Decoupling of Sr-Nd Isotopic Composition Induced by Potassic Alteration in the Shapinggou Porphyry Mo Deposit of the Qinling–Dabie Orogenic Belt, China

: In our previous study on petrogenesis of quartz syenite and granite porphyry, the host rocks of the Late Mesozoic Shapinggou Mo deposit in the Qinling–Dabie orogenic belt, we found that the initial Sr isotopic composition of the host rocks is strongly affected by the degree of K-alteration. Here, we provide further isotopic evidence of the host rocks and their minerals to investigate the geochemical behaviour of trace elements and isotopes during the alteration and to explain the phenomenon of decoupling of Sr–Nd isotopic composition. The quartz syenite and granite porphyry are altered by K-alteration in varying degrees and have high K 2 O and Rb contents and low Na 2 O, CaO, Sr, and Ba contents. Rock samples of both quartz syenite and granite porphyry have variable Rb/Sr ratios and initial 87 Sr/ 86 Sr values (even < 0.70) but contain quite homogeneous ε Nd (t) values ( − 12.8 to − 14.8). Minerals from the rocks of moderate to intense K-alteration have very low initial 87 Sr/ 86 Sr values (even < − 17), while those from the weakly altered rocks have 87 Sr/ 86 Sr(t) values of 0.7044 to 0.7084. The same phenomenon of the decoupling in Sr–Nd isotopic composition can be observed from several Mo deposits within the eastern Qinling–Dabie orogenic belt. This fact suggests similar hydrothermal features and a comparable origin for both the magmatic rocks and hydrothermal ﬂuids in this belt. A comparison between porphyry Mo and porphyry Cu deposits shows that elements and the Rb–Sr isotope system have different behaviours during the K-alteration, implying distinct material sources and igneous rocks for porphyry Mo and porphyry Cu deposits, respectively.

Within the eastern Qinling-Dabie orogenic belt, most of the Mo ores deposited in three pulses occurring at ~220 Ma, ~150-140 Ma, and ~130-110 Ma [27,29,46]. These Mesozoic mineralisation pulses are related to various tectonic environments: (1) in extensional tectonic settings related to the initial exhumation of deep continental subduction, (2) in back-arc extensions related to subduction of the Paleo-Pacific plate (or Izanagi), and (3) in transitional settings from transpression to extension caused by delamination of subducting slabs and thermal erosion [27,29,46].
The Shapinggou porphyry Mo deposit is located in the eastern section of the Beihuaiyang zone and northeast of the intersection between the Shangcheng-Macheng and Tongbai-Tongcheng faults [36,37] (Figure 1). The deposit is located in the centre of the ore district ( Figure 2) and comprises a major part of it, with a confirmed reserve of >2 Mt Mo metal and an average grade of ~0.18% [25,35]. Vein-type Pb-Zn deposits and fluorite deposits surround the Mo deposit and are hosted by several NW-, NE-, and EW-trending faults [35,36].
Intrusive rocks of Late Mesozoic ages (termed as the Yanshanian period in the Chinese literatures) are widely distributed in the Shapinggou ore district and range compositionally from ultra-mafic (olivine pyroxenite and gabbro with cumulatic texture) to intermediate (diorite dike and quartz syenite) to felsic rocks (granodiorite, granite, granite porphyry, etc.). These intrusive rocks can be divided into two magmatic stages on the basis of spatial relationship (Figures 2 and 3) and geochronological and geochemical evidence: magmatic stage I comprises granodiorite, monzogranite and granite, and minor ultra-mafic to intermediate rocks which were emplaced before ca. 120 Ma; magmatic stage II contains the major Mo ore-bearing rocks quartz syenite and granite porphyry and formed between 117 Ma and 111 Ma [43].
Previous studies have shown that the main ore bodies in the Shapinggou deposit have a bell-like morphology ( Figure 3) [3]. The ore is continuously mineralised and of a high grade. Drill cores indicate that the major body of ore occurs predominantly at the top of granite porphyry and in the contact zones between granite and quartz syenite that Within the eastern Qinling-Dabie orogenic belt, most of the Mo ores deposited in three pulses occurring at~220 Ma,~150-140 Ma, and~130-110 Ma [27,29,46]. These Mesozoic mineralisation pulses are related to various tectonic environments: (1) in extensional tectonic settings related to the initial exhumation of deep continental subduction, (2) in back-arc extensions related to subduction of the Paleo-Pacific plate (or Izanagi), and (3) in transitional settings from transpression to extension caused by delamination of subducting slabs and thermal erosion [27,29,46].
The Shapinggou porphyry Mo deposit is located in the eastern section of the Beihuaiyang zone and northeast of the intersection between the Shangcheng-Macheng and Tongbai-Tongcheng faults [36,37] (Figure 1). The deposit is located in the centre of the ore district ( Figure 2) and comprises a major part of it, with a confirmed reserve of >2 Mt Mo metal and an average grade of~0.18% [25,35]. Vein-type Pb-Zn deposits and fluorite deposits surround the Mo deposit and are hosted by several NW-, NE-, and EW-trending faults [35,36].
Intrusive rocks of Late Mesozoic ages (termed as the Yanshanian period in the Chinese literatures) are widely distributed in the Shapinggou ore district and range compositionally from ultra-mafic (olivine pyroxenite and gabbro with cumulatic texture) to intermediate (diorite dike and quartz syenite) to felsic rocks (granodiorite, granite, granite porphyry, etc.). These intrusive rocks can be divided into two magmatic stages on the basis of spatial relationship (Figures 2 and 3) and geochronological and geochemical evidence: magmatic stage I comprises granodiorite, monzogranite and granite, and minor ultra-mafic to intermediate rocks which were emplaced before ca. 120 Ma; magmatic stage II contains the major Mo ore-bearing rocks quartz syenite and granite porphyry and formed between 117 Ma and 111 Ma [43]. enclose the granite porphyry [3]. The high-grade ore is situated in the inner contact zone, and low-grade ore bodies occur in granite and quartz syenite at a substantial distance from the contact zone.  The quartz syenite and granite porphyry formed between 117 Ma and 111 Ma and play a key role in mineralisation [7,40]. They underwent hydrothermal alteration of different types [43], including silicification, K-alteration (K-feldspathisation with minor biotitisation), pyritization-sericitisation-silicification (PSS), and chloritisation. The hydrothermal alteration is strongly developed within the deposit and can be divided into three zones, outwards from the granite porphyry: silicification, K-feldspathisation, and PSS zone. Weak chloritisation zones are also randomly distributed. Spatial relationships between the ore body and the alteration zones indicate that the Mo mineralisation is closely related to the K-feldspathisation and the PSS (Figure 3). Ore-bearing quartz syenite ex- enclose the granite porphyry [3]. The high-grade ore is situated in the inner contact zone, and low-grade ore bodies occur in granite and quartz syenite at a substantial distance from the contact zone.  The quartz syenite and granite porphyry formed between 117 Ma and 111 Ma and play a key role in mineralisation [7,40]. They underwent hydrothermal alteration of different types [43], including silicification, K-alteration (K-feldspathisation with minor biotitisation), pyritization-sericitisation-silicification (PSS), and chloritisation. The hydrothermal alteration is strongly developed within the deposit and can be divided into three zones, outwards from the granite porphyry: silicification, K-feldspathisation, and PSS zone. Weak chloritisation zones are also randomly distributed. Spatial relationships between the ore body and the alteration zones indicate that the Mo mineralisation is closely related to the K-feldspathisation and the PSS (Figure 3). Ore-bearing quartz syenite ex- Previous studies have shown that the main ore bodies in the Shapinggou deposit have a bell-like morphology ( Figure 3) [3]. The ore is continuously mineralised and of a high grade. Drill cores indicate that the major body of ore occurs predominantly at the top of granite porphyry and in the contact zones between granite and quartz syenite that enclose the granite porphyry [3]. The high-grade ore is situated in the inner contact zone, and low-grade ore bodies occur in granite and quartz syenite at a substantial distance from the contact zone.
The quartz syenite and granite porphyry formed between 117 Ma and 111 Ma and play a key role in mineralisation [7,40]. They underwent hydrothermal alteration of different types [43], including silicification, K-alteration (K-feldspathisation with minor biotitisation), pyritization-sericitisation-silicification (PSS), and chloritisation. The hydrothermal alteration is strongly developed within the deposit and can be divided into three zones, outwards from the granite porphyry: silicification, K-feldspathisation, and PSS zone. Weak chloritisation zones are also randomly distributed. Spatial relationships between the ore body and the alteration zones indicate that the Mo mineralisation is closely related to the K-feldspathisation and the PSS (Figure 3). Ore-bearing quartz syenite exhibits silicification, K-feldspathisation, biotitisation, PSS, and partially chloritisation. Most of the alterations within the ore-barren quartz syenite are weakly developed, whereas the ore-bearing rocks show moderate K-feldspathisation and biotitisation or distinct PSS.
The Shapinggou granite porphyry is a small intrusive stock intruding into the quartz syenite and exhibits considerable alterations, including silicification, K-feldspathisation, and PSS [3]. In hand specimens (Figure 4a,c), K-feldspathisation exhibits a K-feldsparisation halo (darker fleshy red) in the contact zone between rock and hydrothermal veins. Correspondingly, K-feldspathisation can be identified by secondary enlargement of the original K-feldspar ( Figure 4d) or by plagioclase replaced by K-feldspar, with fine-grained and vermicular quartz and recrystallised albite around the edges of K-feldspar grains (Figure 4b) under the polarizing microscope. The K-feldspathisation of deeply placed granite porphyry is stronger than that at shallower depths because the latter is overprinted by the PSS, which can be recognised by their alteration mineral assembly of pyrite, sericite (or muscovite), and quartz (Figure 4e,f). hibits silicification, K-feldspathisation, biotitisation, PSS, and partially chloritisation. Most of the alterations within the ore-barren quartz syenite are weakly developed, whereas the ore-bearing rocks show moderate K-feldspathisation and biotitisation or distinct PSS. The Shapinggou granite porphyry is a small intrusive stock intruding into the quartz syenite and exhibits considerable alterations, including silicification, K-feldspathisation, and PSS [3]. In hand specimens (Figure 4a,c), K-feldspathisation exhibits a K-feldsparisation halo (darker fleshy red) in the contact zone between rock and hydrothermal veins. Correspondingly, K-feldspathisation can be identified by secondary enlargement of the original K-feldspar ( Figure 4d) or by plagioclase replaced by K-feldspar, with fine-grained and vermicular quartz and recrystallised albite around the edges of K-feldspar grains (Figure 4b) under the polarizing microscope. The K-feldspathisation of deeply placed granite porphyry is stronger than that at shallower depths because the latter is overprinted by the PSS, which can be recognised by their alteration mineral assembly of pyrite, sericite (or muscovite), and quartz (Figure 4e,f).

Samples and Analytical Methods
The samples of quartz syenite and granite porphyry analysed were collected from seven drill holes in the deposit area. Twelve samples of quartz syenite are fresh to weakly altered rocks. Eighteen samples of granite porphyry show different degrees from weak to intense K-alteration.
Whole rock powder is obtained by grinding more than 5 kg of rock. Minerals are handpicked under a binocular microscope and crushed in agate mortar to approximately 200 mesh. Major element analyses of whole rock are performed by the Australian Laboratory Services Chemex in Guangzhou, China. Major element compositions are determined using a PANalytical Axios X-ray fluorescence (XRF) spectrometer made in Holland. The analytical conditions of the XRF were 50 kV voltage and 50 mA current. Loss on ignition (LOI) is determined by gravimetric method, that is, weighing 1 g sample powder heated to 1100 • C for 1 hour. Analytical uncertainties of major elements are better than 1%. Trace elements of whole rocks and minerals are determined by a Perkin-Elmer ELAN 6100 DRC II ICP-MS located at the University of Science and Technology of China (USTC). According to the analyses of Chinese and United States Geological Survey (USGS) geochemical reference standards GSR-1, AGV-2, and BHVO-2, the relative deviation of accuracy is better than 5% for most elements and about 10% for Rb, Zr, and Hf. Details of the analytical techniques are available elsewhere [47]. Sr and Nd isotopic compositions of whole rocks and rock-forming minerals are measured on a Finnigan MAT-262 thermal ionisation mass spectrometer located in the Laboratory for Radiogenic Isotope Geochemistry at the USTC. Approximately 200 mg of rock or mineral powder is dissolved by a HF-HNO 3 -HClO 4 acid solution for more than 5 days. Rb-Sr and Sm-Nd are separated on quartz columns by ion-exchange chromatography. NIST SRM 987 and La Jolla standard solution are used as standard for Sr and Nd analyses, respectively. More details on chemical processes and analytical techniques are available elsewhere [48,49].

Major and Trace Elemental Geochemistry
The major element contents of whole-rock samples are given in Table S1. Quartz syenite and granite porphyry are characterised by high SiO 2 and K 2 O contents, low CaO and Na 2 O contents, high K 2 O/Na 2 O ratios, high differentiation index values (DI; the sum of the mass percentages of quartz, orthoclase, albite, nepheline, leucite, and kalsilite; 90.36-94.55 and 96.39-99.20, respectively), and low molar Al 2 O 3 /(CaO+Na 2 O+K 2 O) ratio (A/CNK) values, suggesting that they belong to the shoshonite (SHO) series and are metaluminous to weakly peraluminous. Generally, porphyry granites of the Shapinggou Mo deposit with moderate or intense K-alteration exhibited higher SiO 2 and K 2 O contents but lower Na 2 O and CaO contents than those with weak K-alteration.
The trace element contents of whole rocks are given in Table S2 and are plotted in a primitive mantle-normalised spider diagram (Figure 5a). The quartz syenite and granite porphyry shared consistent geochemical features regarding trace elements. Both are depleted in large ion lithophile elements (LILEs; e.g., Ba and Sr), P and Ti; enriched in Rb and high field-strength elements (HFSEs; e.g., Th and U); and weakly enriched in Ta, Nb, Zr, and Hf. Increasing intensity of K-alteration is accompanied by increasing concentrations of Rb and K and decreasing Ba and Sr contents (Figure 5a). Both the quartz syenite and granite porphyry had similar rare earth element (REE) patterns (Figure 5b; data in Table S3), showing strong enrichment in LREEs (La-Nd), depletion in MREEs (Sm-Ho), weak enrichment in HREEs (Er-Lu and Y), and consistently moderate negative Eu anomalies. Granite porphyry samples have lower contents of total REEs and stronger MREE depletion than those of quartz syenite, implying that it is highly differentiated and evolved. This observation is also consistent with their differentiation index values. lower Sr contents and Rb/Sr ratios than weakly altered rocks by one to three orders of magnitude, and Ba contents are lower by two orders of magnitude (Figure 6a-d).  Rb contents of K-feldspar, plagioclase, quartz, and biotite increase an order of magnitude with increasing K-alteration degree, but rocks that have intense K-alteration have lower Sr contents and Rb/Sr ratios than weakly altered rocks by one to three orders of magnitude, and Ba contents are lower by two orders of magnitude (Figure 6a-d). lower Sr contents and Rb/Sr ratios than weakly altered rocks by one to three orders of magnitude, and Ba contents are lower by two orders of magnitude (Figure 6a-d).

Sr-Nd Isotopic Composition
Rb-Sr and Sm-Nd isotopic data of whole-rock samples are given in

Cause for Anomalously Low Initial 87 Sr/ 86 Sr Values
As noted above, intrusive rocks of magmatic stage II in the Shapinggou Mo deposit, including quartz syenite and granite porphyry, are strongly linked to mineralisation. Four lines of evidence are used to evaluate the magmatic origin of these ore-related rocks: (1) close cooling ages and geographic relationships; (2) uniform composition of major elements and trace and rare earth elements; (3) consistent Nd-Hf isotopic compositions, which are notably different from those of intrusive rocks of magmatic stage I [7]; and (4) similar whole-rock initial 87 Sr/ 86 Sr values ( Figure 7) and identical Rb-Sr isochrons (Figure 6f) between quartz syenite and weakly K-altered granite porphyry (ranging from 0.7077 to 0.7090 with a mean value of 0.7080 and from 0.7061 to 0.7094 with a mean value of 0.7076, respectively). Hence, it can be concluded that the magma of both quartz syenite and granite porphyry were derived from a similar source (or sources). The genetic relationship between quartz syenite and granite porphyry implies that they should have uniform initial 87 Sr/ 86 Sr values; however, this does not appear to be supported by the scattered and very low initial 87 Sr/ 86 Sr values obtained for granite porphyry with moderate to strong K-alteration.
On a larger scale, ore-bearing granite porphyries in the eastern Qinling-Dabie Mo belt are characterised by uniformly high concentrations of Si, K, and Rb, but low concentrations of Na, Ca, and Sr, and underwent alteration by K-rich hydrothermal fluids [3][4][5]9,27,29,46,51,52]. When the Sr-Nd isotopic data of the porphyry rocks from several typical deposits in the eastern Qinling-Dabie Mo belt are plotted on the ε Nd (t) versus ( 87 Sr/ 86 Sr) t diagram (Figure 7), all the rocks noticeably have similar ε Nd (t) values (approximately −14; Figure 7) but scattered initial 87 Sr/ 86 Sr values. The similarity in initial ε Nd values implies that all the porphyry bodies, related to the Mo mineralisation in this belt, are derived by melting of the same material sources. Quartz syenite and granite porphyry in the Shapinggou Mo deposit have high concentrations of SiO2, K2O, and Rb and low concentrations of CaO, Na2O, and Sr [3,7,39,40], especially moderately and intensely K-altered rocks. When the geochemical data of whole rocks are plotted on a covariance plot for different compounds, such as SiO2, Na2O, CaO, and Sr, decreasing Na2O and CaO contents are correlated with increasing SiO2 contents, and decreasing Sr contents are correlated with decreasing CaO contents (Figure 8a-d). In the major rock-forming minerals, Rb contents and Rb/Sr ratios are increasing appreciably with increasing K-alteration; at the same time, Sr, Ba, and Pb contents are drastically diminishing (Figure 6a-d). These correlations suggest that extra Si, K, and Rb were added, and Ca, Na, Sr, and Ba were removed from the rocks during the fluid-rock interactions. The following fluid-rock reactions have been suggested to illustrate the progression of the process [54]: K + replaces Na + and Ca 2+ during K-feldspathisation from albite and anorthite (Equations (1) and (2) below), followed by Na + and Ca 2+ being driven into the hydrothermal fluids. Owing to extensive isomorphism between K-Rb and Na-Sr, Rb moves in, whilst Sr is scavenged from rocks with the K-alteration overprint: NaAlSi3O8 (albite) + K + → KAlSi3O8 (K-feldspar) + Na + (1) 3CaAl2Si2O8 (anorthite) + 2K + →2KAlSi3O8 (K-feldspar) + 2Al2O3 + 2CaO + Ca 2+ (2) Minerals from quartz syenite and weakly K-altered granite porphyry have similar concentrations of Rb, Sr, and Ba and Rb/Sr ratios (Figure 6a-d). They fall on an isotopic isochron and have similar initial 87 Sr/ 86 Sr values (Figure 6e,f). From granite porphyry samples, it can be observed that Sr and Ba contents decrease much more remarkably (by 1-3 orders of magnitude) with increasing their alteration degrees (from weak to intense), compared to the increasing range of Rb contents (by 1 order of magnitude). Moderately and intensely K-altered granite porphyry rocks of the Shapinggou Mo deposit have much lower initial 87 Sr/ 86 Sr values than weakly K-altered rocks and are characterised by a large scatter of initial values from 0.505396 to 0.709385 (Figures 7 and 8f), and their minerals have very low initial 87 Sr/ 86 Sr values, especially for biotite, K-feldspar, and plagioclase (up to −17.4, −11.9, and −4.4). The minerals and the stronger K-altered rocks have higher K contents and lower initial 87 Sr/ 86 Sr values (Figure 9). Such occurrences are not observed in quartz syenite, but both quartz syenite and granite porphyry have almost coincident Nd-Hf isotopic compositions (Figure 8e,f). Thus, this difference in Sr isotopic composition between quartz syenite and granite porphyry of the Shapinggou Mo deposit is most likely related to the K-alteration rather than to different magmatic origins. Quartz syenite and granite porphyry in the Shapinggou Mo deposit have high concentrations of SiO 2 , K 2 O, and Rb and low concentrations of CaO, Na 2 O, and Sr [3,7,39,40], especially moderately and intensely K-altered rocks. When the geochemical data of whole rocks are plotted on a covariance plot for different compounds, such as SiO 2 , Na 2 O, CaO, and Sr, decreasing Na 2 O and CaO contents are correlated with increasing SiO 2 contents, and decreasing Sr contents are correlated with decreasing CaO contents (Figure 8a-d). In the major rock-forming minerals, Rb contents and Rb/Sr ratios are increasing appreciably with increasing K-alteration; at the same time, Sr, Ba, and Pb contents are drastically diminishing (Figure 6a-d). These correlations suggest that extra Si, K, and Rb were added, and Ca, Na, Sr, and Ba were removed from the rocks during the fluid-rock interactions. The following fluid-rock reactions have been suggested to illustrate the progression of the process [54]: K + replaces Na + and Ca 2+ during K-feldspathisation from albite and anorthite (Equations (1) and (2) below), followed by Na + and Ca 2+ being driven into the hydrothermal fluids. Owing to extensive isomorphism between K-Rb and Na-Sr, Rb moves in, whilst Sr is scavenged from rocks with the K-alteration overprint: Minerals from quartz syenite and weakly K-altered granite porphyry have similar concentrations of Rb, Sr, and Ba and Rb/Sr ratios (Figure 6a-d). They fall on an isotopic isochron and have similar initial 87 Sr/ 86 Sr values (Figure 6e,f). From granite porphyry samples, it can be observed that Sr and Ba contents decrease much more remarkably (by 1-3 orders of magnitude) with increasing their alteration degrees (from weak to intense), compared to the increasing range of Rb contents (by 1 order of magnitude). Moderately and intensely K-altered granite porphyry rocks of the Shapinggou Mo deposit have much lower initial 87 Sr/ 86 Sr values than weakly K-altered rocks and are characterised by a large scatter of initial values from 0.505396 to 0.709385 (Figures 7 and 8f), and their minerals have very low initial 87 Sr/ 86 Sr values, especially for biotite, K-feldspar, and plagioclase (up to −17.4, −11.9, and −4.4). The minerals and the stronger K-altered rocks have higher K contents and lower initial 87 Sr/ 86 Sr values (Figure 9). Such occurrences are not observed in quartz syenite, but both quartz syenite and granite porphyry have almost coincident Nd-Hf isotopic compositions (Figure 8e,f). Thus, this difference in Sr isotopic composition between quartz syenite and granite porphyry of the Shapinggou Mo deposit is most likely related to the K-alteration rather than to different magmatic origins. Anomalously low and scattered initial 87 Sr/ 86 Sr values, but uniform initial 143 Nd/ 144 Nd values for ore-bearing granite porphyries of many porphyry Mo deposits in the eastern Qinling-Dabie belt, such as the Jinduicheng, Donggou, Tangjiaping, and Shapinggou deposits, have been reported in previous studies [3][4][5][6][7]. Several conditions from K-rich hydrothermal fluids might result in strong isobaric interference on 87 Sr during the analysis [55,56] (this situation, however, is unlikely to occur because of the ionising method of thermal ionisation mass spectrometer); (2) the process of scavenging Sr into fluids from the porphyry body during fluid-rock interactions might lead to fractionation of 87 Sr, and it has been suggested that the diffusion of Sr in feldspar under hydrous conditions is an order of magnitude faster than under anhydrous conditions [57,58]; and (3) abnormally low initial 87 Sr/ 86 Sr values could result from an over-deduction of radiogenic 87 Sr due to additional Rb if the K-alteration occurred several million years later than the formation of granite porphyry. Previous studies have confirmed that the K-alteration is widespread and is correlated with the main episodes of Mo mineralisation in the eastern Qinling-Dabie orogenic belt; it was often observed in the Yuchiling, Shangfanggou, Tianmugou, Tangjiaping, and Shapinggou Mo deposits [12][13][14]33,34]. Figure 7 shows that the porphyry bodies from representative porphyry Mo deposits, such as the Jinduicheng, Donggou, and Tangjiaping Mo deposits, have scattered and shows low initial 87 Sr/ 86 Sr values, similar to the granite porphyry of the Shapinggou Mo deposit. The crust-mantle mixture was suggested to explain low initial 87 Sr/ 86 Sr values (0.701 to 0.705) obtained for the Jinduicheng granite porphyry [6]. However, this suggestion is a misstatement or oversimplification, as initial 87 Sr/ 86 Sr values cannot be used as an isotopic tracer for assessing magmatic origin in this situation. The Rb-Sr isotopic system is readily affected by fluid infiltration and thermal events because both elements are mobile [57]. In contrast, the solubility of Sm and Nd is too low to be mobile in fluids, and they often concentrate in accessory minerals, such as monazite and apatite, which are geochemically inert in fluid activity [2,53,59,60]. Thus, K-alteration can significantly alter the Rb-Sr isotopic system, whereas the Sm-Nd system is insensitive to alteration and can be a reliable indicator for tracing magma sources (Figure 8e,f).

Consequences of K-Alteration for Porphyry Mo and Cu Deposits
An experimental study of Mo geochemical behaviour suggested that Mo prefers alkaline fluids for extraction, dissolution, and migration but acidic fluids for precipitation [61]. According to the spatial relationships between the Shapinggou ore bodies and al- Anomalously low and scattered initial 87 Sr/ 86 Sr values, but uniform initial 143 Nd/ 144 Nd values for ore-bearing granite porphyries of many porphyry Mo deposits in the eastern Qinling-Dabie belt, such as the Jinduicheng, Donggou, Tangjiaping, and Shapinggou deposits, have been reported in previous studies [3][4][5][6][7]. Several conditions can cause anomalously low initial 87 Sr/ 86 Sr values of altered rocks: (1) large amounts of Rb from K-rich hydrothermal fluids might result in strong isobaric interference on 87 Sr during the analysis [55,56] (this situation, however, is unlikely to occur because of the ionising method of thermal ionisation mass spectrometer); (2) the process of scavenging Sr into fluids from the porphyry body during fluid-rock interactions might lead to fractionation of 87 Sr, and it has been suggested that the diffusion of Sr in feldspar under hydrous conditions is an order of magnitude faster than under anhydrous conditions [57,58]; and (3) abnormally low initial 87 Sr/ 86 Sr values could result from an over-deduction of radiogenic 87 Sr due to additional Rb if the K-alteration occurred several million years later than the formation of granite porphyry.
Previous studies have confirmed that the K-alteration is widespread and is correlated with the main episodes of Mo mineralisation in the eastern Qinling-Dabie orogenic belt; it was often observed in the Yuchiling, Shangfanggou, Tianmugou, Tangjiaping, and Shapinggou Mo deposits [12][13][14]33,34]. Figure 7 shows that the porphyry bodies from representative porphyry Mo deposits, such as the Jinduicheng, Donggou, and Tangjiaping Mo deposits, have scattered and shows low initial 87 Sr/ 86 Sr values, similar to the granite porphyry of the Shapinggou Mo deposit. The crust-mantle mixture was suggested to explain low initial 87 Sr/ 86 Sr values (0.701 to 0.705) obtained for the Jinduicheng granite porphyry [6]. However, this suggestion is a misstatement or oversimplification, as initial 87 Sr/ 86 Sr values cannot be used as an isotopic tracer for assessing magmatic origin in this situation. The Rb-Sr isotopic system is readily affected by fluid infiltration and thermal events because both elements are mobile [57]. In contrast, the solubility of Sm and Nd is too low to be mobile in fluids, and they often concentrate in accessory minerals, such as monazite and apatite, which are geochemically inert in fluid activity [2,53,59,60]. Thus, K-alteration can significantly alter the Rb-Sr isotopic system, whereas the Sm-Nd system is insensitive to alteration and can be a reliable indicator for tracing magma sources (Figure 8e,f).

Consequences of K-Alteration for Porphyry Mo and Cu Deposits
An experimental study of Mo geochemical behaviour suggested that Mo prefers alkaline fluids for extraction, dissolution, and migration but acidic fluids for precipitation [61]. According to the spatial relationships between the Shapinggou ore bodies and alteration zones ( Figure 3) and studies on fluid inclusion [25], the main ore bodies occur predominantly at the top of the K-feldspathisation zone and in the overlapping region with the PSS zone. This finding indicates that the main Mo mineralisation or precipitation from fluids began in the later stage of the K-feldspathisation to the PSS, in which fluid boiling and CO 2 separation occurred [25]. Previous studies [15,54,62] have reported that fluids causing K-feldspathisation are alkaline in composition, but those causing PSS are acidic; thus, it can be inferred that the Mo precipitation is triggered by the pH value in the fluid changing from alkaline to acidic. Thus, it can be proposed that the CO 2 separation from fluids during the K-feldspathisation and PSS stages might alter the alkaline-acid balance and trigger Mo precipitation owing to the variable solubility of Mo at different pH values. It has been reported that the rock-and ore-forming processes occurred under a decompression mechanism, likely triggering CO 2 separation from original fluids [43].
In porphyry Cu deposits, it has been documented that metallogenic material (such as metals, sulphur, and fluids) can be provided by mafic magmas [72][73][74]. However, metallogenic materials still remain debated in porphyry Mo deposits. Porphyry Cu deposits are considered to have a close genetic relationship with adakite due to its high oxygen fugacity (f O 2 ) and high contents of sulphur and H 2 O [64,70,75,76]. High f O 2 and high contents of sulphur and H 2 O can contribute to destroying the saturated sulphide in magma sources and release metal and S to initial magma. The metal content increases in earlystage magma due to the high solubility of sulphur and high partition coefficient of Cu (chalcophile elements) between sulphide and silicate. Metal precipitates and porphyry Cu mineralisation forms accompanied sulphate reduction and a corresponding decrease in pH value [70,75,76]. However, the oxygen fugacity of a pure porphyry Mo deposit is hard to predict because limited indicator minerals for redox state have been reported, such as magnetite and hematite for oxidised state or pyrrhotite for reduced state.
Compared to porphyry Mo deposits, such as the Shapinggou deposit mentioned above, porphyry Cu deposits, represented by the Dexing porphyry Cu deposit, have different characteristics in Sr isotopic composition during the K-alteration. Extensively altered rocks associated with the Dexing porphyry Cu deposit have higher Rb and lower Sr concentrations (ranging from 10 1 to 10 2 ppm and from 10 3 to 10 1 ppm, respectively) and higher Rb/Sr ratios (ranging from 10 −1 to 10 1 ; Figure 8) [53] but slightly higher initial 87 Sr/ 86 Sr values (ranging from 0.705 to 0.711; Figure 8e,f) [53] than unaltered rocks from the same rock unit. This implies that hydrothermal fluid introduced additional Rb and removed minor Sr from the igneous rocks. The difference between porphyry Cu and Mo deposits can be explained by the following two factors. Firstly, fresh igneous rocks associated with the porphyry Cu and porphyry Mo deposits have different geochemical compositions. Porphyry bodies related to porphyry Mo deposits are characterised by relatively high K and Rb concentrations and low Ca and Sr concentrations and are of an A-type granite composition (e.g., granite porphyry). In contrast, porphyry rocks related to porphyry Cu deposits are normally of adakitic composition (e.g., granodiorite porphyry or quartz monzonite porphyry): for example, porphyry rocks in the Dexing porphyry Cu deposit [10,42]. These rocks have high Ca and Sr concentrations, and as a result, their Rb/Sr ratios increase much less than those of the granite porphyry related to a Mo deposit after K-alteration by hydrothermal fluids. Secondly, the origin and geochemical composition of the hydrothermal fluids in the alteration can differ substantially between porphyry Cu and Mo deposits, which can be reflected in different types of alterations; the K-feldspathisation is the predominant alteration in porphyry Mo deposits but not in porphyry Cu deposits. Fluids in the Dexing porphyry Cu deposit contain less K and more Ca than those in the Shapinggou porphyry Mo deposit. Furthermore, the varieties in geochemistry and Re contents of molybdenite minerals from the porphyry Cu and Mo deposits in South Qinling and the southern margin of North China [9] also seem to support the above-mentioned interpretation.

Conclusions
Extremely anomalous Sr isotopic compositions, observed in granite porphyry rocks associated with the Late Mesozoic Shapinggou porphyry Mo deposit and their minerals, especially biotite and K-feldspar of high potassic contents, were caused mainly by extensive alteration of K-rich fluids. This explanation may hold for other Late Mesozoic porphyry Mo deposits in the eastern Qinling-Dabie orogenic belt.
Decoupling of Sr-Nd isotopic compositions provides evidence to determine geochemical features or the origin of hydrothermal fluids that have flushed porphyry rocks related to the Mo deposits. On the other hand, caution must be taken when using Sr isotopic composition as a tracer to assess the magmatic origin of porphyry rocks, particularly for those undergoing strong K-alteration by fluids.
Porphyry Mo and Cu deposits show distinguishable behaviours of trace elements and the Rb-Sr isotope system during the alteration of hydrothermal fluid, likely owing to differences in magma origins, hydrothermal alteration types, and rock-/ore-forming materials.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/min11080910/s1, Table S1: Major element contents of quartz syenite and granite porphyry in the Shapinggou porphyry Mo deposit, Table S2: Trace element contents of quartz syenite and granite porphyry, Table S3: Rare earth element contents of quartz syenite and granite porphyry, Table  S4: Whole rock Sr-Nd isotopic composition of quartz syenite and granite porphyry, Table S5: Trace element contents and Sr isotopic composition of minerals from quartz syenite and granite porphyry.
Author Contributions: X.X. and J.H. designed the study. J.H. and F.C. co-wrote the manuscript. J.H., Z.F., Y.A., T.C. and Q.X. carried out the field work and the analyses. All authors have read and agreed to the published version of the manuscript.