The Paleoproterozoic Chibaisong Mafic-Ultramafic Intrusion and Cu-Ni Deposit, North China Craton: SHRIMP Zircon U-Pband Re-Os Geochronology and Geodynamic Implications

Observations of the spatial and temporal setting of magmatic nickel sulfide deposits (or camps comprised of multiple deposits) indicate an empirical relationship with major lithospheric boundaries, and more specifically, to craton margins [1]. This relationship remains poorly documented in the scientific literature. The North China Craton (NCC) is the largest nickel producing area in China (Figure 1). Until now, two mines have been in production, and numerous other deposits have been found along the margin of the NCC. Since the discovery of the Jinchuan deposit in 1979, extensive geological exploration has been carried out in the NCC region, which led to the recognition of the NCC as one of the largest nickel-copper provinces in China. Over the last few years, numerous attempts have been made to directly or indirectly determine the ages of the nickel-copper provinces [2-5]. However, the timing of the magmatic and mineralization events and their correlations within the complexes remain unclear


Introduction
Observations of the spatial and temporal setting of magmatic nickel sulfide deposits (or camps comprised of multiple deposits) indicate an empirical relationship with major lithospheric boundaries, and more specifically, to craton margins [1]. This relationship remains poorly documented in the scientific literature. The North China Craton (NCC) is the largest nickel producing area in China ( Figure 1). Until now, two mines have been in production, and numerous other deposits have been found along the margin of the NCC. Since the discovery of the Jinchuan deposit in 1979, extensive geological exploration has been carried out in the NCC region, which led to the recognition of the NCC as one of the largest nickel-copper provinces in China. Over the last few years, numerous attempts have been made to directly or indirectly determine the ages of the nickel-copper provinces [2][3][4][5]. However, the timing of the magmatic and mineralization events and their correlations within the complexes remain unclear In this study, we present new U-Pb zircon and Re-Os isotopic data for the mineralized ultramafic body from the eastern coppernickel belt in the Jiao-Liao-Ji Belt of the NCC in order to constrain the timing and origin of the mineralization. These new data provide important insights into understanding the mineralization processes and geodynamic environment of the Chibaisong Ni-Cu ore deposit. In addition, an attempt was also made to understand the provenance and geodynamic environment of the NCC copper-nickel province. An understanding of the mineralizing processes and the geodynamics is beneficial for Ni-Cu exploration on other mafic-ultramafic bodies in the region and throughout the NCC.

Local geology
The host rocks of the Chibaisong Cu-Ni deposit are metamorphic rocks of the Sidaolazi Formation of the Archean Anshan Group, which consists predominantly of biotite-amphibole-plagioclase gneiss, migmatite and amphibolite, all of which show strong ductile deformation (Figure3). The Chibaisong intrusions, extending 4800 m long along strike, with a width between 40 and 140 m (Figure 3), and a depth of 1000 m (Figure 4), are composed mainly of diabase gabbro, livinenorite-gabbro, plagioclase lherzolite, norite-gabbro and gabbroporphyrite [28]. The Cu-Ni sulfide ores are hosted in the plagioclase lherzolite and norite-gabbro units. The orientation of the orebody is almost consistent with that of the intrusion, with its northern part striking 5-10° and dips to the SEE at 55 to 86°, and its southern part dipping to the NWW at 63 to 85° [28]. The orebody occurs as a tabular or bedlike unit, with a maximum length up to 200 m and a thickness of up to 40 m at the surface. The exploratory profile section reveals that the ore body extends as much as 1000m down-dip.
Wall-rock alteration related to mineralization in the Chibaisong Cu-Ni deposit includes chloritization, sericitization, carbonatization, serpentinization. Characteristic minerals in the alteration zones are chlorite, serpentine, sericite and carbonate.
For Re-Os dating, we collected 6 samples from fresh open-pit mining faces of the Chibaisong deposit. Samples were collected at a depth of 300 m in the eastern part of ore body ( Figure 3).
Single sulfide grains were also selected at the above-mentioned laboratory. Crude ore samples were jaw-crushed, elutriated by water, separated by Frantz magnetic techniques, panned by alcohol, and finally hand sorted under the microscope. The weight of six pure pyrrhotite samples was about 5g each. To avoid Re contamination from steel and tungsten carbide crushing equipment, the pure pyrrhotite powders from the Fujia deposit were pulverized to b200-mesh grain size using a ceramic jaw crusher and agate mill. The purity of the samples was greater than 99% prior to Re-Os isotopic analyses.

Analytical procedures
Cathodoluminescence (CL) images: The CL images were obtained using a CAMECA SX50 microprobe at the Institute of Geology and Geophysics, Chinese Academy of Sciences in Beijing, in order to identify internal textures and choose potential target sites for later U-Pb analyses. The working conditions during CL imaging were at 15 kV.

SHRIMP U-Pb isotope analytical techniques:
Zircon-U-Th-Pb analyses were performed using the Sensitive High-Resolution Ion Microprobe (SHRIMP II) at the Beijing SHRIMP Laboratory, Chinese Academy of Geological Sciences. Details of the analytical methods and procedures for zircon dating using SHRIMP have been described by [29][30][31][32]. Analytical target sites within zircons were chosen for SHRIMP U-Pb zircon dating. Due to small amounts of 207 Pb formed in young (i.e., ~1000 Ma) zircons, which results in low count rates and high analytical uncertainties, the determination of the ages for young zircons has to be based primarily on their 206 Pb/ 238 U ratios; older zircon ages are derived from 207 Pb/ 206 Pb ratios [33]. During the SHRIMP analyses, the intensity of the preliminary ion current was 2.5-4.5 nA and spot size averaged 25-30 μm, using peakjumpingscanningmode. The measured 206 Pb/ 238 U ratios were used for inter-element fractionation corrections with TEMORA zircon standards (417 Ma; 206 Pb/ 238 U=0.06683). The SL 13 zircon standard was used for calibration of U, Th and Pb contents. Common lead corrections were based on measured 204 Pb assuming that the 206 Pb/ 238 U-206 Pb/232Th age was accordant. The standard sample was measured after every three points, in order to insure the reliability of the measured results and then to monitor the stability of the instrument. The isotopic ratio and element content data were handled using ISOPLOT programs [34]. The age uncertainties are cited as 1σ, and the weighted average ages of 206 Pb/ 238 U used have a confidence level (2σ) of 95%.

Re-Osisotope analytical technique
Re-Os isotopic analyses were performed at the National Research  Center of Geoanalysis. The details of the chemical procedure are described by [35][36][37], and are briefly summarized as follows.
Enriched 190 Os and enriched 185 Re were obtained from the Oak Ridge National Laboratory, USA. A Carius tube (a thick-walled borosilicate glass ampoule) digestion was used. The weighed sample was loaded in the Carius tube through a thin necked long funnel. The mixed 190 Os and 185 Re spike solutions and 2 ml of 12M HCl and 6 ml of 15 M HNO 3 were loaded while the bottom part of the tube was frozen at -80°C to -50°C in an ethanol-liquid nitrogen slush; the top was then sealed using an oxygen-propane torch. The tube was then placed in a stainless-steel jacket and heated for 24 hours at 230°C. Upon cooling, the bottom part of the tube was refrozen, the neck of the tube was broken, and the contents of the tube were poured into a distillation flask and the residue was washed out with 40 ml of water.
Osmium was distilled at 105-110°C for 50 minutes and trapped in 10 ml of water. The residual Re-bearing solution was saved in a 150 ml Teflon beaker for late Re separation. The water trap solution was used for ICP-MS (TJA X-series) determination of the Os isotope ratio.
The Re-bearing solution was evaporated to dryness, and 1 ml of water was added twice, followed by heating to near-dryness, the 10 ml of 20% NaOH was added to the residue followed by Re extraction with 10 ml of acetone in a 120 ml Teflon separation funnel. The water phase was then discarded and the acetone phase washed with 2 ml of 20% NaOH. The acetone phase was transferred to a 120 ml Teflon beaker that contained 2 ml of water. After evaporation to dryness, the Re was picked up in 1 ml of water that was used for the ICP-MS determination of the Re isotope ratio. Cation-exchange resin was used to remove Na when the salinity of the Re-bearing solution was more than 1 mg/ml [38].
The ICP-MS analysis was conducted on a TJA PQ ExCELL ICP mass spectrometer. The instrument was optimized to: >5×10 4 cps for 1 ng ml-1115 In and >5×10 4 cps for 1 ng ml -1238 U. Data acquisition was performed in peak-jumping mode, 3 points/u; dwell time=15 ms/point and number of scan=200 for 5 ppb of Re in solution. The reproducibility by ICP-MS is 0.3% (RSD, 2S, n=5); by using water as an absorbent for OsO 4 , the sensitivity of Os by ICP-MS increases significantly. For 0.2 ppb of Os solution, the reproducibility is 0.3% (RSD, 2S, n=5).
If a minor 190 Os signal was observed when measuring Re, the 187 Re signal was appropriately corrected for 187 Os using the 187 Os/ 190 Os ratio of the spiked Os solution. Conversely, if a minor 185 Re signal was observed while analyzing the Os-bearing solution, 187 Os was appropriately corrected for 187 Re using the measured 185 Re/ 187 Re of the spiked sample. The corrections were generally minor and constituted no more than 0.1% of the isotope signal. The maximum correction percentage that we used was always less than 1%.
The mass fractionation can be corrected using an interlaboratory isotope reference standard. Using the λ 238 U value of [39] and λ 235 value of [40], a value for λ 187 Re of 1.6689 ± 0.0031×10 -11 a -1 was determined. This value is nominally higher (ca. 0.1% and ca. 0.2%) than the value determined by [40], but is within the calculated uncertainty. So, we used the λ 187 Re1.666 ± 0.005×10 -11 a -1 determined by [40].  Average blanks for the total Carius tube procedure as described above were ca. 10 pg Re and ca. 0.1 pgOs. 187 Os was not detected. Three reference materials were used to check the analytical results.
The uncertainty in each individual age determination was about 1.02% including the uncertainty of the decay constant of 187 Re, uncertainty in isotope ratio measurement, and spike calibrations. The decay constant used for 187 Re of 1.666×10 -11 a -1 has an absolute uncertainty of ± 0.017 (1.0%) [40].

SHRIMP U-Pb dating results
Twelve analyses of 12 grains from Chibaisong olivine-noritegabbro were analyzed on the CZ3 standard ( Figure 5). Of these analyses (Table 1), two age populations are identified (Figure 6), of which one population consists of eight grains gave a Th/U ratio larger than 4.0, indicating an igneous origin, and yield a 207 Pb/ 206 Pb weighted average age of 2188 ± 8 Ma, which is considered as the best estimate of the eruption age of this rock. Another age group consists of four grains with Th/U ratios of less than 1.0, these zircons have Th and U contents and Th/U ratios ranging respectively from 84 to 393 ppm, 175 to 823 ppm and 0.33 to 0.48, which are different from typical metamorphic zircons and are thus regarded to be detrital zircons that suffered variable Pb loss in metamorphism. Thus, the concordant age of 2188 ± 8 Ma may be the best constraint for estimating the maximum age of the Chibaisong deposit.

Re-Os dating results
The abundance of Re and Os and the osmium isotopic compositions of the Cu-Ni sulfide ores from the Chibaisong mine are shown in Table  2. Re and Os concentrations for the massive sulfide of the Chibaisong deposit are high, ranging from 0.47 to 13.97 ppb Os and 19 to 490 ppb Re ( Table 2). The analyses on six samples define a Re-Os isochron with an age of 1885 ± 94 Ma, with initial 187 Re/ 188 Os ratio of 0.80 ± 0.16 and a mean square of weighted deviation (MSWD) of 0.17 (Figure7). The isochron age was calculated by means of the 187 Re decay constant of 1.666×10 -11 /year [40] using the ISOPLOT software (Model 3; [34]). This isochron age can reflect the ore-forming age of the Chibaisong Cu-Ni deposit. The combination of high Os concentrations coupled with low Re/Os ratios is not typical of most low-tenor, basalt hosted magmatic sulfides (e.g. Duluth Complex), which exhibit much lower Os concentrations, coupled with higher Re/Os ratios [41]. This suggests a parental magma for the Chibaisong sulfides that had both a higher Os concentration and a lower Re/Os ratio than a typical basaltic magma.

Initial 187 Os/ 188 Os and source of ore-forming metals
The Re-Os isotope system has been recognized as a geochemical tool not only for directly dating mineralization but also for defining the ore forming process of Cu-Ni sulfide deposits, and thus is a powerful tracer of sulfide ore formation and can be a highly sensitive monitor of the extent of crustal involvement during ore genesis [42,43]. Since the initial 187 Os/ 188 Os ratios of the crust (0.2-10) are higher than those of the mantle (0.11-0.15, [44]), 187 Os/ 188 Os ratios can be used to readily discern different sources [45].
The Chibaisong Cu-Ni sulfides ores possess an initial 187 Os/ 188 Os ratio of 0.80 ± 0.16 (Figure 7), which is considerably greater than chondritic or value those of the mantle 187 Os/ 188 Os ratio at 1885 Ma [35,46,47], indicating that crustal components were involved in the Os source of the Chibaisong ores.
In order to describe the Os isotopic composition at a given time, a parameter of γ Os was introduced by [46][47][48]. Owing to high Re/Os ratios in the crust, γOs will have a large positive value with increasing crustal material entering the magmatic or ore-forming systems, and, by contrast, Re-loss in the systems can cause   This implies that ores and melts at the Chibaisong deposit were all dominated by mantle sources, although some minor amount of crustal material may have been assimilated into the melts.

Age of mineralization and its geodynamic significance
The Chibaisong intrusion was previously considered to have been linked to a post-orogenic extensional setting in the Early Cretaceous [5]. However, our new SHRIMP and Re-Os geochronological results indicate that the Chibaisong mafic-ultramafic intrusion, along with the Cu-Ni deposit, was probably related to the Paleoproterozoic mantle plumes or superplume, rather than in the Early Cretaceous postorogenic extension setting.
This new SHRIMP and Re-Os age (1885 ± 94Ma) for sulfide mineralisation at the base of the Chibaisong intrusion is marginally younger than the previously determined ages of the nearby Chibaisong intrusion (2188 ± 8Ma). Working on the geochronology of dyke swarms in the NCC, Peng et al. [53] identified four dyke suites with SHRIMP   [56]; Duluth Complex [57], Sudbury [58], and Noril'sk-Talnakh [46]; Komatiite-associated Ni sulfide ores from Kambalda [42]; Archean lithospheric mantle xenoliths from the Kaapvaal and Siberian Cratons [48]; mantle melts, which include komatiites [42,49] and basalts [60][61][62]; Lewisian lower crustal gneisses; metalliferous S-rich sediments; and chondritic mantle abundances [48]. Os uncertainty <1% at 2 standard errors of the mean, including error in mass spectrographic analysis, fractionation correction and measured isotope ratios in the samples. negative γ Os [49]. According to the formula of [48], and with a decay constant of 1.666×10 -11 /year [40], the γ Os values for the Chibaisong Cu-Ni sulfide ores were calculated based on the isochron age of 1885 ± 94Ma and their corresponding initial 187 Os/ 188 Os ratios. Table 2 shows that the γ Os values for the Chibaisong Cu-Ni sulfide ores range from 87 to 488. The initial 187 Os/ 188 Os ratio of 0.80 ± 0.16 for the Chibaisong sulfides is higher than that of 0.1089 ± 0.00035 reported for the uncontaminated Archeankomatite-related Cu-Ni sulfide ores [50]. These data reflect that mafic-ultramafic magmas were mixed with crustal components during the uprised of the magma or within magma chambers in the crust. This process is analogous to the crustal contamination of the Voisey's Bay magma by the Tasiuyakparagneiss where siderophile and chalcophile elements were selectively incorporated and the case of the Jinbulake intrusion in Xinjiang where crustal sulfides were selectively incorporated to magmas [51]. Such features can also be seen in the common Os vs. Re/Os diagram (Figure 8).

Sample
Fu [52] reported the δ 34 S values of sulfides from the Chibaisong ore body range from -1.1 to +0.7‰, with a mean of -0.03‰, and δ 18 O values of plagioclase from the Chibaisong intrusion range from +6.1 to +7.7‰, with a mean of +7.15‰ [52]. The initial 87 Sr/ 86 Sr ratios of the intrusions vary from 0.703 to 0.708 [52]. These data support the hypothesis that the ore-forming metals were derived mainly from a mantle source and the crust was not a major contributor of volatiles.
U-Pb ages of 2147 ± 5, 1929 ± 8, 1834 ± 5 and 1778 ± 3 Ma. On the basis of this geochronology [52], Peng et al. [53] suggested that the older group of dyke suites could have been related to the amalgamation of the NCC to the Columbia supercontinent, whereas the younger suites were related to the breakup of Columbia and a mantle plume event at 2147-1778 Ma [52]. Mafic dykes in the Chibaisong region, have SHRIMP U-Pb and LA-ICP-MS ages of 2188 ± 8 and 2145 ± 19 Ma, respectively, confirming the Paleoproterozoic rifting processes within the NCC. The Chibaisong ultramafic-mafic intrusion was located within the Paleoproterozoic Jiao-Liao-Ji Belt which is considered as an intercontinental rift zone that developed about 2.2 Ga and the rifting basin was closed at about 1.93-1.90 Ga, followed by a post-orogenic extensional event at 1.88-1.85 Ga [15][16][17]53,54]. The Chibaisong ultramafic-mafic intrusion was likely the earliest product of this postorogenic extensional event.

Conclusions
i. Zircon SHRIMP U-Pb ages of the Chibaisongmagmatic Ni-Cu sulfide deposits indicate a maximum depositional age of 2188 ± 8 Ma. This result is consistent with the geological fact that ore-bearing maficultramafic intrusions intruded into the Sidaolazi Formation of the Anshan Group.
ii. Six massive Ni-Cu sulfide ores of the Chibaisong deposit yielded a Re-Os isotopic isochron age of 1885 ± 94Ma, indicating that the oreforming age was Paleoproterozoic.
iii. Massive Ni-Cu sulfide ores of initial 187 Re/ 188 Os ratio of 0.80 ± 0.16, and their γOs values ranged from +24 to +622, indicating that ore-forming materials came mainly from the mantle with crustal Os contamination, suggesting a mixture of mantle and crustal sources.
iv. The Chibaisong Cu-Ni deposit was located within the Paleoproterozoic Jiao-Liao-Ji Belt which was the earliest product of the post-orogenic extensional event.