Chemical and Isotopic Composition of Sulfide Minerals from the Noho Hydrothermal Field in the Okinawa Trough

Abstract

Studies of the element contents and isotopic characteristics of sulfide minerals from seafloor hydrothermal sulfide deposits are a significant method of investigating seawater-fluid mixing and fluid-rock and/or sediment interactions in hydrothermal systems. The seafloor hydrothermal sulfide ores from the Noho hydrothermal field (NHF) in the Okinawa Trough (OT) consist of pyrrhotite, isocubanite, sphalerite, galena, and amorphous silica. The Rh, Ag, Sb, and Tl contents mostly increase in galena as the fluid temperature decreases in the late ore-forming stage. In the sulfide minerals, the rare earth elements are mainly derived from the hydrothermal fluids, while the volcanic rocks and/or sediments are the sources of the sulfur and lead in the sulfide minerals. After the precipitation of galena, the redox state becomes oxidizing, and the pH value of the fluid increases, which is accompanied by the formation of amorphous silica. Finally, neither pyrite nor marcasite has been observed in association with pyrrhotite in the NHF sulfides, likely indicating that the amount of sulfur was limited in this hydrothermal system, and most of the residual Fe was incorporated into the sphalerite. This suggests that the later pyrite and/or marcasite precipitation in the seafloor hydrothermal sulfide deposit is controlled by the sulfur content of the fluid. Furthermore, it is possible to use hydrothermal sulfides and their inclusions to trace subseafloor fluid circulation processes.

1. Introduction

The majority of seafloor hydrothermal sulfide deposits contain a mixture of sulfide minerals (e.g., pyrrhotite, pyrite, marcasite, chalcopyrite, isocubanite, sphalerite, wurtzite, and galena), secondary Cu sulfides (bornite, covellite, digenite, and chalcocite), sulfosalts (tennantite-tetrahedrite among many others) [1], sulfates (anhydrite and barite), and amorphous silica in different abundancies. For example, pyrrhotite, which contains Cu, Co, Mn, Ni, Sn, Pb, Cr, Ti, Ag, and Se [2,3,4], is commonly found as a major phase in many seafloor hydrothermal sulfide deposits (e.g., [5,6,7,8,9,10]). Its in-situ chemical composition and genesis have been investigated in several hydrothermal fields, including the Bent Hill field on the northern Juan de Fuca Ridge (JdFR) [11], the Broken Spur vent field on the Mid-Atlantic Ridge (MAR) [12], the Lucky Strike vent field on the MAR [13], and the Ashadze hydrothermal field on the MAR [14]. However, the trace element contents of pyrrhotite are variable [1]. It is known that the pyrrhotite from the southern JdFR has higher silver (Ag) contents (0.04–0.07 wt.%; [15]) than that from the Endeavour Segment of the JdFR (<0.04 wt.%; [16]) and the Rainbow hydrothermal field on the MAR (0.02 wt.%; [17]). In addition, the Cu content of pyrrhotite from the Endeavour Segment of the JdFR (0.065 wt.%; [16]) is higher than that from the southern JdFR (0.01–0.06 wt.%; [15]).

Sphalerite, also one of the most common minerals in seafloor hydrothermal sulfides and chimneys (e.g., [18,19,20,21]), contains Mn, Fe, Co, Ni, Cu, Ag, Sn, Pb, Hg, Cd, Ge, Ga, In, and Sb [22,23,24]. Its Cd content can be very high (200–11,900 ppm) [1]. Sphalerite associated with arc-related geologic settings generally contains higher levels of As (1000–10,000 s ppm), Pb (1000–10,000 s ppm), and Sb (100–1000 s ppm) than that in mid-ocean ridge (MOR) hydrothermal systems (10–100 s ppm) [1]. The FeS content of sphalerite can be useful in determining its formation temperature and the sulfur fugacity of the fluid in the chimneys in seafloor hydrothermal fields [20].

The presence of significant amounts of galena is characteristic of arc and back-arc basin (BAB) hydrothermal sulfide deposits [10,20,25,26,27]. Galena contains Cu, Ag, Sb, Bi, Cd, As, In, Mn, Tl, and Zn [28]. Particularly high Cd (800–1100 ppm) and Zn (2000–3400 ppm) contents have been measured in the galena from the Palinuro Volcanic Complex, Hook Ridge, and the Endeavour Segment of the JdFR [16,29]. Furthermore, the highest Sb (2160 ppm) and Ag (9400 ppm) contents in galena have been observed in arc-related seafloor hydrothermal fields [1].

However, the in-situ element contents and their distributions in sulfide minerals from the Noho hydrothermal field (NHF) in the Okinawa Trough (OT) are poorly documented. To reveal the formation processes and metal sources of the sulfide minerals in the seafloor hydrothermal systems, we studied and characterized the mineralogical, chemical, and isotopic compositions of the sulfide minerals from the seafloor hydrothermal sulfide deposits in the NHF.

2. Geological Setting

The OT is an active BAB located behind the Ryukyu trench–arc system, extending approximately 1200 km from Taiwan island to Kyushu Island in the western Pacific (Figure 1). It was formed by the active subduction of the Philippine Sea Plate under the Eurasian Plate in the Early Pleistocene (2–1.5 Ma; [30,31,32]). However, the OT is divided into the southern (SOT), middle (MOT), and northern (NOT) segments by the Kerama and Tokara faults [33]. The latest phase of rifting started <100 ka ago according to the ages of the zircon in the OT volcanic rocks, but the exact ages of the earlier rifting phases may be considerably older (108 Ma to 2.7 Ga) [34]. Furthermore, the SOT is characterized by a sedimentary cover that is much thinner (1–3 km thick) than the very thick cover in the NOT (~8 km) [31,35,36]. The OT volcanic rocks vary from tholeiitic basalt to calc-alkaline andesite to rhyolite [37], including basaltic andesite and dacite in the MOT [38].

However, the presence of high ³He contents in the seawater [40] and a high heat flow [41] suggest that there are numerous active seafloor hydrothermal fields in the MOT [42]. Indeed, several seafloor hydrothermal fields have been discovered in the MOT [43,44,45,46,47]. For example, the NHF (27°31.1′ N, 126°59.0′ E, 1581 m) was surveyed in 2016 during seafloor hydrothermal activity off the Iheya Ridge in the MOT using a television (TV)-log grabber (Figure 1) during the HOBAB4 cruise. The survey was conducted near the Noho site (27°31.3′ N, 126°59.1′ E, 1595 m), approximately 3 km SE of the Clam hydrothermal field in the Sakai field discovered by the Japan Oil, Gas, and Metals National Corporation (JOGMEC) ([48]; JOGMEC news release on 4 December 2014; NT15-13 cruise report available at http://www.godac.jamstec.go.jp/darwin/cruise/Natsushima/NT15-13/). The exact position of the Noho (JOGMEC) site had not yet been published at that time [42]. The NHF likely corresponded to the acoustic water column anomaly site [42]. It is mainly associated with basaltic and trachyandesitic rocks (H4-TVG5-2; Figure 1) which are located on the outside slope of the Iheya Ridge and are covered by several meters of pumice-rich sediments. The site includes active hydrothermal vents, chimneys, hydrothermal sulfide deposits, and chemosynthetic communities. However, there is a coupling relationship between the tectonic setting, magmatism, mineralization, fluid-rock interactions, and sedimentary processes in the NHF hydrothermal system and the response, adaptation, record, and activity of organisms. It is interesting to investigate the synergetic metallogenic mechanism of the magma, fluid, rocks, sediments, seawater, and organisms in the NHF hydrothermal field. In the future, it will be possible to use hydrothermal sulfide and its inclusions to trace the deep circulation processes of the fluids in the NHF. Furthermore, we can combine geophysical prospecting, geochemical exploration, and biological exploration techniques with drilling techniques to explore the seafloor polymetallic sulfides and hydrothermal vents, and we can use the four systems of exploration techniques to understand the resource potential of the seafloor polymetallic sulfide deposits in the NHF.

3. Sampling and Methods

Seafloor hydrothermal sulfide samples H4-TVG5-1-1 and H4-TVG5-3-1 were collected from the NHF using a TV grabber (Figure 1; Table S1). To determine the morphology, structure, and chemical compositions of the minerals, reflected and transmitted light microscopy, scanning electron microscopy (SEM), and energy dispersive spectrometry (EDS) were used to analyze thin sections of selected samples. Furthermore, the in-situ major and trace element contents of the samples were analyzed, as well as their S and Pb isotopic compositions.

3.1. SEM Analysis

Pyrrhotite, isocubanite, sphalerite, galena, and amorphous silica were studied using the TESCAN VEGA 3 LMH SEM with an Oxford INCA X-Max EDS at the Institute of Oceanology, Chinese Academy of Sciences, Qingdao, China. The standard analytical operating conditions included an accelerating voltage of 20 kV, a beam intensity of 15 nA, and a working distance of ~15 mm. The following standards were used for the EDS measurements and calibrations: (1) sulfides–pyrrhotite (Fe, S), galena (Pb, S), sphalerite (Zn, S), chalcopyrite (Cu, Fe, S), greenote (Cd), stibnite (Sb), and proustite (Ag, As); (2) silicates–olivine (Ni) and basalt glass (Ti, Si); (3) carbonates–calcite (O); (4) metals–cobalt (Co); (5) oxides–Cr-spinel (Mn); and (6) system standard library (Au, Se). The EDS detection limits for Au, S, Pb, Ag, Cd, Sb, O, Se, As, Zn, Cu, Ni, Co, Fe, Mn, and Ti were 0.1 wt.% (Table S2).

3.2. Major Element Analysis

The major element compositions of the pyrrhotite, sphalerite, and galena were analyzed using the JXA-8230 electron microprobe analyzer (EMPA) at the State Key Laboratory of Continental Dynamics, Northwest University, Xi’an, China. The instrument was operated with a 15 kV acceleration voltage, 10 nA beam current, and 2 μm beam diameter. The following reference materials were used for the wavelength-dispersive spectrometry measurements and calibrations: (1) metals–Co, Au, Se, Ni, Mn, and Ti and (2) natural sulfides–FeS₂, CuFeS₂, ZnS, PbS, Sb₂S₃, Ag₃AsS₃, and CdS. The counting time for the peak and background of Au, S, Pb, Ag, Cd, Sb, Se, As, Zn, Cu, Ni, Co, Fe, Mn, and Ti are 20–60 s and 15 s, respectively. The detection limits of these elements were 0.02–0.11 wt.% (Table S3).

3.3. Trace Element Analysis

The trace element contents of the pyrrhotite, sphalerite, and galena were determined using the laser ablation inductively coupled plasma mass spectrometer (LA-ICP-MS) at the State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Wuhan. A GeoLas Pro 193 nm ArF excimer laser was used to analyze the samples. The laser energy was 80 mJ, the frequency was 6 Hz, and the diameter ablation spot was 44 μm. The ion-signal intensities were acquired using an Agilent 7500 x ICP-MS instrument. Helium was used as the carrier gas which was mixed with argon via a T-connector before entering the ICP-MS. Each measurement incorporated approximately 30 s of background acquisition (gas blank) followed by 50 s of data acquisition from the sample. The ICPMSDataCal software was used to quantitatively calibrate the trace element contents [49]. Fe, Cu, Zn, and Pb were selected as the internal standards for the data reduction. Reference glass NIST 610 was used as the external standard, which was analyzed every 10 spots to monitor the instrument drift. The accuracy was determined with respect to reference glass NIST 610 and was assessed to be better than 10% (1σ).

3.4. In-Situ S Isotope Analysis

The in-situ S isotope compositions of the sphalerite, pyrrhotite, and galena were determined using double side-polished slices (DSPS) and the 193 nm femtosecond laser-ablation multi-collector ICP-MS (fs-LA-MC-ICP-MS) at the State Key Laboratory of Continental Dynamics (SKLCD), Northwest University, Xi’an, China. The equipment consisted of a 193 nm NWRfemto (RESOlution M-50, ASI) laser ablation system coupled with a Nu Plasma II MC-ICP-MS (NU Instruments, Ltd., Wrexham, UK). The ablated materials were transported into the plasma using He as the carrier gas. In a cyclone coaxial mixer, Ar gas was mixed with the carrier gas before being transported into the ICP’s torch. The energy fluence of the laser was approximately 3.5–4 J/cm². The beam diameter was 35 μm with a laser repletion rate of 3–4 Hz for a single spot analysis. Standard PSPT-3 (sphalerite; δ³⁴S_V-CDT = 26.5 ± 0.2‰) was used as the certified reference material. The obtained δ³⁴S_V-CDT of Standard PSPT-3 in this study is 26.36 ± 0.29‰ (2SD; n = 21; Table S4), which is consistent with the recommended values (δ³⁴S_V-CDT = 26.5 ± 0.2‰; [50]). The detailed analysis parameters are described in Bao et al. [50], Chen et al. [51], and Yuan et al. [52].

3.5. In-Situ Pb Isotope Analysis

The in-situ Pb isotope compositions of the galena were determined using DSPSs and the fs-LA-MC-ICP-MS at the SKLCD. To remove potential contamination, anhydrous ethanol was used to carefully clean the surfaces of the DSPSs prior to the laser ablation analysis. The analytical spots were carefully selected to prevent the possible influence of inclusions and impurities. High temperature-activated carbon was used to filter the Hg contained in the carrier gas, which lowered both the Hg background and the detection limit. Internal Tl isotope reference NIST SRM997 was used in conjunction with external reference NIST SRM 610 to correct for fractionation and mass discrimination effects. The Tl solution was introduced through the CETAC Aridus II desolvation nebulizer system. The exponential law correction method for Tl normalization with an optimally adjusted Tl ratio was used to obtain Pb isotope data with high precision and accuracy [53]. The measured isotope ratios were in agreement with the reference and the published values within 2σ measurement uncertainties [50,52,53,54]. Standard CBI-3 (Natural galena) was used as certified reference material. The measured Pb isotope ratios of standard CBI-3 in this study (²⁰⁸Pb/²⁰⁶Pb = 2.1246 ± 0.0010, ²⁰⁷Pb/²⁰⁶Pb = 0.8696 ± 0.00006, ²⁰⁶Pb/²⁰⁴Pb = 17.963 ± 0.0017, ²⁰⁷Pb/²⁰⁴Pb = 15.621 ± 0.0025, and ²⁰⁸Pb/²⁰⁴Pb = 38.165 ± 0.0048; SD; N = 5; Table S4) were consistent with the recommended values (²⁰⁸Pb/²⁰⁶Pb = 2.1247 ± 0.0000, ²⁰⁷Pb/²⁰⁶Pb = 0.8694 ± 0.00001, ²⁰⁶Pb/²⁰⁴Pb = 17.964 ± 0.0014, ²⁰⁷Pb/²⁰⁴Pb = 15.622 ± 0.0013, and ²⁰⁸Pb/²⁰⁴Pb = 38.168 ± 0.0028; [52]).

4. Results

4.1. Mineralogy of Hydrothermal Sulfide Samples

Seafloor hydrothermal sulfide ore samples H4-TVG5-1-1 (~8 cm in size) and H4-TVG5-3-1 (~24 cm in size) are black to dark grey in color and contained pores up to 0.2 mm (H4-TVG5-1-1) and 1 mm (H4-TVG5-3-1) in diameter. They are composed of sulfide minerals (Figure 2), i.e., pyrrhotite, galena, sphalerite, and isocubanite, as well as amorphous silica. The coarser-grained aggregates of pyrrhotite (up to 100 μm), sphalerite (up to 50 μm), galena (up to 30 μm), and isocubanite (up to 20 μm) comprise >70% of the sulfide samples by volume in both samples (Figure 2 and Figure 3).

4.1.1. Pyrrhotite

Pyrrhotite is the most abundant mineral. It occurs as randomly oriented laths that range in size from a few to hundreds of micrometers. The laths form boxwork textures (Figure 2) shaped similar to blades or hexagonal plates (Figure 2 and Figure 4a). Some of the pyrrhotite boxwork laths contain interstitial sphalerite, minor isocubanite, and late galena (Figure 2). The whitish rim around pyrrhotite is likely a thin oxidized pyrrhotite-rim, similar to the rims commonly observed in pyrrhotite from other hydrothermal fields (Figure 2f).

4.1.2. Sphalerite

The sphalerite from the NHF is red and Fe-rich (Table S3). It is intimately connected with the pyrrhotite and galena (Figure 2 and Figure 4), intergrown with the pyrrhotite and isocubanite (Figure 2h), overgrown by amorphous silica, and contains local inclusions of galena. The sphalerite commonly occurs as hydrothermal aggregates that are intimately intergrown within the interstices between the pyrrhotite laths where it forms rounded or elongated anhedral grains up to several tens of micrometers across (Figure 2f).

4.1.3. Galena

Galena is the major sulfide mineral in both samples (Figure 2 and Figure 4). It is disseminated throughout the samples, appearing as subhedral or anhedral crystals (Figure 2), and ranges in size from a few to tens of micrometers (Figure 2 and Figure 4). Most of the drop-like, oval, or elongated inclusions of galena crystals up to 25 μm in size are found in the sphalerite (Figure 2d). This indicates that there is a positive correlation between the Zn and Pb in the sulfide minerals [10]. The galena is often surrounded by amorphous silica (Figure 2e). It is also found as intergrowths of minute (5–40 μm) euhedral to subhedral crystals in the pores within the pyrrhotite–sphalerite aggregates (Figure 2h).

4.1.4. Isocubanite and Amorphous Silica

The isocubanite coexists with the sphalerite and galena in hydrothermal sulfide samples H4-TVG5-1-1 and H4-TVG5-3-1 (Figure 2). When present, amorphous silica occurs as a thin late-stage rim around the sulfide minerals and may overgrow the sphalerite, pyrrhotite, and galena (Figure 2).

4.2. Major and Trace Element Contents of Sulfide Minerals

4.2.1. Major Element Contents of Sulfide Minerals

The major element concentrations of the sulfide minerals are presented in Tables S2 and S3. According to Belzile et al. [55] and Fallon et al. [1], pyrrhotite can be monoclinic (less Fe-deficient: 45.9–46.8 mol% Fe) or hexagonal (more Fe-deficient: 47.2–48.0 mol% Fe) (Table S2). Most sphalerite has low Cu contents (0.10–0.46 wt.%) and very high Fe (12.46–19.53 wt.%) and Mn (0.10–6.27 wt.%) contents (Figure 4; Table S2). The FeS contents of the sphalerite slabs range from 21.95 to 34.78 mol% (Table S2). In contrast, the Fe contents (1.09 to 2.61 wt.%) of the FeS-bearing sphalerite from the sulfide chimneys in the North Konll, Iheya Ridge, are reportedly very low (~4 mol%; Ueno et al. [20]). The major elements in the galena are Pb (83.68–87.17 wt.%) and S (13.31–13.69 wt.%) (Tables S2 and S3), and those in the isocubanite are Zn (0.80–2.12 wt.%) and Pb (0.38–2.35 wt.%) (Table S2), while most of the amorphous silica contains significant amounts of Fe (0.46–9.64 wt.%) and Zn (0.13–4.86 wt.%) (Table S2).

4.2.2. Trace Element Contents of Sulfide Minerals

The trace element concentrations of the sulfide minerals are presented in Table S5. The pyrrhotite is relatively enriched in Ge (28.0–97.3 ppm) and Mo (1.21–3.40 ppm). The Ge content (28.0 to 97.3 ppm) of the pyrrhotite samples (H4-TVG5-1-1) from the NHF are noticeably higher than those of the galena (0.51–1.09 ppm) and sphalerite (3.63–4.84 ppm) crystals (Table S5; Figure 5).

The sphalerite contains Cd (318–388 ppm), Sn (145–182 ppm), Au (0.67–0.82 ppm), Th (4.85–5.09 ppm), Mo (0.08–0.34 ppm), Tl (0.11–0.67 ppm), and Bi (0.03–0.15 ppm). The Th contents of the sphalerite samples from the NHF are usually higher than those of the pyrrhotite and galena from sample H4-TVG5-1-1 (Table S5; Figure 5).

The Mo (0.07–2.44 ppm), Cd (3.11–70.1 ppm), V (0.41–22.3 ppm), Cr (2.82–369 ppm), Ni (0.45–6.48 ppm), and Cu (12.9–4420 ppm) contents of the galena from the NHF are much more variable than those of the pyrrhotite and sphalerite, covering the largest range (Tables S2, S3 and S5; Figure 5). The As (1.74–2233 ppm), Se (17.0–109 ppm), Fe (3128–101,010 ppm), and Zn (829–127,104 ppm) contents of the galena samples from the NHF also span large ranges (Tables S2, S3 and S5). Se is concentrated in the galena (17.0–109 ppm) and pyrrhotite (13.2–52.7 ppm) (Table S5), whereas Tl was detected in the galena (0.49–4.09 ppm), pyrrhotite (0.12–0.64 ppm), and sphalerite (0.11–0.67 ppm) (Table S5). The galena is more enriched in Ag (693–1940 ppm) and Sb (1608–5129 ppm) than the pyrrhotite and sphalerite (Table S5; Figure 5). In particular, the Ag contents (693–1940 ppm) of most of the galena samples are significantly higher than those of the pyrrhotite (0.49–9.28 ppm) and sphalerite (42.4–53.9 ppm) samples (Figure 5). There is a positive correlation between the Ag and Bi contents of the galena. Most of the pyrrhotite (Th 0.08–2.38 ppm) and sphalerite (Th 4.85–5.09 ppm) contain more Th than the galena (Th 0.04–2.37 ppm), while the galena (7.52–16.8 ppm) has significantly higher Rh contents than the pyrrhotite (0.08–0.14 ppm) and sphalerite (0.04–0.05 ppm) (Table S5; Figure 5). Furthermore, a significant positive correlation was observed between the Au and Mo contents of the galena from the NHF.

4.3. Rare Earth Element Compositions of Sulfide Minerals

The rare earth element (REE) contents were normalized to Chondrite values (subscript CN) [56]. The Eu and Ce anomalies were assessed using the following equations: (Eu/Eu*)_CN = (2 × Eu_CN)/(Sm_CN + Gd_CN) and (Ce/Ce*)_CN = (2 × Ce_CN)/(La_CN + Pr_CN). The REE data for the Yonaguni Knoll IV vent fluids, Iheya Ridge volcanic rocks, and MOT sediments used for comparison are from Hongo et al. [57], Li et al. [39], and Xu et al. [58], respectively.

The total REE contents (∑REEs) of the pyrrhotite, sphalerite, and galena from the NHF are highly variable (2.01–36.7 ppm) (Table S5). Most of the ∑REE contents of the pyrrhotite (2.32–36.5 ppm) and galena (2.01–36.7 ppm) are higher than those of the sphalerite (10.2–13.2 ppm) (Table S5). Among the studied samples, the pyrrhotite from the NHF had the highest ∑REE contents (36.5 ppm, H4-TVG5-1-1, point 2–3) (Table S5).

The Chondrite-normalized REE distribution patterns of the pyrrhotite, sphalerite, and galena from the NHF are presented in Figure 6. The REE patterns of most of the sulfide minerals exhibit light rare earth element (LREE) enrichment (LREE/HREE = 8.09–32.0), variable La_CN/Lu_CN ratios (0.98 and 50.0), and Eu ((Eu/Eu*)_CN = 0.47–4.13) and Ce ((Ce/Ce*)_CN = 1.70–3.85) anomalies (Figure 6a–c). Compared with the Eu anomalies of the pyrrhotite ((Eu/Eu*)_CN = 1.44–1.68) and sphalerite ((Eu/Eu*)_CN = 3.61–4.13) from the NHF, the galena has smaller Eu anomalies ((Eu/Eu*)_CN = 0.47–0.83). Moreover, most HREE contents are around the detection limit, which has been marked in dash lines (Figure 6) and were only used to qualitatively understand their low contents.

4.4. In-Situ S and Pb Isotopic Compositions of the Sulfide Minerals

Five sulfur isotope analyses were conducted on the sphalerite, pyrrhotite, and galena grains from sample H4-TVG5-1-1. The δ³⁴S values obtained for the sulfide minerals are consistent and range from 3.58 to 5.69‰. One sphalerite sample from hydrothermal sulfide sample H4-TVG5-1-1 yielded the highest δ³⁴S values (5.69‰). The galena and pyrrhotite are characterized by lower sulfur isotope values (<4.16‰). The lowest δ³⁴S value (3.58‰) was for a galena sample (Table S6).

The galena from sample H4-TVG5-1-1 from the NHF has uniform Pb isotopic compositions, with a narrow range of Pb isotopic values (Table S6; Figure 7). The ²⁰⁷Pb/²⁰⁴Pb and ²⁰⁸Pb/²⁰⁴Pb ratios of the galena from the NHF are within or close to the range of values for the Iheya Ridge volcanic rocks (basalt and trachyandesite) and the JADE sediments in the MOT (Figure 7).

5. Discussion

5.1. Seawater–Fluid Mixing and Element Enrichments

The NHF sulfide minerals often contain well-developed pyrrhotite blades, xenomorphic isocubanite, interstitial sphalerite, and sharp-edged galena crystals, which are the result of rapid cooling during seawater–fluid mixing (Figure 2) [10]. Mo-enriched sulfide is characteristic of high-temperature paragenetic mineral associations [61]. The Mo contents (1.21–3.40 ppm) of the pyrrhotite samples from the NHF are mainly higher than those of the sphalerite (0.08–0.34 ppm) and galena (0.07–2.44 ppm) in sample H4-TVG5-1-1 (Table S5; Figure 5). This suggests that the Mo-enrichment of the pyrrhotite may have occurred at high temperatures.

The sphalerite grains from the NHF are Fe-rich (Tables S2 and S3); their FeS mol% contents are significantly higher than those from the PACMANUS hydrothermal field (0.00–2.22) in the eastern Manus Basin, SW Pacific [62]. These values are identical to previously reported values [63] for Vienna Woods (3.35–10.65) and sediment-poor JADE (0.00–1.93) samples and are consistent with those (23.48–50.81) reported for the sediment-rich Hakurei samples from the OT [62]. Such FeS contents indicate low-sulfidation conditions [64]. The sphalerite from the NHF is enriched in Th compared to the pyrrhotite and galena (Table S5; Figure 5). This suggests that the abundance of sphalerite controls the Th enrichment in seafloor hydrothermal sulfides. The Th/U ratios (1.86–4.76) of the pyrrhotite from the NHF are consistent with those of the Iheya Ridge basalt (2.64; Li et al. [39]), trachyandesite (4.13; Li et al. [39]), and the MOT metalliferous sediments (0.18–6.11; Yang et al. [65]), so pyrrhotite serves as the source of the Th and U in the subseafloor hydrothermal fluids. Furthermore, the Ge contents (0.51–97.3 ppm) of the sulfide minerals from the NHF in the OT are higher than those (Ge 0.0055 ppm) of seawater [59]. The positive correlation between the Ge and Pd contents indicates an affinity between these elements and the pyrrhotite (Figure 8a).

The galena from the NHF is commonly associated with sphalerite and was observed to form partial rims around the sphalerite crystals (Figure 2d). The galena inclusions in the sphalerite indicate that the galena post-dates the sphalerite (Keith et al., 2016), and it may have formed through the following reaction: ZnS (s) + Pb²⁺ (aq) → PbS (s) + Zn²⁺ (aq). This would result in relatively Zn-rich galena (up to 1.83 wt.%) (Tables S2 and S3) and needs to be confirmed in a future study. However, this indicates a varying fluid chemistry and a change in the solubility as the temperature decreases or increased dilution during seawater–fluid mixing [66]. In the latter case, pyrrhotite would form later than sphalerite/galena, since the fluid is in the interior of both samples H4-TVG5-1-1 and H4-TVG5-3-1.

The Ag contents (0.49–9.28 ppm) of the pyrrhotite from sample H4-TVG5-1-1 (Table S5) are lower than those from the Rainbow hydrothermal vent field on the MAR (200 ppm), the southern JdFR (400–700 ppm), and the Endeavour Segment of the JdFR (<400 ppm) [15,16,17]. In addition, the Ag contents exhibit a positive correlation with the higher Bi contents (Figure 8c).

The Rh (7.52–16.8 ppm), Bi (0.14–1.38 ppm), Ag (693–1940 ppm), Tl (0.49–4.09 ppm), and Sb (1608–5129 ppm) contents of the galena from the NHF are variable and are noticeably higher than the Rh contents of the sulfides from the PACMANUS hydrothermal field (0.0004–0.3556 ppm) in the eastern Manus Basin, southwestern (SW) Pacific [67], the Turtle Pits, and the Logatchev hydrothermal field on the MAR (0.0012–0.149 ppm) [68]. The values of the galena from the NHF suggest that the chemical compositions of the fluids changed with time during the galena deposition. These values are also significantly higher than those of the sphalerite and pyrrhotite from the NHF (Table S5; Figure 5), the vent fluid (Bi = 0.000647 ppm) in the OT [69], and seawater (Bi = 0.00003 ppm, Ag = 0.002 ppm, Tl = 0.013 ppm, Sb = 0.2 ppm) [59]. Furthermore, the enrichment of Rh (7.52–16.8 ppm), Bi (0.14–1.38 ppm), Ag (693–1940 ppm), Tl (0.49–4.09 ppm), and Sb (1608–5129 ppm) in the galena may be the result of the galena precipitating later from late-stage fluids and rapid cooling due to seawater–fluid mixing (Table S5; Figure 5). However, Rh is more compatible in a monosulfide-solid solution (MSS) than in an intermediate solid solution (ISS) and has been found to be more soluble than Pt and Pd [70]. Due to these properties, Rh may crystallize as chalcogen (Se, Te, As, Sb, Bi)-rich discrete precipitate (e.g., RhSeS or RhSbS) before or synchronously with other chalcogen-bearing phases that are associated with late-stage low-sulfur precious metal haloes around hydrothermal sulfides. Similar associations have been documented at various locations around the Sudbury metallogenic deposit, Ontario [71,72]. Consequently, the Ag content is positively correlated with the Sb content for the galena from the NHF and in neighboring sphalerite (Figure 8d), which may also indicate the presence of Sb with the Ag incorporated into the galena from the sphalerite during mineralization. The Rh (7.52–16.8 ppm), Bi (0.14–1.38 ppm), Ag (693–1490 ppm), Tl (0.49–4.09 ppm), and Sb (1608–5129 ppm) contents of the galena are higher than those of the original hydrothermal sphalerite and the earlier pyrrhotite in the NHF (Table S5; Figure 5), which indicates that the incorporation of these trace elements into the galena, pyrrhotite, and sphalerite is controlled by their partition coefficients. This implies that the back-arc galena-rich hydrothermal sulfide deposit may be an Rh-Ag-Sb enriched deposit.

Compared with the PACMANUS hydrothermal mineralization (average Se content of pyrite is 5.97–7.39 ppm; [1]), most of the Noho sulfide minerals are enriched in Se (13.2–109 ppm) (Table S5), which was concentrated in the galena and pyrrhotite via seawater–fluid mixing (Table S5). This is similar to the seafloor hydrothermal mineralization in the Kulo Lasi volcano, Futuna, and SW Pacific. This implies that the Noho basalt-, trachyandesite-, and sediment-hosted hydrothermal mineralization may be enriched in Se [10] based on the fact that it was found in large quantities in the pyrrhotite and based on the reducing conditions during the fluid-rock and/or sediment interaction.

In addition, the Pb contents of the pyrrhotite (up to 1.33 wt.%) and sphalerite (up to 2.93 wt.%) are comparable to other BAB mineralization in the PACMANUS field (0.176 wt.% to 2.35 wt.% Pb in sphalerite) in the SW Pacific [73,74], but they are higher than those in the Trans-Atlantic Geotraverse (TAG) hydrothermal field (<0.1 to 0.16 wt.% Pb in sphalerite) on the MAR [1,75]. However, some of the sphalerite grains exhibit noticeably high Pb contents (up to 2.93 wt.%) (Table S2) which may indicate spectral contamination from nearby galena grains, or true chemical contamination, caused by the diffusion of the chemical components across grain boundaries. The seafloor pyrrhotite (Fe = 58.5–62.0 wt.%; Pb = 0.02–1.33 wt.%; S = 37.8–40.9 wt.%; Zn = 632–5300 ppm; Cu = 130–529 ppm) and galena (Fe = 0.10–1.32 wt.%; Pb = 81.1–88.8 wt.%; S = 12.7–14.5 wt.%; Zn = 829–127,104 ppm; Cu = 12.9–4420 ppm) from the NHF contain different proportions of major (Fe, Pb, and S) and minor (Cu and Zn) elements which may also indicate probable spectral contamination from the adjacent sphalerite and subsurface Cu-sulfide inclusions (isocubanite, up to 30.83 wt.% Cu) (Table S2).

5.2. Variable REE Contents and Sources of REEs

The ∑REEs contents of the pyrrhotite analyzed in this study vary significantly (2.01–36.7 ppm), and they do not exhibit systematic variations with the Fe (58.5–59.8 wt.%) and S (39.5–40.3 wt.%) contents. The degree of LREE and heavy rare earth element (HREE) fractionation in the NHF pyrrhotite (LREE/HREE) is highly variable, up to 32.0 (Table S5). The ∑REE contents and ranges of the pyrrhotite (Table S5) exceed those of the sphalerite. The substitutions of REEs into the pyrrhotite are similar to those involving galena and sphalerite [76,77], indicating the significant influence of the REEs’ larger ionic radii.

The REE patterns of the pyrrhotite and sphalerite from the NHF contain positive Eu anomalies (1.44–4.13) (Table S5; Figure 6), similar to the values for the vent fluids in the OT [57,78] and almost all hydrothermal fields globally. These similarities indicate that the pyrrhotite and sphalerite inherited the positive Eu anomalies of the vent fluids [77,79,80,81]. The stability of the soluble Eu²⁺ species increases in contact with Cl^– complexation, low- to high-temperature acidic fluids, and under reducing conditions [77,82,83,84,85,86]. Therefore, the positive Eu anomalies of the pyrrhotite and sphalerite with high Eu content (0.04 to 0.27 ppm) may have been induced by Cl-complexation under high-temperature, low-pH, and strongly reducing fluid conditions [77,81].

The REE patterns of the galena from the NHF are characterized by negative Eu anomalies (0.47–0.83) (Table S5; Figure 6c), which are indicative of relatively low-temperature seawater [87]. A large proportion of the Eu in the low-temperature fluid is trivalent because divalent Eu is stable at temperatures of greater than ~250 °C [83]. The decreasing formation temperature of galena strongly correlates with the decreasing Eu²⁺/Eu³⁺ ratios of the vent fluids [77]. The accumulation of Eu²⁺ in galena formed at medium to low temperatures is also reduced. The stability of the soluble Eu²⁺ species has been reported to decrease in association with SO₄²⁻ complexation, high- to low-temperature acidic fluids, and under reducing conditions [83]. Therefore, the negative Eu anomaly of the galena were most likely induced by lower Eu contents, medium (300–200 °C) to low (<200 °C) temperatures (Figure 3), and/or mixing between fluids and seawater and the galena chemistry. Consequently, the Eu content of the fluids may have influenced the Eu anomaly in the precipitated galena and its depositional conditions and processes. The negative Eu anomalies are related to the lower Eu contents of the galena (0.03–0.14 ppm) (Table S5), which have been interpreted to have formed at medium to low temperatures and from less concentrated Eu-bearing- fluids [80,81,88].

The REEs in the sulfide minerals from the NHF may indicate the sources and evolution of the seafloor hydrothermal fluids. Previous studies of seafloor hydrothermal sulfides from the Rainbow, Broken Spur, and TAG hydrothermal fields on the MAR have revealed that the REEs were mainly derived from the fluids [79,89] and were incorporated into the sulfide minerals during seawater–fluid mixing (e.g., Bau and Dulski [90] and Zeng et al. [81]). The REE patterns of the sulfide minerals from the NHF are comparable to those of the Yonaguni Knoll IV vent fluids (Figure 6d) [57]. Thus, the REEs in the sulfide minerals from the NHF were most likely derived from fluids which leached the REEs from the local sub-seafloor volcanic rocks and/or sediments and deposited them in the Noho hydrothermal sulfide deposits (Figure 6d) [58,81,91]. Admittedly, more detailed works focusing on how REE enters sulfides need to be conducted in the future.

5.3. Sources of Sulfur in Sulfide Minerals

The sulfur isotope compositions of the sulfides and their possible geological reservoirs are important indicators that constrain the sources of the reduced sulfur [92,93]. The NHF hydrothermal sulfide deposit is hosted in volcanic rocks and sediments. Igneous rock-derived sulfur, including that in the MORBs, mantle peridotites, and calc-alkaline volcanic rocks (e.g., andesites or rhyolites; Zeng et al. [21]) show δ³⁴S values very close to 0‰ (+0.1 ± 0.5‰) [94,95,96,97]. However, the δ³⁴S values of the Quaternary volcanic rocks from the Japanese Island Arc (+4.4 ± 2.1‰; Ueda and Sakai [98]) and the volcanic rocks from the northern Mariana Trough (+2.0 to +20.7‰; Woodhead et al. [99]) are significantly higher. Although volcanic sulfur is rapidly leached by high-temperature (300–500 °C) fluids [100], no significant isotopic fractionation occurs during the leaching, transport, and reprecipitation of sulfide minerals [101]. The seawater sulfur in the NHF was most likely derived from seawater and/or fluids that leached sulfur from the sediments. Corresponding to volcanic sulfur, seawater sulfate has a δ³⁴S value of +21‰ [102] and H₂S-containing seafloor hydrothermal fluids related to sulfide and sulfate mineralization usually have δ³⁴S values of 1.5‰ to 7‰ (e.g., Michard et al. [103]; Butterfield et al. [104]; Shanks et al. [96]; and Shanks [105]). These values are between the δ³⁴S values of volcanic rocks (+0.1 ± 0.5‰; [94,95,96,97], the Japanese Island Arc volcanic rocks (+4.4 ± 2.1‰; Ueda and Sakai [98]), the northern Mariana Trough volcanic rocks (+2.0 to +20.7‰; Woodhead et al. [99]), and sediments (δ³⁴S values similar to those of seawater sulfate, which can be affected by microorganisms >+21‰) and seawater (+21‰). Therefore, the sulfur isotopic compositions of sulfide minerals can serve as evidence of the sources of the sulfur [21,106].

The δ³⁴S values of the SO₄ and H₂S from the OT vent fluids range from 20.6‰ to 25.7‰ (average of 22.4‰, n = 9) and –0.2‰ to 12‰ (average of 5.9‰, n = 26), respectively [69,78,107,108,109,110,111]. The δ³⁴S values of the sulfide mineral samples from the NHF are all within a narrow range (3.58–5.69‰; Table S6) which overlaps with the overall range for sulfide minerals from global seafloor hydrothermal fields (0.0–9.6‰; n = 1901; Zeng et al. [21]). This indicates a heavy sulfur contribution from reduced seawater sulfate, falling within the range between the seawater δ³⁴S values for SO₄ (20.99‰; [102]) and vent fluid H₂S (−0.2 to 12‰; [69,78,107,108,109,110,111]).

The sulfur isotope compositions of the sulfide minerals from the NHF may have been influenced by the mixing of seawater and/or sediment SO₄ with volcanic rock-derived sulfur during the sulfide formation according to the following reaction:

M²⁺ (aq) + SO₄²⁻ (aq) + 4H₂S (aq) → MSx (s) + yS (s) + 4H₂O (aq)

where M²⁺ is Fe²⁺, Pb²⁺, or Zn²⁺; and MS is pyrite (x = 2, y = 3), galena (x = 1, y = 4), or sphalerite (x = 1, y = 4).

According to a simple two end-member mixing model,

(δ³⁴Smix = X × δ³⁴S_{seawater and/or sediment} SO₄ + (1 − X) × δ³⁴S_{volcanic rock}

where X is the amounts of seawater and/or the sediment SO₄ component, and δ³⁴S_mix, δ³⁴S_{seawater and/or sediment} SO₄ (21‰), and δ³⁴S_{volcanic rock} (0.1‰, volcanic rock-derived sulfur) are the sulfur isotope compositions of sulfide minerals or vent fluids, seawater and/or sediment SO₄ and volcanic rock-derived sulfur, respectively, the sulfur isotope composition of sulfide minerals can be obtained via mixing sulfur from seawater and/or sediment SO₄ (17–19%, n = 5) with sulfur from volcanic rocks (81–83%, n = 5). This mixing likely occurred in the downwelling limb of the hydrothermal convection cell in the NHF. This further suggests that the proportions of the sulfur in NHF sulfide minerals contributed by seawater and/or sediment SO₄ were less than that contributed by the volcanic rock-derived sulfur. The sulfide minerals in sample H4-TVG5-1-1 precipitated in the NHF primarily where the hydrothermal fluids are dominant along the tube interiors. In addition, disulfides and sulfates tend to be more abundant where mixing with seawater promotes oxidation of S²⁻ in the distal portions of the sample. The δ³⁴S values of the sphalerite from the NHF are higher than those of the galena and pyrrhotite from the NHF (Table S6), and they are significantly lower than those of the sphalerite from the JADE hydrothermal field which has been reported to be +5 to +6‰ [20] and +6.6 to +7.1‰ [112]. These values have been significantly influenced by the reduction in seawater and/or sediment sulfate under highly oxidizing fluid conditions during sphalerite formation [113,114].

In addition, the decrease in the δ³⁴S values from the sphalerite to the galena [115] suggests that the sphalerite-galena pair (point 8-9-3 for sphalerite and point 8-9-5 for galena) in the NHF represent sulfur isotopic equilibrium. Based on the known temperature-dependent fractionation factors [93,116], the formation temperature of the coexisting sphalerite-galena pair was estimated to be 314 °C and 318 °C using the equation for the sulfur isotope geothermometer (Table S6).

5.4. Fluid-Rock Interactions and Source of Lead in Galena

The majority of the Pb isotope ratios of the galena plot were in or near the fields of the basalts and andesites from the NHF and the JADE sediments from the MOT (Figure 7). This suggests that the MOT volcanic rocks and/or sediments are the main source of the Pb in the seafloor vent fluids (e.g., Vidal and Clauer [117]; Chen [118]; Hegner and Tatsumoto [119]; Hinkley and Tatsumoto [120]; Fouquet and Marcoux [121]; Halbach et al. [38]; Charlou et al. [122]; Yao et al. [123]; and Zeng et al. [21]). The range of the Pb isotope compositions of the galena (Table S6) is small compared to those of the volcanic rocks and sediments in the OT. The Pb isotope ratios of the galena from the NHF are the same as those of the volcanic rocks and sediments in the OT, which indicates that the heterogeneous Pb isotopes (e.g., Allègre et al. [124] and Hamelin et al. [125]) were extracted from the volcanic rocks and/or sediments via fluid-rock and/or sediment interactions and were homogenized during fluid circulation, supporting the conclusions of previous studies [21,121,126]. This indicates that the Pb isotope compositions of the galena can be used to infer the Pb isotope compositions of the local volcanic rocks, sediments, and fluids. The Pb isotope ratios of the galena from the NHF are close to the average Pb isotope ratios of the volcanic rocks and sediments in the MOT (Figure 7b), suggesting that the volcanic rock and/or sediments were likely the sources of the Pb in the galena from the NHF. In addition, these ratios plot in or near the field of the west Philippine Sea sediments and altered oceanic crust basalts (Figure 7b) [60], implying that sedimentary and subducted oceanic crust components were a possible source of the Pb in the galena from the MOT.

5.5. Hydrothermal Temperature and Redox Conditions during Sulfide Formation

The equilibrium sulfur isotope temperature (314 °C and 318 °C) of the sphalerite-galena pair (Table S6) suggests that the dominance of sphalerite is a characteristic of the Noho hydrothermal sulfides with formation temperatures of >300 °C. Based on the observed microtextural relationships of the minerals and the interpreted paragenetic sequences (Figure 3), the sulfide mineral crystallization sequence was as follows. First, the pyrrhotite formed, followed by isocubanite and sphalerite. In the final stage, galena, and then amorphous silica is precipitated. The amorphous silica is concluded to have formed during the late stage since it precipitates at low temperatures (i.e., from ~50 °C to <200 °C; Figure 3) [10]. The overall mineral crystallization sequence from pyrrhotite to isocubanite to sphalerite to galena and finally to amorphous silica is indicative of hot, reduced hydrothermal fluids mixing with cooler, more oxygenated seawater [18]. This also emphasizes the importance of the interactions between the mineralizing fluid and the coexisting seawater during the formation of this hydrothermal sulfide deposit in the NHF. However, replacement of fine-grained sulfides with coarse-crystalline sulfides indicates a reaction with hotter and, at least in part, more reduced ascending fluids from which pyrrhotite and sphalerite may also have precipitate [127,128]. In addition, the decrease in the Fe/S ratio of the pyrrhotite from 0.96 to 0.92 (Table S2) indicates an increase in the level of pyrrhotite oxidation (fO₂ increase or fS₂ decrease) [1].

Interestingly, no pyrite or marcasite was observed in samples H4-TVG5-1-1 or H4-TVG5-3-1. The lack of these minerals sheds important light on the formation processes of the seafloor hydrothermal sulfides in the NHF. For instance, it is known that pyrite is much less soluble than pyrrhotite (approximately 6–10 orders of magnitude less soluble at 200–250 °C; [129]). Just after the pyrrhotite precipitated, the residual Fe precipitated to form Fe-rich sphalerite. This did not leave enough sulfur to form abundant pyrite and/or marcasite. Thus, the lack of pyrite and/or marcasite when pyrrhotite is present merely indicates low sulfur activity. However, according to the mineral paragenesis, the pyrrhotite, isocubanite, and sphalerite were formed during the early ore-forming stage, while the sphalerite, galena, and amorphous silica were formed during the late stage. The pyrrhotite, which represents the earliest mineralization stage, is occasionally overgrown by sphalerite and late-stage galena (Figure 2 and Figure 3).

Even so, the accurate sulfide mineralizations model requires drilling samples and data. Therefore, our TV-grabber collected samples are focused on exploring the element sources (e.g., REE, S, and Pb) and physicochemical conditions (e.g., temperature, fO_2, and fS₂) of the sulfide mineral formation.

6. Conclusions

The mineral assemblage of the seafloor hydrothermal sulfide ores from the NHF is dominated by pyrrhotite, sphalerite, galena, minor isocubanite, and amorphous silica. The early sulfides (pyrrhotite, isocubanite, and sphalerite) are partly or completely surrounded by the later minerals (sphalerite, galena, and amorphous silica). The late overgrowth of the pyrrhotite by sphalerite and the sphalerite by galena indicates an unstable hydrothermal system.

Most of the measured Rh, Bi, Ag, Tl, Se, and Sb contents of the pyrrhotite and sphalerite are significantly lower than those of the galena, indicating enrichment of these trace elements in the galena during the seawater–fluid mixing. This suggests that the galena-rich hydrothermal sulfide deposit in this BAB may be Rh-Ag-Sb-enriched.

The REEs in the sulfide minerals from the NHF were likely sourced from the seafloor hydrothermal fluids. The REE contents and patterns of the sulfide minerals are related to the mineral chemistry, but they were also influenced by the physiochemical compositions, REE contents and patterns of the fluids, the degree of mixing between the fluids and seawater, and interactions with sub-seafloor rocks and/or sediments. The pyrrhotite and the sphalerite precipitated at higher temperatures in association with acidic or reducing fluids have positive Eu anomalies. Thus, the Eu anomaly values may indicate the Eu content and temperature of the source fluids.

The sphalerite has higher δ³⁴S values than the pyrrhotite or galena. The higher δ³⁴S values of the sulfide minerals from the NHF indicate a heavy sulfur contribution from reduced seawater and/or sediment sulfate, which may indicate that the sulfur in the sulfides was mainly derived from the volcanic rocks.

The Pb isotope compositions of the galena are similar to those of the associated volcanic rocks and/or sediments in the NHF and the JADE sediments in the MOT, suggesting that the galena inherited the Pb isotope compositions of the host rocks in the Noho sub-seafloor hydrothermal systems during the galena formation via seawater-fluid mixing and fluid-rock and/or sediment interactions. The Pb isotope compositions of the galena from the NHF are very homogenous and have a narrow range, plotting in or near the field of the large Pb isotope dataset for the MOT volcanic rocks, western Philippine Sea sediments, and altered oceanic crust basalts. This implies that the isotopic composition of the Pb in the galena is important for understanding the influence of plate subduction on back-arc hydrothermal and magmatic systems.

The formation temperature of the coexisting sphalerite and galena in the NHF was estimated to be 314 °C and 318 °C. The higher sulfur fugacity of the fluid resulted in lower formation temperatures and greater Fe contents in the sphalerite. Altogether, these facts suggest that the sphalerite was deposited under reducing fluid conditions. After the pyrrhotite precipitated, most of the remaining Fe was used for the formation of Fe-rich sphalerite precipitates. These processes consumed all of the available sulfur and prevented the formation of pyrite and/or marcasite.