5	0.7	2.6	1.9	1.2	<0.5	<0.5	<0.5
Cs	5.5	4.9	4.4	63.8	6.1	7.8	7.6	7.3
La	114	130	101	91.6	98.9	9.6	10.4	8.2
Ce	246	320	264	243	260	18.8	20.6	16
Pr	30.7	41.5	37.6	35.3	37.1	2.37	2.51	2.03
Nd	110	156	146	142	148	8.9	9	7.7
Sm	16.2	22.2	22.2	22.3	22.1	2.4	2.5	2.1
Eu	1.39	1.03	0.89	0.89	0.95	0.45	0.43	0.41
Gd	7.1	7.6	7.2	8.7	7.7	2.5	2.7	2
Tb	0.7	0.6	0.6	0.7	0.6	0.4	0.4	0.3
Dy	3.4	3	2.6	2.9	2.5	2.1	2.1	1.5
Ho	0.5	0.4	0.4	0.4	0.4	0.3	0.3	0.2
Er	1.4	1.1	1	1.1	0.9	0.6	0.7	0.6
Tm	0.18	0.14	0.14	0.15	0.12	0.09	0.1	0.09
Yb	1	0.8	0.8	0.8	0.7	0.5	0.5	0.5
Lu	0.15	0.1	0.12	0.13	0.1	0.07	0.07	0.06
Hf	10.8	10.5	10.3	10	9.7	1.7	1.7	1.5
Ta	0.6	0.4	0.3	0.4	0.3	0.8	0.9	1.3
W	<1	2	3	<1	<1	1	<1	2
Tl	1.5	2.1	2.1	2.6	2.3	1	0.9	1
Pb	57	47	56	39	32	31	34	28
Bi	<0.4	<0.4	<0.4	<0.4	<0.4	1	0.7	1
Th	72.3	133	137	130	135	4.2	4.6	3.5
U	8.8	13.4	12	9.9	12.8	4.5	4.7	5.4

). Lithogeochemical analyses required lithium metaborate/tetraborate fusion, followed by inductively coupled plasma mass spectrometry (FUS-ICP-MS). Fused sample is diluted and analyzed by Perkin Elmer Sciex ELAN 6000, 6100, or 9000 ICP/MS. Three blanks and five controls (three before sample group and two after) are analyzed per group of samples. Duplicates are fused and analyzed every 15 samples. Instrument is recalibrated every 40 samples (http://www.actlabs.com/). The geochemical diagrams containing whole-rock geochemical data were created using the open-source software GCDkit (version 4.1.3) [33].

4.2. Isotopic (Sr-Nd) Geochemistry

The isotopic Sr-Nd signatures were determined in the Laboratory of Isotope Geology of the University of Aveiro (Portugal) for five samples of Pedregal diatexite and three samples of metatexite. After the sample dissolution, the target elements were purified using ion chromatography at two stages: (1) separation of Sr and REE on a cation exchange column with AG8 50 W Bio-Rad resin and (2) purification of Nd relative to the other lanthanides with Ln Resin (EiChrom Technologies) columns. The values of ⁸⁷Sr/⁸⁶Sr and ¹⁴³Nd/¹⁴⁴Nd were obtained using thermal ionization mass spectrometer VG Sector 54. The isotopic ratios were corrected for mass fractionation considering ⁸⁸Sr/⁸⁶Sr = 0.1194 and ¹⁴⁶Nd/¹⁴⁴Nd = 0.7219. During this study, the SRM 987 and the JNdi-1 standards gave average values of ⁸⁷Sr/⁸⁶Sr = 0.710263(16) (N = 14; conf.lim = 95%) and ¹⁴³Nd/¹⁴⁴Nd = 0.5121030(68) (N = 13; conf.lim = 95%). The Rb and Sr concentrations of the samples with higher Rb/Sr ratios were determined by isotopic dilution using a spike of ⁸⁷Rb/⁸⁴Sr.

4.3. Rutile Geochemistry

Twenty-five rutile grains of a Pedregal diatexite sample (P2) were analysed on a mount to obtain trace elements compositions.

Selected capsules in which large areas of crystal-free glass were observed were used for in situ trace element determinations by laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) at the University of Granada (Spain). LA-ICP-MS analyses were performed with a 213 nm Mercantek Nd-YAG laser coupled to an Agilent 7500 ICP-MS system with a shielded plasma torch, using the NIST-610 glass as standard. The ablation was carried out in a He atmosphere. The laser beam was fixed on a 95 μm-wide square section. The spot was pre-ablated for 45 s using a laser repetition rate of 10 Hz and 40% output energy. Then, the spot was ablated for 60 s at 10 Hz with laser output energy of 75%; see [34] for analytical details on standards.

4.4. U-Pb Geochronology

One sample of the Pedregal diatexite was selected for U-Pb zircon geochronology (P2 sample). The U-Pb analyses were done using a Sensitive High Resolution Ion Microprobe (SHRIMP IIe/mc) at IBERSIMS lab (University of Granada). Hand-picked zircons from the studied sample, several grains of the TEMORA and one grain of the SL13 zircons standards, plus a few grains of the GAL zircon, were cast on a 3.5 cm-diameter epoxy mount (megamount), polished and documented using optical (reflected and transmitted light) and scanning electron microscopy (secondary electrons and cathodoluminescence). After extensive cleaning, mounts were coated with gold (80 microns thickness) and inserted into the SHRIMP for analysis. Each selected spot was rastered with the primary beam during 120 s prior to the analysis and then analyzed during six scans following the isotope peak sequence ¹⁹⁶Zr₂O, ²⁰⁴Pb, ^204.1 background, ²⁰⁶Pb, ²⁰⁷Pb, ²⁰⁸Pb, ²³⁸U, ²⁴⁸ThO, ²⁵⁴UO. Every peak of every scan was measured sequentially 10 times with the following total counting times per scan: 2 s for mass 196; 5 s for masses 238, 248, and 254; 15 s for masses 204, 206, and 208; and 20 s for mass 207. The primary beam, composed of ¹⁶O¹⁶O₂⁺, is set to an intensity of 4 to 5 pA, with a Kohler aperture of 120 microns, which generates 17 × 20 micron elliptical spots on the target. The secondary beam exit slit was fixed at 80 microns, achieving a resolution of about 5000 at 1% peak height. All calibration procedures were done on the standards included on the same mount. Mass calibration was done on the GAL zircon (c.a 480 Ma, very high U, Th, and common lead content) [35]. Analytical sessions started measuring the SL13 zircon [36], which was used for isotope concentration standard (238 ppm U). The TEMORA zircon (417 Ma, [37]), used as isotope ratios standard, is then measured every four unknowns. Data reduction is done with the SHRIMPTOOLS software (available from www.ugr.es/~fbea), specifically developed for IBERSIMS by F. Bea. Errors (95% confidence level) are calculated as the standard error of the linear prediction at the mid-point of the analysis. ²⁰⁶Pb/²³⁸U is calculated from the measured ²⁰⁶Pb+/²³⁸U⁺ and UO⁺/U⁺ following the method described by [38]. For high-U zircons (U > 2500 ppm), ²⁰⁶Pb/²³⁸U is further corrected using the algorithm of [39]. Though seldom necessary, the software also permits corrections for instrumental drift with time using the sequence of replicate measurements of the TEMORA zircon. These are also used for calculating the point-to-point error for the analytical session, to which the error for the final age calculations must be expanded.

4.5. Zircon Oxygen Isotopes

Following U-Pb analysis, the mounts were cleaned, re-polished, and coated with a 30 nm-thick gold layer for oxygen isotope analyses. For this purpose, the SHRIMP primary ion optics were set with a 120 µm Kohler aperture to produce a ~18 µm diameter spot on the mount surface. The Cs gun was set to yield a ~8 nA Cs+ beam. The e-gun to neutralize Cs ions on nonconductive materials was set to an intensity of about 1 µA. The spots to be analyzed were burned for around 5 min before measurements. During this time, the secondary beam and the e-gun were fully optimised to maximise the ¹⁶O signal. Measurements were performed in 2 sets of 10 scans each. Each scan lasted 10 s, meaning the real data collection time was 200 s per spot. The EISIE (electron induced secondary ion emission) background was recorded for 10 s before and after each set and was subtracted from the ¹⁸O and ¹⁶O counts. We used the TEMORA-II zircon as a standard, measuring it every 3 unknowns and cross-checking the 91500 zircon again every 20 unknowns. The reproducibility of the standards was excellent: ∂¹⁸O = 8.22 ± 0.08 (2 s) for the TEMORA and ∂¹⁸O = 9.98 ± 0.26 (2 s), respectively. Data reduction was performed using the POXY program, which was developed by P. Lanc and P. Holden at the Australian National University.

5. Results

5.1. Whole-Rock Geochemistry

The diatexites are peraluminous, with an A/CNK parameter ranging from 1.18 to 1.62 and an ASI ranging from 1.23 to 1.71. They have a magnesian and alkali to alkali–calcic signature (Figure 5).

However, the diatexites peraluminosity is probably influenced by alteration effects, since the LOI is higher than 2 wt.% [41].

The major elements composition of the diatexites (Figure 6; Table 1) shows concentrations of SiO₂ ranging from 65 to 69 wt.%, as well as high concentrations of Al₂O₃ (15.71–16.68 wt.%) and K₂O (5.21–5.96 wt.%). Diatexites also have high content of minor elements; for instance, TiO₂ (0.676–0.886 wt.%) and P₂O₅ (0.42–0.55 wt.%). The trace element composition (Figure 7) shows that the diatexites are enriched in Zr (388–435 ppm), La (91.6–130 ppm), Ce (243–320 ppm), and Th (72.3–137 ppm).

The metatexites are more siliceous, SiO₂ (72–74 wt.%), than diatexites. These rocks yield high content of Al₂O₃ (14–15 wt.%), although these concentrations are lower in diatexites. High concentrations of Na₂O (3.3–3.5 wt.%) are observed in metatexites, oppositely to diatexites (Na₂O = 2.01–2.89 wt.%). The minor elements composition of the metatexites are very different from the diatexites because they have low values of TiO₂ (0.10–0.12 wt.%). The trace elements composition of these rocks is clearly lower than that of the diatexites for Zr (51–57 ppm), La (8–10 ppm), and Ce (16–21 ppm).

The rare earth elements (REE) signature of the diatexites exhibits an almost flat pattern from La to Pr (Figure 8), but it does not continue towards Sm without significant fractionation of the light rare earth elements (LREE–La/SmN = 2.59–4.43). The heavy rare earth elements (HREE) show a slightly more fractionated pattern (Gd/Yb)_N = 5.74–8.89). The negative Eu anomaly (Eu/Eu* = 0.20–0.40) is also evident. The low HREE content contrasts with the high Zr concentrations, suggesting different mineralogical controls. Conversely, the high LREE composition and Th content could be controlled by the presence of monazite [42]. By contrast, the REE abundances in the metatexites are lower than in diatexites, showing a fractionation of the LREE (La/Sm)_N = 2.46–2.62) and HREE (Gd/Yb)_N = 3.23–4.36), as well as a less pronounced negative Eu anomaly (Eu/Eu* = 0.51–0.61).

The trace element contents of the Pedregal diatexites and metatexites were compared with the average composition of materials similar to the country rocks of the Douro–Beiras Supergroup: NIBAS (Neoproterozoic Iberian Average Shale) [44]. The diatexites show a positive anomaly for Zr, LREE, Hf, Th, and U, and a negative anomaly for Y, Yb, Lu, and Ta in the multi-element diagram. Notably, the metatexites exhibit opposite behaviour for the elements in which the diatexites are enriched (Figure 9).

5.2. Sr-Nd Isotopic Geochemistry

The initial Sr-Nd ratios were recalculated using the 313 Ma age given by the analyzed zircons of the Pedregal diatexite (Figure 10;

Table 2. Whole-rock Sr and Nd of the Pedregal Migmatitic Complex (DTX—diatexite; MTX—metatexite).

Sample	P1	P2	P3	P6	P7	P8	P9	P10
Lithology	DTX	DTX	DTX	DTX	DTX	MTX	MTX	MTX
Rb (ppm)	266	383	321	403	383	174	164	171
Sr (ppm)	228	109	68	93	82	62	56	37
87Rb/86Sr	3.3804	10.1900	13.7074	12.6695	13.6581	8.1639	8.5228	13.4654
±2δ	0.10	0.29	0.39	0.36	0.39	0.23	0.24	0.38
87Sr/86Sr	0.72249	0.75125	0.77340	0.76940	0.77241	0.76314	0.76763	0.80116
±2δ	0.00002	0.00002	0.00002	0.00002	0.00003	0.00002	0.00003	0.00002
87Sr/86Sr313	0.7077	0.7067	0.7135	0.7140	0.7127	0.7275	0.7304	0.7423
Sm (ppm)	16.2	22.2	22.2	22.3	22.1	2.4	2.5	2.1
Nd (ppm)	110	156	146	142	148	8.9	9	7.7
147Sm/144Nd	0.0891	0.0861	0.0920	0.0950	0.0903	0.1631	0.1680	0.1650
±2δ	0.003	0.002	0.003	0.003	0.003	0.012	0.012	0.009
143Nd/144Nd	0.5121	0.5120	0.5121	0.5121	0.5121	0.5124	0.5124	0.5124
±2δ	0.00002	0.00002	0.00001	0.00001	0.00001	0.00001	0.00002	0.00001
εNd313	−5.78	−7.08	−7.10	−7.14	−6.85	−3.14	−3.43	−3.00
TDM1 (Ma)	1207	1263	1308	1334	1280	2018	2225	2057
TDM2 (Ma)	1485	1587	1589	1591	1569	1278	1301	1267

λRb [50], λSm [51], ¹⁴⁷Sm/¹⁴⁴Nd_CHUR and ¹⁴³Nd/¹⁴⁴Nd_CHUR [52], ¹⁴⁷Sm/¹⁴⁴Nd_DM, ¹⁴³Nd/¹⁴⁴Nd_DM, and ¹⁴⁷Sm/¹⁴⁴Nd_CC [53], T_DM1 = 1/λ × ln(¹⁴³Nd/¹⁴⁴Nd − ¹⁴³Nd/¹⁴⁴Nd_DM)/(¹⁴⁷Sm/¹⁴⁴Nd − ¹⁴⁷Sm/¹⁴⁴Nd_DM) + 1), T_DM2 = 1/λ × ln(¹⁴³Nd/¹⁴⁴Nd −(eλt − 1) × (¹⁴⁷Sm/¹⁴⁴Nd − ¹⁴⁷Sm/¹⁴⁴Nd_CC), ¹⁴³Nd/¹⁴⁴Nd_DM/(¹⁴⁷Sm/¹⁴⁴Nd_CC − ¹⁴⁷Sm/¹⁴⁴Nd_DM) + 1).

).

The metatexites have Rb (164–174 ppm), Sr (37–62 ppm), Sm (2.1–2.5 ppm), and Nd (7.7–9 ppm) compositions considerably lower than the diatexite (Rb: 266–403 ppm; Sr: 68–228 ppm; Sm: 16.2–22.2 ppm; Nd: 110–156 ppm). The ⁸⁷Sr/⁸⁶Sr₃₁₃ (0.7067–0.7140) and εNd313 (−5.78 to −7.14) for the diatexites are low comparatively to the metatexites (⁸⁷Sr/⁸⁶Sr₃₁₃: 0.727–0.7423; εNd₃₁₃: −3.00 to −3.43), revealing a divergence between both lithologies. This fact was not expected considering the field, petrographic, and geochemical features of the Pedregal diatexites and metatexites [13]. The ¹⁴⁷Sm/¹⁴⁴Nd ratio of the metatexites (0.1631–0.1680) is close to the mean for the Iberian Massif metasediments (0.120) [45] probably because of the paleosomatic contribution of the Douro–Beiras Supergroup metasediments in the metatexites. However, this isotopic ratio is slightly higher, which is probably related to the garnet content of the metatexites. Diatexites and metatexites yield slightly different two-stage Nd model ages (diatexites TDM2: 1.45–1.59 Ga; metatexites TDM2: 1.27–1.30 Ga).

5.3. Rutile Trace Elements Composition

The analysed rutiles from the Pedregal diatexite exhibit subhedral to euhedral grains, sometimes with an oval shape and a brownish colour. In the diatexite, rutile occurs as single grains in the matrix between the rock-forming minerals and as needle shape inclusions in the quartz and biotite. The needle-like morphology implies that rutile grew during prograde metamorphism and with the increasing of the metamorphic grade the rutile assumes a grain morphology [54].

In general, the rutile is a host of Cr and high-field-strength elements, mainly Nb, Ta, and Zr. In the Pedregal diatexite, Cr contents in rutile grains vary from 17 to 30 ppm, while Nb concentrations exhibit a large range with values between 104 and 8711 ppm. Tantalum also reveals a large spread varying from 5 to 106 ppm. The Zr composition of the rutile is roughly constant (60–86 ppm) with exception of two outliers with 4671 and 770 ppm. Hafnium presents the same behaviour of Zr, with concentration of 4 to 6 ppm and the same outliers with higher contents (19 ppm and 25 ppm). Curiously, high values of LREE and Th are shown by at least for 6 grains.

The Zr-in-rutile geothermometer was calculated, and three geothermometers were tested to understand if there were relevant differences (

Table 3. Rutile trace elements composition and respective Zr-in-rutile thermometer results for the Pedregal diatexite.

Points	Nb (ppm)	Ta (ppm)	Cr (ppm)	Zr (ppm)	Hf (ppm)	LREE (ppm)	Th (ppm)	T (°C) [55]	T (°C) [56]	T (°C) [57] P = 6	T (°C) [57] P = 8	T (°C) [57] P = 10
1	6642.26	82.56	25.07	72.21	4.93	0.21	0.15	537	539	534	543	551
2	397.48	32.11	27.06	72.39	5.13	0.06	0.23	537	540	534	543	551
3	8710.66	105.54	25.59	71.71	4.99	0.05	0.09	536	539	534	542	551
4	221.49	14.41	18.63	67.52	4.75	0.16	0.15	528	535	530	538	547
5	176.65	12.02	21.45	60.25	4.57	286.00	93.22	514	528	523	532	540
6	276.14	21.88	24.83	62.80	4.37	34.49	9.83	519	531	526	534	543
7	116.26	5.12	30.03	62.99	3.77	0.53	3.03	519	531	526	534	543
8	360.34	30.74	25.37	62.23	4.41	0.16	0.10	518	530	525	533	542
9	335.12	28.09	25.75	66.59	4.82	1.92	2.59	527	534	529	538	546
10	495.76	39.17	27.37	65.08	4.77	5.60	1.63	524	533	528	536	545
11	489.30	36.64	26.97	69.71	4.96	0.16	0.05	532	537	532	540	549
12	219.23	13.25	19.69	59.97	4.11	2161.58	606.02	513	528	523	531	540
13	276.67	20.87	22.85	65.11	4.64	4937.09	2366.31	524	533	528	536	545
14	119.12	5.51	17.35	61.54	3.80	0.09	0.67	516	529	524	533	541
15	187.28	8.15	20.55	60.80	4.26	47.03	11.62	515	529	524	532	541
16	103.63	5.29	21.61	66.22	4.53	0.29	0.08	526	534	529	537	546
17	473.17	56.95	21.59	78.63	5.47	6346.64	3479.42	548	545	539	548	557
18	430.12	38.72	19.31	78.46	5.65	0.75	0.10	548	545	539	548	556
19	348.16	32.87	20.34	70.82	5.42	0.51	0.34	534	538	533	541	550
20	204.44	12.33	19.61	4671.22	19.41	370.96	279.31	1070	938	907	919	932
21	156.18	7.52	24.96	73.82	5.22	0.02	0.10	540	541	535	544	553
22	136.68	5.46	24.80	65.87	4.52	0.02	0.04	525	534	528	537	546
23	423.94	35.52	27.01	770.07	24.98	645.19	53.30	839	726	710	721	731
24	220.48	14.06	20.88	71.21	5.24	1.30	0.28	535	538	533	542	550
25	251.38	18.59	30.25	85.84	6.27	1.09	1.00	559	551	545	553	562

). Zack et al. [55] present a Zr-in-rutile thermometer that gave temperatures from 513 to 559 °C for the rutile grains in this study. Watson et al. [56] modified the geothermometer considering experimental data at ~10 kbar, which results in temperatures between 528 and 551 °C for the analysed grains. At last, Tomkins et al. [57] calibrated three pressure dependent Zr-in-rutile thermometers for the α-quartz, β-quartz, and coesite field. Here, the β-quartz field thermometer was chosen because this is a high-temperature quartz, and in this study, we are working with high-temperature lithologies. Additionally, β-quartz field the thermometer was tested for three different pressures (6, 8, and 10 kbar), which are acceptable pressures for the Variscan HT-LP Barrovian metamorphism type. For the 6 kbar pressure, the temperatures are between 523 and 545 °C. When increasing the pressure for 8 kbar, the temperatures slightly increase for 531–553 °C. At 10 kbar, once again, the temperatures lightly rise to values between 540 and 562 °C. In each Zr-in-rutile geothermometer exposed in this work, there are two outliers (point 20 and 23 of Table 3) that correspond to the analyzed grains with higher Zr and Hf contents. These outliers are not considered in the discussion. Despite the application of the Zr-in-rutile thermometers from different authors, the resulting temperatures are very similar with a minimum temperature of 513 °C and a maximum temperature of 562 °C.

5.4. U-Pb Geochronology

The analysed zircons (30 grains) do not reveal a homogeneous morphology exhibiting rounded, oval, and prismatic crystals (Figure 11a). As crystal shape, the zircon internal textures are also heterogeneous, exhibiting oscillatory zoning typical of igneous zircon, convoluted zoning, and recrystallized borders typical of high-temperature metamorphism [58,59,60]. Some zircon crystals include xenocrystic cores.

The analysed zircons are characterized by a wide range of values for U (295–1084 ppm) and Th (39–424 ppm) with Th/U ratios between 0.05–0.41 (

Table 4. SHRIMP U–Pb zircon data of the Pedregal diatexite.

				Common-Lead Uncorrected					Variscan Final Ages (Ma)
ID	U (ppm)	Th (ppm)	Th/U	207Pb/235U	±2σ	206Pb/238U	±2σ	Rho *	206Pb/238U	±2σ	207Pb/235U	±2σ	Concordance (%) **
P2-1.1	438.54	111.94	0.26	0.35658	0.02062	0.04771	0.00214	0.55922	300	13	310	16	97
P2-12.1	641.72	214.85	0.34	0.36409	0.00490	0.05031	0.00053	0.56126	316	3	315	4	100
P2-14.2	669.21	161.19	0.25	0.35142	0.00434	0.04792	0.00036	0.44080	302	2	306	3	99
P2-20.1	523.48	125.99	0.25	0.36264	0.00199	0.04970	0.00011	0.27911	313	1	314	2	100
P2-21.1	646.92	105.69	0.17	0.36709	0.00373	0.05029	0.00037	0.51975	316	2	318	3	100
P2-23.1	319.38	59.59	0.19	0.36740	0.00780	0.04996	0.00048	0.32631	314	3	318	6	99
P2-24.2	667.59	133.07	0.20	0.36622	0.00202	0.05087	0.00021	0.54078	320	1	317	2	101
P2-3.1	569.74	107.25	0.19	0.35214	0.00446	0.04879	0.00054	0.62785	307	3	306	3	100
P2-5.1	545.45	124.50	0.23	0.34628	0.01603	0.04899	0.00194	0.61610	308	12	302	12	102
P2-6.1	467.78	92.70	0.20	0.35649	0.01592	0.04848	0.00193	0.64032	305	12	310	12	99
P2-8.1	759.53	177.77	0.24	0.36748	0.01078	0.04849	0.00115	0.57984	305	7	318	8	96
P2-8.2	364.41	148.06	0.42	0.35620	0.01764	0.04691	0.00218	0.67581	296	13	309	13	96
P2-9.1	515.34	191.55	0.38	0.35892	0.00602	0.04934	0.00050	0.43734	311	3	311	5	100
				Common-Lead Uncorrected							Inherited Final Ages (Ma)
ID	U (ppm)	Th (ppm)	Th/U	207Pb/206Pb	±2σ	207Pb/235U	±2σ	206Pb/238U	±2σ	Rho *	206Pb/238U	±2σ	207Pb/235U	±2σ	207Pb/206Pb	±2σ	Concordance (%) **
P2-10.1	179.10	164.87	0.94	0.06022	0.00086	0.83557	0.03238	0.10064	0.00360	0.66536	618	21	617	18	611	31	100
P2-11.1	228.86	343.07	1.54	0.06147	0.00045	0.84107	0.01033	0.09923	0.00090	0.53425	610	5	620	6	656	16	98
P2-13.1	338.63	190.71	0.58	0.06132	0.00044	0.79991	0.00890	0.09461	0.00073	0.49734	583	4	597	5	651	15	98
P2-15.1	756.04	304.45	0.41	0.05988	0.00078	0.72902	0.01614	0.08829	0.00155	0.56965	545	9	556	10	599	28	98
P2-17.1	665.98	269.52	0.42	0.07250	0.00316	1.39082	0.06675	0.13914	0.00275	0.29604	840	16	885	29	1000	86	95
P2-18.1	187.86	137.25	0.75	0.06026	0.00093	0.81887	0.01576	0.09855	0.00107	0.40548	606	6	607	9	613	33	100
P2-18.2	526.20	401.42	0.78	0.06161	0.00041	0.80328	0.01102	0.09456	0.00108	0.59907	582	6	599	6	661	14	97
P2-2.2	163.45	112.18	0.70	0.11033	0.00319	4.44407	0.25007	0.29213	0.01407	0.61622	1652	71	1721	48	1805	52	96
P2-22.1	268.95	98.35	0.38	0.05889	0.00059	0.77191	0.00987	0.09506	0.00067	0.39711	585	4	581	6	563	22	101
P2-7.1	129.83	79.45	0.63	0.07135	0.00090	1.49943	0.04354	0.15243	0.00395	0.64252	915	22	930	18	967	26	98

* Rho is calculated by SHRIMPTOOLS software, and it corresponds to the correlation between the given errors of the ²⁰⁷Pb/²³⁵U and ²⁰⁶Pb/²³⁸U ratio. ** Concordance = (²⁰⁶Pb/²³⁸U age)/(²⁰⁷Pb/²³⁵U age) × 100.

).

The ²⁰⁶Pb/²³⁸U ages of the Pedregal diatexite was defined from 13 concordant points, which gives a concordia age of 313.3 ± 0.5 Ma (MSWD = 0.75) (Figure 11b). The inherited zircon (10 concordant ages) exhibits a group of ages around 600 Ma (545–618 Ma; Ediacaran age) and other two groups with minor representation with ages around 900 Ma (Neoproterozoic age) and 1700 Ma (upper Paleoproterozoic age; Table 4).

5.5. Zircon Oxygen Isotopes

The zircon oxygen isotopes results (δ¹⁸O) range from 2.70 to 11.94 ‰ (n = 46). The low δ¹⁸O values (2.70–5.29 ‰; n = 8) are associated to fractures/inclusions in the zircon crystals, and they must not be considered to further interpretations. Thus, in the discussion, we will only consider the δ¹⁸O range between 6.34 and 11.94 ‰, with an average of 9.58 ± 0.08 ‰ (n = 38;

Table 5. δ¹⁸O (‰) results of the Pedregal diatexite and normalised to VSMOW.

ID	δ18O (‰)	±Error (95%)	18O/16O	Error (95%)	Comment	206Pb/238U Age (Ma)	±2σ
P2-1.1	9.65	0.06746	0.00202	0.00000014	-	300	13
P2-10.1	4.63	0.12962	0.00201	0.00000026	fracture or inclusion	618	21
P2-11.1	9.47	0.07264	0.00202	0.00000015	-	610	5
P2-12.1	8.93	0.08526	0.00202	0.00000017	-	316	3
P2-13.2	10.98	0.09169	0.00203	0.00000019	-	-	-
P2-14.1	4.65	0.06794	0.00201	0.00000014	fracture or inclusion	-	-
P2-15.1	9.25	0.08628	0.00202	0.00000017	-	545	9
P2-16.1	11.13	0.07682	0.00203	0.00000016	-	326	4
P2-17.1	6.58	0.09385	0.00202	0.00000019	-	840	16
P2-18.1	11.94	0.11393	0.00203	0.00000023	-	606	6
P2-19.1	10.73	0.08436	0.00203	0.00000017	-	291	4
P2-2.1	10.40	0.09662	0.00203	0.00000020	-	-	-
P2-20.1	9.67	0.12326	0.00202	0.00000025	-	313	1
P2-21.1	5.52	0.07376	0.00202	0.00000015	fracture or inclusion	316	2
P2-22.1	11.23	0.06784	0.00203	0.00000014	-	585	4
P2-23.1	9.83	0.06893	0.00202	0.00000014	-	314	3
P2-24.1	7.48	0.05721	0.00202	0.00000012	-	321	6
P2-25.1	9.39	0.09217	0.00202	0.00000019	-	-	-
P2-26.1	6.34	0.07965	0.00202	0.00000016	-	-	-
P2-27.1	10.98	0.09700	0.00203	0.00000020	-	-	-
P2-28.1	7.12	0.09780	0.00202	0.00000020	-	-	-
P2-29.1	10.55	0.06217	0.00203	0.00000013	-	-	-
P2-3.1	10.46	0.04913	0.00203	0.00000010	-	307	3
P2-30.1	10.51	0.11472	0.00203	0.00000023	-	-	-
P2-31.1	7.85	0.07252	0.00202	0.00000015	-	-	-
P2-32.1	7.01	0.06468	0.00202	0.00000013	-	-	-
P2-33.1	10.13	0.07407	0.00203	0.00000015	-	-	-
P2-34.1	9.54	0.10252	0.00202	0.00000021	-	-	-
P2-35.1	10.13	0.05394	0.00203	0.00000011	-	-	-
P2-36.1	9.72	0.08780	0.00202	0.00000018	-	-	-
P2-37.1	8.77	0.07616	0.00202	0.00000015	-	-	-
P2-38.1	9.57	0.10258	0.00202	0.00000021	-	-	-
P2-39.1	9.32	0.06220	0.00202	0.00000013	-	-	-
P2-4.1	10.34	0.07804	0.00203	0.00000016	-	-	-
P2-40.1	3.80	0.15000	0.00201	0.00000030	fracture or inclusion	-	-
P2-41.1	5.29	0.05377	0.00202	0.00000011	fracture or inclusion	-	-
P2-41.2	4.60	0.08730	0.00201	0.00000018	fracture or inclusion	-	-
P2-42.1	9.47	0.06931	0.00202	0.00000014	-	-	-
P2-43.1	9.39	0.10278	0.00202	0.00000021	-	-	-
P2-44.1	8.94	0.07377	0.00202	0.00000015	-	-	-
P2-45.1	2.70	0.21000	0.00201	0.00000042	fracture or inclusion	-	-
P2-5.1	10.05	0.06455	0.00203	0.00000013	-	308	12
P2-6.1	10.65	0.07805	0.00203	0.00000016	-	305	12
P2-7.1	4.90	0.08337	0.00202	0.00000017	fracture or inclusion	915	22
P2-8.2	9.54	0.08794	0.00202	0.00000018	-	296	13
P2-9.1	10.96	0.08797	0.00203	0.00000018	-	311	3

) as the original zircon isotope composition of the studied diatexite.

6. Discussion

6.1. Petrogenesis of the Pedregal Migmatites

The geological setting, petrological characteristics, whole-rock geochemistry, and rutile trace element geochemistry provide insights into the partial melting process that formed the Pedregal migmatites. In addition to the field association of the Pedregal diatexites and metatexites (e.g., field contacts and local breccia-like structure; Figure 2d,e), the geochemical features of the Pedregal Migmatitic Complex suggest a possible genetic relation resulting from the sequential melting processes between the lithologies and a progressive residuum-melt separation. This process of melt segregation and separation from the residuum is manifested through the negative correlation in the Harker diagrams (Figure 6 and Figure 7), and it is also evidenced in the multi-elements diagram (Figure 9), particularly for the LREE, Zr, Hf, and Th, which are enriched in diatexites and depleted in metatexites (Table 1). Yet the Pedregal diatexites reveal higher contents of FeO, MgO, and TiO₂ and lower concentrations of SiO₂ (Figure 6) when compared with the metatexites, suggesting that the diatexites are enriched in residual biotite (Figure 3b). In addition, the whole-rock geochemical composition of the diatexites shows an enrichment in high-field-strength elements (HFSE) such as LREE, Hf, Zr, Ti, Nb, Ta, Th, and U compared to the metatexites. All these aspects point to the residual character of the Pedregal diatexites [61,62]. The SiO₂ enrichment of the Pedregal metatexites relative to the diatexites can be explained by melt injection into the foliation plans of the host metasediments [62,63], giving origin to micaschist-derived metatexites. The melt injected into the host micaschists of the Douro–Beiras Supergroup corresponds to a first stage of melting, leaving a residual melt that corresponds to the Pedregal diatexites.

Considering the host rocks of the migmatites are of metapelitic character, it seems that partial melting reactions occurred through the melting of metapelitic rocks. Although the major elements signature, specially the Al₂O₃/TiO₂ vs. CaO/Na₂O (e.g., [14]), indicates that the metatexites have a metapelitic source, while the diatexites reveal a mixture between the metapelitic and psammitic source (Figure 12a) [64]. Nevertheless, rutile trace element composition, particularly log (Cr/Nb), indicates a pelitic provenance for the Pedregal diatexites (Figure 12b) [65].

The partial melting conditions of the potential metapelitic source to form the Pedregal migmatites involved the dehydration melting of muscovite and biotite at high temperatures (~800 °C) [66]. The migmatites reveal a depletion in Sr and Ba compositions as the Rb/Sr ratio increases, which match with the fluid-absent melting of muscovite and culminate with the biotite melting as well (Figure 13) [14,67]. The dehydration melting of muscovite is also noticed by the massive occurrence of secondary muscovite in the migmatites, which indicates the melting of the primary muscovite during the prograde path, and the secondary muscovite grew in the retrograde one. Still, at the thin section scale, the dehydration melting of biotite is observed by the corroded aspect of biotite. Moreover, this observation is supported by the petrographic observations since the metatexites exhibit garnet, K-feldspar, and sillimanite as peritectic minerals, which are products of the dehydration melting of biotite and muscovite, respectively. In the case of diatexites, garnet is not observed, but the K-feldspar and sillimanite are still present (muscovite dehydration melt products), plus hercynite spinel (biotite dehydration melt product). The hercynite can be a product of the reaction biotite + sillimanite, but also from the reaction garnet + sillimanite [68,69]. Thus, probably, the garnet is not a peritectic mineral in diatexites because it was consumed during the reaction to form hercynite. It is important to mention that near the study area occurs a gneiss–migmatite outcrop (Lavadores–Madalena) where hercynite is also present in its mineral assemblage [11]. The presence of the hercynite and the absence of garnet in the Pedregal diatexites are indicators of high-temperature, low-pressure metamorphism [70], as observed for the other anatectic complexes in the CIZ [11].

The zircon saturation temperatures (TsatZr) [71] for the metatexites are around 720 °C and higher for the diatexites that are approximately 900 °C (Table 1). The TsatZr estimated for the metatexites are closer to the metamorphic peak temperatures obtained for the metatexites of the Porto–Viseu Metamorphic Belt (769–812 °C; [28]) and to the minimum metamorphic peak conditions of another CIZ anatectic complex, the Figueira de Castelo Rodrigo–Lumbrales Anatectic Complex (T = 761 ± 50 °C; [15]). For the last anatectic complex, Ferreira et al. [71] estimated maximum metamorphic peak temperatures between 716 ± 48 °C and 869 ± 21 °C. The zircon saturation temperatures estimated to the Pedregal migmatites correspond to a maximum approximation of the melting temperatures, especially for the diatexites, which exhibit a relative higher proportion of inherited zircon (33–10 of 30 analysed grains; [72]) and an anomalous Zr whole-rock concentration (388–433 ppm), which can contribute for the overestimation of the TsatZr for the diatexites [73]. Yet the presence of hercynite in the diatexites cannot be ignored, indicating that the dehydration melting of biotite (800–850 °C; [74]) probably reached a maximum melting temperature of 900 °C [5,68,69,75]. Moreover, the 900 °C melting temperature is not too far from the 869 ± 21 °C calculated for the diatexites of the Figueira de Castelo Rodrigo-Lumbrales Anatectic Complex [76]. The retrograde reaction temperatures are registered in the analysed rutiles. When applied, the different Zr-in-rutile thermometers agree with a retrograde temperature around 550 °C (Table 3).

6.2. Significance of the U-Pb Age

The U-Pb geochronological data point to a Variscan age of (313.3 ± 0.5 Ma) for the Pedregal diatexite matching with ages of the syn Variscan magmatism of the CIZ. The most important meaning of this result is that the migmatization age in the CIZ probably must be redefined. The U-Pb results in this work reveal that the migmatization in the study area occurred during the D3 deformation phase and not exclusively in the D₂ as established until the moment for the CIZ migmatites [25]. For the migmatites of the Porto-Viseu Metamorphic Belt, it has been considered that the metamorphic peak occurred during the D₂ and the retrograde/cooling took place during the D₃ phase [77]. In contrast, the Pedregal diatexite reveals a peak metamorphic age synchronous with the D₃ deformation phase and the emplacement of syn-tectonic granites, as observed for other migmatites and anatectic areas of the CIZ [76,78]. In the Figueira de Castelo Rodrigo Anatectic Complex, Ferreira et al. [71] estimated U-Pb migmatization ages between 343.7 ± 2.5 and 314.8 ± 1.3 Ma, and a peak metamorphic age of 316.8 ± 2.0 Ma, implying that anatexis event in the CIZ started during the D₂ and culminated during the D₃ Variscan phase. In the same case study, Valverde-Vaquero et al. [79] also concluded that the HT–LP metamorphism is not synchronous in the CIZ, and some of this metamorphism occurred between 316–314 Ma, synchronous with the D₃ granite emplacement.

The group of ages ~600 Ma confirms the existence of zircons with inherited cores, and this age should correspond to restitic cores in the anatectic process of the detrital zircons that belong to Douro–Beiras Supergroup metasedimentary sequence in the CIZ [80]. The detrital zircon ages of the Pedregal diatexite around 600 Ma are compatible with the youngest zircon ages (630–545 Ma) of the Ediacaran metasedimentary sequence of Iberia [81]. This metasedimentary sequence represents the late Cadomian flysch, where there could exist contribution of zircons resulting from the arc-magmatism and/or Cadomian orogeny.

6.3. Source of Migmatites

In order to assess the most probable protolith of the migmatites, the Sr-Nd isotopic compositions recalculated for migmatization age (313 Ma) and the two-stage Nd model ages of the Pedregal Migmatitic Complex (Table 2) were compared with the following formations and anatectic complexes (Figure 10): Douro–Beiras Supergroup [16,45,46,47,48], which represents the host rocks of the anatectic complex; Ollo de Sapo magmatism [49]; stromatic metatexites and leucosomes of the Mindelo Migmatitic Complex [16], which is located in the NW limit of the Porto–Viseu Belt in the same alignment of the Pedregal Migmatitic Complex; and the Fânzeres granite, which occurs in the Pedregal geological context (Figure 1b), and it has a migmatite character and a whole-rock geochemical composition akin to the Pedregal metatexites [82].

For comparison, the Pedregal diatexites reveal isotopic affinities with the Ollo de Sapo formation and the Mindelo stromatic metatexites and leucosomes. Conversely, the metatexites are similar to the Fânzeres granite, the Douro–Beiras Supergroup, particularly for the Beiras Group, and also to the Ollo de Sapo magmatic formation.

Additionally, the zircon δ¹⁸O compositions for the Pedregal diatexite reflect an upper crustal origin of the source materials, and they are similar to zircon δ¹⁸O compositions from melts derived from metasedimentary protoliths and also S-type granites ([83] and references therein). The δ¹⁸O (VSMOW) vs. zircon U-Pb age (Figure 14) reveal that the Variscan zircon ages have a similar oxygen isotopic signature to the detrital zircons, particularly to the population between ca. 550–650 Ma (Douro–Beiras Supergroup depositional ages). Thus, zircon oxygen isotopes also indicate the Douro–Beiras Supergroup (Beiras group) as a possible source for the Pedregal diatexite. This is also supported by the Ediacaran inherited ages of the Pedregal diatexite (545–618 Ma), which are within the range of maximum depositional ages for the Beiras Group (544–578 Ma; [26,80]). Moreover, these data are corroborated by the Portugal Geological Map at 1/500,000 scale (north sheet), where the metasedimentary sequence of the Douro–Beiras Supergroup that hosts the Pedregal migmatites is mapped as Beiras group [84].

The field association and the trace element composition indicate a genetic link between the Pedregal migmatites, exhibiting a process of residuum-melt separation as mentioned before. However, the initial Sr-Nd isotopic composition suggests no petrogenetic relation between the two groups of migmatites (metatexite to diatexite) for having different isotopic ratios. Moreover, the Pedregal diatexite does not exhibit affinity with the Sr-Nd composition of the Douro–Beiras Supergroup datasets (Figure 10), which is the opposite of what is demonstrated with the zircon δ¹⁸O compositions and zircon-inherited ages. Nevertheless, isotopic disequilibrium during the partial melting process can explain the dissimilarities between the Pedregal diatexites and metatexites and the possible protolith.

6.4. Isotopic Disequilibrium During Partial Melting

Sr and Nd isotopic systems are widely applied to study the petrogenetic links between metamorphic and magmatic rocks because they often preserve their initial signatures and the nature of the protolith. Partial melting can generate melts in isotopic disequilibrium with their source rocks. This disequilibrium melting and, consequently, isotopic disequilibrium during anatexis can be explained by several factors [70,85,86,87,88,89,90,91,92]: (i) melt–segregation and extraction rates higher than the rate of chemical equilibration between melt and the residual solid; (ii) accessory phases dissolution was not sufficient to saturate the melt; (iii) the isotopic composition of the reactant minerals and the stoichiometry of the melting reactions involved in the partial melting process; (iv) heterogeneous source; (v) tectonic setting (syntectonic anatexis can favor disequilibrium melting).

The Sr-Nd initial isotopic compositions of the Pedregal diatexites and metatexites reveal differences and isotopic variability, up to four orders of magnitude for εNd₃₁₃ and 0.04 for ⁸⁷Sr/⁸⁶Sr₃₁₃, where the metatexites are more radiogenic (Figure 10). Looking at the actual isotopic compositions, the differences remain for the Nd compositions, but the measured Sr isotopic ratios display more similar values between metatexites and diatexites (Table 2). Despite the Sr-Nd initial composition, the field and geochemical characteristics of the Pedregal Migmatitic Complex suggest a possible genetic relation between the lithologies, resulting from the sequential melting processes and a progressive residuum-melt separation, which gave origin to the metatexites and residual diatexites.

The variability of the Sr and Nd isotopes in granites and migmatites can be attributed to the anatexis of different metasedimentary components through different melting reactions involving micas and accessory phases, suggesting that partial melting occurred under disequilibrium conditions causing isotopic disequilibrium [93]. The dehydration melting of micas can cause a disequilibrium on the Rb-Sr isotopic system and influence the Rb/Sr ratio of the melt, depending on which mica, muscovite, or biotite has more intervention in the melting reactions [4,92]. The Rb- and Sr-bearing minerals (e.g., muscovite and biotite; plagioclase, and K-feldspar, respectively) control the Rb-Sr system [89]. The progression from muscovite to biotite dehydration melting increases the ⁸⁷Sr/⁸⁶Sr in the melt due to the highly radiogenic character of biotite (high Rb/Sr ratio compared to muscovite; [92,94]). The diatexites reveal an ⁸⁷Sr/⁸⁶Sr₃₁₃ less radiogenic than the metatexites (Figure 15a), which is not expected since the biotite breakdown was more present in the diatexites genesis at higher temperatures. Such Sr isotopic disequilibrium could be caused by the presence of residual biotite in the Pedregal migmatites, which retain their isotopic composition during partial melting reactions [95]. This is particularly pertinent given the residual nature of the diatexites, as discussed previously.

The presence in the melt of the LREE rich minerals has important implications for trace element and Nd isotopic compositions (e.g., monazite and apatite) of the products of crustal anatexis [86,89,96]. The high values of P₂O₅ and LREE of the Pedregal diatexite and the correlation between εNd₃₁₃ vs. P₂O₅ (Figure 15b) and LREE (Figure 15c) of the migmatites point to the influence of the monazite and apatite in the εNd₃₁₃ concentrations. However, the diatexites are less radiogenic than the metatexitic rocks. The opposite was expected because the diatexites are more melt-enriched than the metatexites. The dehydration reactions favoured the dissolution of apatite to the melt that gave origin to the diatexites, and the metatexites have a lack of accessory minerals compared to the diatexites. As mentioned before, the dehydration melting of micas are involved in the origin of the migmatites, and these reactions dissolve apatite over monazite [21]. The last one is preferentially dissolved in the fluid-present reactions [89]. For this reason, the monazite present in the diatexites is probably causing the Nd isotopic disequilibrium because it behaved as a refractory mineral during the partial melting and inherited the isotopic signature of the source. Another reason for the difference in Sm/Nd isotopic composition between the migmatites is the presence of garnet in the metatexites. Besides monazite, garnet also affects the Nd isotopic disequilibrium since its retention in the residuum or its separation from a segregated melt arise melts with low Sm/Nd isotopic ratio [96]. Garnet remains in the residuum after residuum-melt segregation, and part of it was consumed during the partial melting to form hercynite present in the Pedregal diatexites, which also explain the lower Nd radiogenic composition of these rocks. The diatexites yield higher contents of Y (Figure 15d) and HREE (Figure 15e) comparatively with the metatexites, but it is not related with the garnet, which is absent in the diatexites but rather incorporated in zircon [42].

The whole-rock and isotopic composition similarities among other lithologies of the studied area (Figure 10), for instance, between the Pedregal metatexites and the Fânzeres granite [82], confirm that sequential melting processes and, consequently, residuum-melt separation [5] took place in the Pedregal Migmatitic Complex and indicates that there exists a genetic link between the region lithologies. Thus, the Pedregal diatexite isotopic dissimilarities are reinforced by an isotopic disequilibrium process that, as discussed before, were caused by the dehydration melting of micas and the residual character of accessory minerals.

7. Conclusions

The autochthonous domain of the Iberian massif (CIZ) was affected by high-temperature metamorphism during the Variscan orogeny. Several anatectic complexes are distributed along the CIZ, where migmatites and S-type granites association are observed. The Pedregal Migmatitic Complex occurs at the NW limit of the Porto–Viseu Metamorphic Belt, and it is characterised by the occurrence of diatexites and metatexites hosted by a Ediacaran–Cambrian metasedimentary sequence (Douro–Beiras Supergroup).

The petrogenesis of the Pedregal migmatites implied the partial melting of a pelitic source with melt-residuum separation through the dehydration melting of muscovite and biotite. Particularly for the diatexites, the dehydration melting of biotite also involved the sillimanite and garnet to produce hercynite. The metamorphic peak occurred at 313.5 ± 0.5 Ma during the D₃ and coeval with the syn-D₃ granite emplacement, reaching maximum melting temperatures around 800–900 °C.

The geological setting, zircon oxygen isotope, detrital zircon ages, and isotopic compositions indicate the metasediments of the Douro–Beiras Supergroup, particularly the Beiras group, as the most probable source of the Pedregal migmatites, with some contribution of the Ollo de Sapo magmatism. Considering the residual character of the diatexites, the dissimilar isotopic compositions of the Pedregal migmatites suggest a disequilibrium melting and consequent isotopic disequilibrium caused by the residual mineral phases, which control the radiogenic composition. For the Rb-Sr isotopic system, the residual biotite is responsible for the whole-rock isotopic disequilibrium retaining the isotopic composition of the source. The same is observed for the Sm-Nd isotopic composition where the residual monazite and garnet in the source retains the Nd radiogenic content, causing the disequilibrium between the diatexites (lower εNd₃₁₃) and metatexites (higher εNd₃₁₃).