Mineralogy and Stable Isotopes of Tetradymite from the Dashuigou Tellurium Deposit, Tibet Plateau, Southwest China

Due to the very limited quantity of associated tetradymite, both mineralogical and geochemical research on tetradymite is scarce and incomplete. By taking advantage of the discovery of the Dashuigou tellurium deposit in Tibet Plateau, the authors conducted mineralogical studies on tetradymite. The authors present new mineralogical data including reflectance, micro-pressure hardness, chemical compositions focusing on its chemical formula Bi2.00Te1.89S0.95~1.00, unit cell parameters (a = 4.239 Å, c = 29.595 Å) and lattice parameters (a = 10.172 Å, v = 154.391 Å3), pyroelectricity (N type), and stable isotopes including sulfur and lead. The authors find that: δ34S‰ of tetradymite varies between −0.5~2.1 with a 15-sample average of 0.56, similar to that of meteorites and rocks from the mantle, indicating that the sulfur is from the mantle; lead isotopes of the tetradymite formed in the late metallogenic epoch is different from that of both pyrite and pyrrhotite formed in the early metallogenic epoch, further indicating that the three minerals formed in different metallogenic epochs; lead isotope compositions reveal that tellurium ore bodies emplaced in a quick process mainly in the form of ore magma; lead of the deposit is primarily from the mantle with some captured from the Earth’s crust. These findings help fill in large gaps of information for the mineral tetradymite.


Geology of the Deposit
The strata of the area are low-grade metamorphic rocks of the lower-middle Triassic age, including marble, slate and schist. The main wall-rocks of the ore bodies are schist and slate. All of the Triassic strata make up a NNE-trending dome. The geological and geochemical characteristics in the area indicate that the protolith of the tellurium ore veins' direct wall-rocks is poorly differentiated mantle-derived basalt [3][4][5][6][7] .
Both faults and folds are well-developed in the area. The annular and linear structures together make up special "Ø" pattern structures, which control the formation of different types of endogenetic mineral deposits, including the Dashuigou tellurium deposit.
No intrusive rocks emerge within a 5 km radius around the deposit. Only two small Permian ultrabasic-basic rock bodies emerge within a 10 km range of the deposit. Large neutral, acid and alkaline intrusive bodies exist beyond 10 km, which are unrelated to the deposit.
The Te content in the granites is under 1 × 10 −7 , which is similar to its Clark value in the Earth's crust. Te in the metamorphic rocks is slightly higher than in the granites and varies slightly between metamorphic rocks of different geological times, while being relatively higher in the Triassic metamorphic rocks. Of the metamorphic rocks in the same geological time, the Te content in the slate and schist is higher than in the marble. Te content in rocks of the same stratohorizon of the same geological time also varies; namely, it is higher in rocks within the mining area than in those beyond the mining area. Te content is closely related to the intensity of alterations; that is, the ore-forming elements are not derived from the country rocks, but instead from the mantle.
The deposit is located at the northeastern end of the Triassic metamorphic dome. The ore bodies are controlled by and fill a group of shear fractures. Nine tellurium ore veins have been discovered, which strike from 350 to 10 degrees and dip at 55 to 70 degrees westward. Widths of the ore bodies vary between 25 and 30 cm. The narrow ore bodies are in the shape of lenticular veins and have sharp contact with the wall rocks.
The altered rocks occur in narrow bands ranging between several centimeters and one meter in thickness. Altered zones beside the massive ore veins are narrower, at only several centimeters wide. The dominant alterations include dolomitization, silicification, biotitization, muscovitizaion, tourmalinization, sericitizaion, greisenization, and chloritization.

Mineralogy of tetradymite
Mineragraphy. Tetradymite is the most common telluride and makes up more than 90% of all the tellurides of the deposit. It occurs as a silvery-white fine-to coarse-grained flake with one group of complete cleavages and lower Moh's hardness (1.4 ~ 2.1). Reflectance of the tetradymite under the four standard wavelengths is listed in Table 1 with a reflectance color of yellowish white. Table 2 lists its micro-pressure hardness.
Members of the tetradymite group present complex problems, many of which remain unsolved due to incomplete data 15 .
The group of compounds in the four-compound system Bi-Te-Se-S represents a particular challenge, not least because the number of natural occurrences of the phases that have been comprehensively documented remains limited 16 . The difficulties are compounded by the rarity of the species, the varying quality and limited   www.nature.com/scientificreports www.nature.com/scientificreports/ quantity of published data including reflectance and micro-pressure hardness of tetradymite, and by the invariably smaller-grained size and intergrown character of the mineral. Since the published data on reflectance and micro-pressure hardness of tetradymite is not plentiful, it does not permit a comparison and confirmation regarding the normality of these results. chemical compositions & formula. Four of the nine tellurium veins in total had been mined out by the time this research started. As a result, all samples were collected from the five remaining veins. Chemical compositions of the tetradymite analyzed by electronic probe are listed in Table 3.
Based on the results presented in Table 3, chemical compositions of the tetradymite samples collected from different ore bodies of the mine are very similar: Te content varies between 34.38~34.80% with a maximum difference of 0.42% and average of 34.60%; Bi content varies between 59.48~60.29% with a maximum difference of 0.81% and average of 60.08%; S content varies between 4.38~4.81% with a maximum difference of 0.43% and average of 4.61%.
Synthetic tetradymite (Bi 2 Te 1.9 S 1.1 ), which is similar to the tetradymite from the Dashuigou deposit, and two compounds, Bi 48 Te 21 S 31 and Bi 5 Te 3 S 2 , were generated at 400 °C. Sulfur-rich tetradymite appears more chemically stable than stoichiometric Bi 2 Te 2 S 16 .
The contents of trace elements in the tetradymite are similar to those in the mine's pyrrhotite and pyrite. The only difference is tetradymite is richer in gold than both the pyrite and pyrrhotite, which is identical to the observation results under microscope study of the minerals. This indicates that gold formed in the tellurium epoch mentioned above.
X-Ray power diffraction data. Due to insufficient quantity of the tetradymite mineral, single-crystal X-ray studies are lacking and previous X-ray diffraction data and unit cell parameters of tetradymite were mainly obtained via crystal powder photography, the accuracy of which is not very satisfactory.
The quantitative phase analysis of one powder sample (#SD-40, Fig. 3 Experimental method and procedure. The sample SD-40 was reduced to the optimum grain-size range for quantitative X-ray analysis (<10 μm) by grinding under ethanol in a vibratory McCrone Micronizing Mill for 10 minutes. Continuous-scan X-ray powder-diffraction data were collected over a range 3-80 °2 with CoKα radiation on a Bruker D8 Advance Bragg-Brentano diffractometer equipped with an Fe filter foil, 0.6 mm (0.3°) divergence slit, incident-and diffracted-beam Soller slits and a LynxEye-XE detector. The long fine-focus Co X-ray tube was operated at 35 kV and 40 mA, using a take-off angle of 6°.
Results. The X-ray diffractogram was analyzed using the International Centre for Diffraction Database PDF-4 and Search-Match software by Bruker. X-ray powder-diffraction data of the sample were refined with Rietveld program Topas 4.2 (Bruker AXS). The results of quantitative phase analysis by Rietveld refinements are given in  Table 6. Comparison of X-ray diffraction data of tetradymite from Dashuigou deposit and tetradymite/ kawazulite from deposits of other countries. Table 4. These amounts represent the relative amounts of crystalline phases normalized to 100%. The Rietveld refinement plot is shown in Fig. 4. Lattice parameters and volumes are given in Table 5. Another X-ray crystal powder diffraction analysis done in lab of the Research Center of Standard Materials, China Academy of Metrological Sciences, Beijing of China reveals that the tetradymite from the Dashuigou tellurium deposit is different than that from the Paonia mine in Colorado, US, but similar to kawazulite from the Kawazu mine in Japan. Even so, tetradymite from the Dashuigou deposit is lacking in Se compared to the Japanese kawazulite (Tables 6 and 7 and Fig. 5).
Kawazulite, Bi 2 Te 2 Se, which was originally described by Kato (1970) as the Se analogue of tetradymite, is isostructural with tetradymite, which has been synthesized by Glaz (1967), Evdokimenko and Tsypin (1971), and Abrikosov and Beglaryan (1973). The compositional limits of tetradymite have been known to be Bi 2 Te 2 S-Bi 2 Te 1.7 S 1.3 As determined by Kuznetsov and Kanishcheva (1970). Pauing (1975) explained why the substitution of Te by S increases the chemical stability. The calculated intensities of an X-ray powder-diffraction pattern of kawazulite with the ordered tetradymite structure type are similar to the visually estimated observed X-ray powder-diffraction intensities of kawazulite (PDF 29-248). The visually estimated observed intensity data is not sufficiently accurate to exclude partial ordering of Se and Te 15 . Extensive solid-solution between S-and Se-bearing end members can be documented in many series; for instance, tetradymite -kawazulite, and continuous solid-solution between tetradymite and kawazulite is well developed 16 .
The experimental and calculated two theta values with hkl indices are listed in Table 8.   www.nature.com/scientificreports www.nature.com/scientificreports/ pyroelectricity. It is believed that pyroelectricity of minerals can be used to determine their origin and process of formation. Unfortunately, there existed no pyroelectricity data of tetradymite prior to the research of this paper.
Pyroelectricity of minerals can be divided into N (electron) type, P (electron hole) type and the mixed N and P type.
Pyroelectricity characteristics of the tetradymite from Dashuigou deposit are listed in Table 9.
The tetradymite from Dashuigou deposit is completely of N conduction type. All of its pyroelectricity is negative, and the values are both close to each other and vary between −209.0 ~ −274.0 µV/°C. With a maximum difference of −65.0 µV/°C and an average of −243.12 µV/°C, the data implies that the tellurium veins were all formed in the same geological event and from the same source.
The negative pyroelectricity of the tetradymite resulted from insufficient sulfur, As and Se impurities, and other isomorphous mixtures of Te in tetradymite.

Sulfur and Lead isotopes of tetradymite
Sulfur isotope. Sulfur isotope results of the dominant sulfides collected from various veins of the deposit are provided in Table 10. It can be seen that sulfur isotopes of various sulfides formed in different veins of different metallogenic epochs and/or stages are very close to each other, varying within a narrow range with an average below 1‰, a crest value around 0.6‰, and a clear tower effect (Fig. 6).
Cao and Luo researched and published their δ 34 S CDT ‰ results of 1.13~3.17 with an average of 1.96 17 , which indicated that sulfur isotopes of the Dashuigou deposit were very homogeneous and thus originated from the deep mantle.
In general, sulfur isotopes of the deposit are very similar to those of meteorite, moon rock and mantle-derived materials, indicating that the sulfurs are from the mantle. This finding is in agreement with Cao and Luo's studies 17 . The δ 34 S CDT ‰ of series #16 & #22 samples in Table 10, which are collected from regional country rocks far from the deposit, clearly deviate from that of samples collected from the deposit.
The series #16 sample is coarse-to very coarse-grained cubic pyrite in the upper Permian metamorphosed basalt to the south of the deposit, while the series #22 sample is chalcocite from the Shaoyaocao copper showing, demonstrating a connection to the basic to ultra-basic intrusive to the east of the deposit. In theory, δ 34 S CDT ‰ of these two samples should be close to that of mantle-derived materials (−3 ~ 2), since their wall rocks are from the mantle. In fact, they are not similar to each other, which may be owed to the post metamorphism after their formation.
δ 34 S CDT ‰ of all the pyrrhotite samples varies between −3.1 ~ 2.1 with an 8-sample average of 0.175, close to that of the meteorite, indicating that they are mantle derived. Meanwhile, δ 34 S CDT ‰ of all the pyrite samples is between 1.4 ~ 2.8 with a 6-sample average of 1.717, also close to that of meteorite and implying that they too are mantle derived.
δ 34 S CDT ‰ of both the deposit's pyrrhotite and pyrite formed in the same metallogenic epoch are in a similar narrow scope. For samples #SD-29 and SD-55 in Table 10, δ 34 Spyrite‰ <δ 34 Spyrrhotite‰, suggesting that the sulfur exchange between pyrite and pyrrhotite of these samples did not reach balance. For samples #SD-23 and SD-41 in Table 10, δ 34 Spyrite‰> δ 34 Spyrrhotite‰, indicating that sulfur exchange between pyrite and pyrrhotite of these samples reached balance.
As shown in Table 10 and briefly discussed above, δ 34 S CDT ‰ of both pyrrhotite and pyrite varies within a very narrow range, indicating that sulfur exchange between pyrite and pyrrhotite became highly uniform and achieved balance. As a result, δ 34 S Σs ‰, the general sulfur component of the deposit's metallogenic hydrothermal solution, can be calculated by the following equation 18 :   The general sulfur component of the late tellurium metallogenic epoch's hydrothermal solution is close to that of the meteorite and demonstrates that the deposit's sulfur is derived from the mantle.
Per discussions on sulfur isotopes of the deposit's minerals above, the authors' preliminary conclusions are that δ 34 S of both single sulfide minerals from different veins of different metallogenic epochs and the general total sulfur isotope components of the deposit's metallogenic hydrothermal solutions vary within very narrow scopes with very small ranges close to 0.0‰. As a result, the deposit's sulfur is very close to that of the meteorite and may be derived from the mantle.  Table 10  www.nature.com/scientificreports www.nature.com/scientificreports/ Lead isotope. Lead isotope results of samples collected from the study area are listed in Table 11, and further summarized in Table 12. Model lead ages in Table 11 vary significantly between 0.00 ~ 916.25 Ma, likely indicating that the lead isotopes mainly consist of more radioactive lead which is not homogeneous and thus results in strong anomalous model ages. Figure 7 shows the components of lead in the study area. According to Zhang 18 , the area to the left of 206 Pb/ 204 Pb = 18.5 is generally the evolutionary area of normal lead, while to the right is the evolutionary area of anomalous leads. This confirms that lead isotopes in the study area, especially those of ore samples from the deposit, are not uniform. As a result, the model lead ages in Table 11 does not make sense geologically.
In general, the lead isotopes of pyrite are more homogeneous than that of pyrrhotite, with the least uniformity in tetradymite. This can be seen in Fig. 8, as some samples fall out of the semilunar area composed of the µ = 10 (0 time) line and the evolutionary curve, indicating that there exists anomalous lead. This further confirms that stable lead isotopes in the area are a mixture of both normal and anomalous ones.
Parent rock of the slate is mantle derived basaltic rock 3,5,6 . A comparison of slate's lead isotope (series #24 sample in Table 11) with that of material with proven origins is presented in Table 13. It can be seen that the deposit's wall rock slate is mantle derived, though post metamorphism made it deviate from the lead of pure mantle materials.
Upon synthesizing lead isotope results of slate and schist from the middle-lower Triassic ages published by former researchers in the study area (Table 14), lead isotope deviations caused by the post metamorphism become clear.
Lead isotope of the deposit's middle-lower Triassic marbles in Table 12 is similar to that of the upper crust in Table 13. Lead isotope of the lower Permian metamorphosed basalt as well as that of pyrite in the basalt are between those of the mantle and the lower crust. Likewise, lead isotope of the granite of the Indo Chinese epoch is similar to that of the upper crust, though it was possibly contaminated with the lead from the crust.
In comparison to Table 14, it can be seen that lead isotopes of the granite at the Niubeishan outside the deposit and those of the wall rocks at the deposit, including slate, schist, marble and metamorphosed basalt as well as the pyrite in the basalt, are similar to that of the post Cambrian rocks.  As demonstrated in Tables 11, 12 and 13, the isotopes of pyrite and pyrrhotite of the same metallogenic epoch of the deposit are similar to each other and similar to that of the mantle, indicating that the ores' leads are derived from the mantle. Lead isotopes of the tetradymite formed in the late metallogenic epoch are both similar to and different from that of pyrite and pyrrhotite formed in the early metallogenic epoch of the deposit, implying that lead isotopes of the late metallogenic epoch are mixtures of lead from both the mantle and the crust (Fig. 9). Table 11 clearly shows that lead isotopes of tetradymite from different ore veins are different from each other. This likely implies that:   www.nature.com/scientificreports www.nature.com/scientificreports/ • Emplacement of the ore bodies completed in such a short period of time that lead isotopes in the ore did not have sufficient time to homogenize; • When quickly uplifting and emplacing, the ore bodies rose in the form of ore pulp, during which lead isotopes from the crust were captured and mixed with the original mantle derived isotopes. This means that the lead isotope compositions can be used to determine the metallogenic mechanism of mineral deposits.
It is interesting to note that lead isotope of tetradymite from Ore vein #I-10 (series #22 sample in Table 11) is different from that of the series #24 sample collected at the footwall of Ore vein #I-10 in the same table. This may further confirm that tellurium ore bodies emplaced in the form of ore magma in a very quick process, during which captured and mixed leads from both the mantle and crust and made compositions of the lead isotopes very complicated.
It can be concluded on a preliminary basis that lead isotopes of the research area consist of more radioactive lead and are very inhomogeneous. The lead isotopes of the wall rocks slate, schist and metamorphosed basalt as well as the pyrite in the basalt are mantle derived, though overlapped with characteristics of lead of the later geological events. Lead isotopes of both pyrite and pyrrhotite are close to each other and mantle derived, while that of tetradymite is different from the former, indicating that they formed in different metallogenic epochs. Lead isotope compositions reveal that tellurium ore bodies emplaced in the form of ore magma in a quick process. Finally, lead of the deposit was mainly from the mantle but also partially captured some lead from the crust.

conclusions
• Chemical formula of the tetradymite from the Dashuigou deposit is Bi 2.00 Te 1.89 S 1.00 , which contains more Te but less S compared to the tetradymite with a chemical formula of Bi 2.00 Te 1.65 S 1.35 on the JCPDS card in 1979, indicating that the Dashuigou tetradymite was formed in an Te-rich but S-poor environment. • The negative values of the Dashuigou tetradymite's pyroelectricity indicate that it formed in an environment with low sulfur fugacity. • The sulfur isotope δ 34 S of both single sulfide minerals including tetradymite from different veins of different metallogenic epochs, and the general total sulfur isotope components of the deposit's metallogenic hydrothermal solutions, vary within a very narrow scope and a very small range close to 0.0‰, indicating that the deposit's sulfur is similar to that of meteorites and may be derived from the mantle. • Lead isotopes of samples from the research area consist of more radioactive lead and are very inhomogeneous; lead isotopes of the wall rocks slate, schist and metamorphosed basalt as well as the pyrite in the basalt are mantle derived, though overlapped with characteristics of lead of the later geological events; lead isotopes of both pyrite and pyrrhotite are close to each other and mantle derived, while that of tetradymite is different from the former, demonstrating that they formed in different metallogenic epochs; lead isotope compositions reveal that tellurium ore bodies emplaced in a quick process mainly in the form of ore magma; lead of the deposit is mainly from the mantle but captured some other lead from the crust.

Data availability
The data that support the findings of this study is available from the authors upon reasonable request; see authors' contributions for specific data sets.