Pb-Pb and U-Pb Dating of Cassiterite by In Situ LA-ICPMS: Examples Spanning ~1.85 Ga to ~100 Ma in Russia and Implications for Dating Proterozoic to Phanerozoic Tin Deposits

: This paper investigates applicability of cassiterite to dating ore deposits in a wide age range. We report in situ LA-ICPMS U-Pb and Pb-Pb dating results ( n = 15) of cassiterite from six ore deposits in Russia ranging in age from ~1.85 Ga to 93 Ma. The two oldest deposits dated at ~1.83–1.86 Ga are rare metal Vishnyakovskoe located in the East Sayan pegmatite belt and tin deposits within the Tuyukan ore region in the Baikal folded region. Rare metal skarn deposits of Pitkäranta ore ﬁeld in the Ladoga region, Fennoscandian Shield are dated at ~1.54 Ga. Cassiterite from the Mokhovoe porphyry tin deposit located in western Transbaikalia is 810 ± 20 Ma. The youngest cassiterite was dated from the deposits Valkumei (Russian North East, 108 ± 2 Ma) and Merek (Russian Far East, 93 ± 2 Ma). Three methods of age calculations, including 208 Pb/ 206 Pb- 207 Pb/ 206 Pb inverse isochron age, Tera-Wasserburg Concordia lower intercept age, and 207 Pb-corrected 206 Pb*/ 238 U age were used and the comparison of the results is discussed. In all cases, the dated cassiterite from the ore deposits agreed, within error, with the established period of magmatism of the associated granitic rock. Constraining the age of tin mineralization is critical for developing robust genetic models of the deposits. Previously published geochronology of tin/rare metal deposits in Russia used dating of spatially related granites and/or minerals coexisting with tin ore (e.g., [25–36]). In this paper, we report the ﬁrst LA-ICPMS Pb-Pb and U-Pb ages for cassiterite from several tin and rare metal deposits in Russian Karelia, Siberia, and the Russian far east obtained using previously dated by ID-TIMS primary matrix-matched reference materials. Replicate analyses of a secondary matrix-matched material were used to estimate external age uncertainty of the reported age values. The results are used to provide recommenda-tions for the LA-ICPMS U-Pb dating of the Proterozoic and Phanerozoic cassiterite. for comparison pluton [30,66]. This pluton belongs to the Chaun igneous complex, composed of quartz monzonites, monzogranites, granites, zinnwaldite granites and ongonites, and topaz-bearing albite-rich microgranites. Mineralization occurs in a north-northwest-trending zone alone the contact of the pluton. of situ analyses to target inclusion- and fracture/vein-free parts of crystal CL of cassiterite most grains. The oscillatory zoning is commonly at grain margins. Using back-scattered electron and energy-dispersive spectroscopy (BSE-EDS) spot analyses, no mineral inclusions within the cassiterite were found; ( B ) L-304 (Tuyukan ore region, September deposit). The cassiterite exhibits light-colored growth banding in CL interrupted with compositional sector zoning. Minor later staged light-colored veinlets are common throughout. BSE-EDS analyses revealed quartz and plagioclase within the cassiterite matrix; ( C ) L-490 (Tuyukan ore region, Silvery deposit). CL images of the Silvery cassiterite indicate proliﬁc oscillatory zoning with typically ﬁne-scale light-colored banding and coarser-scale dark banding. The grains host inclusions of quartz, apatite, and K-feldspar as discovered with BSE-EDS inspection; ( D ) 31 (Pitkäranta Mining District, Old Mine Field deposit). These grains yielded minimal CL response, although sector zoning throughout the grains and thin, dark, U-rich growths at the crystal margins may be discerned. BSE-EDS evaluation indicate seemingly inclusion-free grains; ( E ) S-23-29 (Western Transbaikalia, Mokhovoe deposit). The cassiterite grains are characterized by extremely ﬁne-scaled zoning, followed by darker sector zoning. BSE-EDS review found no inclusions; ( F ) S-19-7 (Western Transbaikalia, Mokhovoe deposit). CL evidence shows that sector zoning played a dominant role followed by a period of uniformly dark CL activation with sporadic trace element mobility as evidenced by disruptive growth banding in distinct sections of the grains. Inclusions of scheelite and possible clay mineralogy were discovered with BSE-EDS analyses; ( G ) W-1 (Russian Northeast, Valkumei deposit). The Valkumei cassiterite is characterized by distinct oscillatory zoning, commonly with lesser U cores, U-rich mantle areas followed by late-stage U-poor rims. The larger grain at left is showing signs of crystal strain. BSE-EDS examination disclosed inclusion free grains; ( H ) MK-1 (Russian Far


Introduction
Cassiterite (SnO 2 ), a main ore mineral in tin deposits, is suitable for in situ U-Pb isotopic dating using laser ablation inductively coupled plasma mass spectrometry (LA-ICPMS) because of its relatively high U/Pb ratios and typically low initial "common" Pb (Pb c ). This has been demonstrated in a series of publications within the last decade (e.g., [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16]. Also, Neymark et al. [13] showed that Proterozoic low-Th cassiterites can be dated by an inverse Pb-Pb isochron approach using 208Pb as a common Pb proxy. Cassiterite is highly resistant to acid dissolution [17], which makes it difficult to achieve the complete acid digestion and spike homogenization needed for accurate isotope dilution and thermal ionization mass spectrometry (ID-TIMS) U-Pb dating. However, several recent publications [18][19][20] showed that complete dissolution of cassiterite and reliable ID-TIMS dating results are achievable. These publications provide the basis for the usage of better characterized matrix-matched reference materials for in situ analyses of cassiterite.
Tin reserves in Russia were estimated at 350,000 metric tons or 7.5 percent of world tin reserves in 2016 [21]. Tin in Russia is mined at both primary (lode) and secondary (placer) deposits. Unlike the situation in most other major producing countries (e.g., Malaysia, Indonesia, Nigeria, Brazil), in Russia the overwhelming majority of reserves is found not in placers, but in primary ores in lode deposits [22]. Approximately 95% of Russia's known tin reserves are located in its Far East economic region [23,24]. Additional tin-bearing deposits of less economic significance are known to exist in the Russian Karelia (the historic mining district of Pitkäranta) and in Siberia (e.g., the East Sayan and Baikal region). The tin ore deposits discussed in this paper differ in age, composition, and the nature of the associated granitoids of variable geochemical types formed in different geodynamic settings. The Pitkäranta mining district deposits are associated with the rocks of the anorthosite-rapakivi-granite magmatic association formed in an intraplate geodynamic setting. The Salmi batholith, which contains highly differentiated Li-F granites, is a part of The tin ore deposits discussed in this paper differ in age, composition, and the nature of the associated granitoids of variable geochemical types formed in different geodynamic settings. The Pitkäranta mining district deposits are associated with the rocks of the anorthosite-rapakivi-granite magmatic association formed in an intraplate geodynamic setting. The Salmi batholith, which contains highly differentiated Li-F granites, is a part of the Ladoga-Dalekarlian magma belt (1.66-1.53 Ga), which stretches along the southern boundary of the Fennoscandian shield for almost 2000 km [38]. Deposits of rare metal pegmatites of the East Sayan belt [39] and tin ore deposits of the Tuyukan ore region are Minerals 2021, 11,1166 3 of 30 associated with granite batholiths of the transregional South Siberian post-collision belt (1.88-1.84 Ga), stretching over the southern margin of the Siberian craton for more than 2500 km [40,41]. The belt is dominated by S-and A-type granites, sometimes forming bimodal magmatic associations with mafic rocks. The Mokhovoe deposit is associated with a granite-rhyolite volcano-plutonic association with geochemical characteristics of S-type granites formed during the final accretionary stage of the Baikal-Muya orogen of the Central Asian fold belt [42,43]. The deposits of the Russian Far East and North-East, Merek and Valkumei, are associated with the activity of the late Mesozoic Pacific active continental margin of the North Asian continent [44,45].

Vishnyakovskoe Rare Metal Ore Deposit
The Vishnyakovskoe large rare-metal (Ta, Li, Be, Sn, Rb, Cs) ore deposit is located in the northwestern part of the East Sayan pegmatite belt, which stretches for more than 500 km along the southwestern boundary of the Siberian Craton ( Figure 2) and includes several pegmatite fields. The Vishnyakovskoe pegmatite field lies in northwest part of the belt and is located within the Paleoproterozoic Elashsky graben near the contact with the granite of the Sayan complex [46]. The deposit contains the highest quality of tantalum ores in Russia [47]. Pegmatite veins are found in ortho-amphibolites and are characterized by gentle bedding. They form vein series, where individual bodies are up to two km long and up to 12 m thick. The main ore minerals of pegmatites are represented by manganotantalite, ixiolite, wodginite, columbite, microlite, spodumene, petalite, montebrasite, eucryptite, lepidolite, cassiterite, rynersonite, and beryl [48].
Minerals 2021, 11, x. https://doi.org/10.3390/xxxxx www.mdpi.com/journal/minera the Ladoga-Dalekarlian magma belt (1.66-1.53 Ga), which stretches along the souther boundary of the Fennoscandian shield for almost 2000 km [38]. Deposits of rare meta pegmatites of the East Sayan belt [39] and tin ore deposits of the Tuyukan ore region ar associated with granite batholiths of the transregional South Siberian post-collision be (1.88-1.84 Ga), stretching over the southern margin of the Siberian craton for more tha 2500 km [40,41]. The belt is dominated by S-and A-type granites, sometimes forming b modal magmatic associations with mafic rocks. The Mokhovoe deposit is associated wit a granite-rhyolite volcano-plutonic association with geochemical characteristics of S-typ granites formed during the final accretionary stage of the Baikal-Muya orogen of the Cen tral Asian fold belt [42,43]. The deposits of the Russian Far East and North-East, Mere and Valkumei, are associated with the activity of the late Mesozoic Pacific active cont nental margin of the North Asian continent [44,45].

Vishnyakovskoe Rare Metal Ore Deposit
The Vishnyakovskoe large rare-metal (Ta, Li, Be, Sn, Rb, Cs) ore deposit is located i the northwestern part of the East Sayan pegmatite belt, which stretches for more than 50 km along the southwestern boundary of the Siberian Craton ( Figure 2) and includes sev eral pegmatite fields. The Vishnyakovskoe pegmatite field lies in northwest part of th belt and is located within the Paleoproterozoic Elashsky graben near the contact with th granite of the Sayan complex [46]. The deposit contains the highest quality of tantalum ores in Russia [47]. Pegmatite veins are found in ortho-amphibolites and are characterize by gentle bedding. They form vein series, where individual bodies are up to two km lon and up to 12 m thick. The main ore minerals of pegmatites are represented by mangan otantalite, ixiolite, wodginite, columbite, microlite, spodumene, petalite, montebrasite, eu cryptite, lepidolite, cassiterite, rynersonite, and beryl [48].   The Sayan complex is composed of granitoids of two intrusive phases. The first phase is represented by biotite-amphibole granodiorites and granites; the second one, biotite and two-mica granites, biotite, muscovite, and tourmaline-containing leucogranites, pegmatitic, and aplitic veins. The geochemical characteristics of these granites correspond to the post-collisional S-type granites. The U-Pb zircon ages of the Sayan complex granites range from 1869 ± 6 to 1855 ± 5 Ma [49]. Makagon et al. [50] published Rb-Sr ages of 1490 Ma and 1480 Ma for pegmatites of the Vishnyakovskoe deposit and associated metasomatic rocks, respectively. More recently, Sal'nikova et al. [48] reported an older ID-TIMS U-Pb age of 1838 ± 3 Ma for manganotantalite from rare metal pegmatites at the Vishnyakovskoe deposit. The studied cassiterite sample 2622 was also separated from rare metal pegmatites at the Vishnyakovskoe deposit.

Tonod Uplift Quartz-Vein Tin Deposits
Tin ore deposits of the Tuyukan ore region are associated with granites of the Chuya-Kodar complex in the northern part of the Baikal folded region within the Baikal-Patom fold-thrust belt, which is mainly composed of shelfal carbonate-terrigenous sedimentary rocks of the passive southern margin of the Siberian craton ( Figure 3). The northern segment of this belt includes a series of uplifts (called Chuya, Tonod, Necher, and Kutim) where Paleoproterozoic intrusive and supracrustal rocks of the folded basement are exposed. The Sayan complex is composed of granitoids of two intrusive phases. The first phase is represented by biotite-amphibole granodiorites and granites; the second one, biotite and two-mica granites, biotite, muscovite, and tourmaline-containing leucogranites, pegmatitic, and aplitic veins. The geochemical characteristics of these granites correspond to the post-collisional S-type granites. The U-Pb zircon ages of the Sayan complex granites range from 1869 ± 6 to 1855 ± 5 Ma [49]. Makagon et al. [50] published Rb-Sr ages of 1490 Ma and 1480 Ma for pegmatites of the Vishnyakovskoe deposit and associated metasomatic rocks, respectively. More recently, Sal'nikova et al. [48] reported an older ID-TIMS U-Pb age of 1838 ± 3 Ma for manganotantalite from rare metal pegmatites at the Vishnyakovskoe deposit. The studied cassiterite sample 2622 was also separated from rare metal pegmatites at the Vishnyakovskoe deposit.

Tonod Uplift Quartz-Vein Tin Deposits
Tin ore deposits of the Tuyukan ore region are associated with granites of the Chuya-Kodar complex in the northern part of the Baikal folded region within the Baikal-Patom fold-thrust belt, which is mainly composed of shelfal carbonate-terrigenous sedimentary rocks of the passive southern margin of the Siberian craton ( Figure 3). The northern segment of this belt includes a series of uplifts (called Chuya, Tonod, Necher, and Kutim) where Paleoproterozoic intrusive and supracrustal rocks of the folded basement are exposed.   The tin deposits and prospects of the Tuyukan ore region are located within the Tonod uplift. Tin mineralization is related to the Paleoproterozoic coarse-grained post-collision S-type tin-bearing granites of the Chuya-Kodar complex. Tin ore deposits and occurrences are linear stockwork zones with quartz and quartz-muscovite veins, plus veins and linear zones of tourmaline-muscovite-quartz, microcline-muscovite-quartz, albite-quartz, and muscovite-quartz metasomatites with cassiterite mineralization (Figures 2B,C and 3C-H in Neymark et al. [16]). The deposits are confined to the ore-controlling faults of northeastern strike and are associated with small stocks and dikes (sometimes explosive breccias) of the Neoproterozoic Yazov porphyry granite and mainly hosted by the Chuya-Kodar granites. The main ore mineral is cassiterite. The ore also contains subordinate amounts of arsenopyrite, galena, sphalerite, chalcopyrite, and pyrrhotite. Previous dating of the Yazov granites and ores [16,51] confirmed the~720 Ma age of the granites and rules out any genetic relations between the tin ore dated at~1850 Ma and the Yazov granites. The geochronological data suggested that the tin ore is related to the older Paleoproterozoic host Chuya-Kodar coarse-grained porphyry two-mica granites dated at 1846 ± 8 Ma [40]. In this paper we use all U-Pb-isotope analyses for cassiterite samples L-304, L-316, L-490, and L-528 from Neymark et al. [16] to construct a combined Pb-Pb isochron to improve the precision of the ore age estimate.

Pitkäranta Mining District (PMD) Tin-Rare Metal-Base Metal Skarn Deposits
The historic mining district of Pitkäranta in the Ladoga region, Fennoscandian Shield (Figures 1 and 4), was exploited for Fe, Cu, Zn, Pb, Sn, and Ag in the nineteenth and twentieth centuries. It is located on the northeastern shore of Lake Ladoga, NW Russia and extends for~40 km from north to south at the western edge of the Salmi anorthositerapakivi granite batholith and has a maximal width of four to five km. The main type of ore deposits in the region is represented by skarn deposits with a multi-metal Fe, Sn, rare metal, base metal, and fluorite mineralization. The Pitkäranta ores constitute a classical skarn type, formed when granite-related hydrothermal fluids reacted with marble [52]. The deposits are related to the carbonate horizons of the Pitkäranta suite which encloses the gneiss-granite domes. Linear greisen zones carrying Be, Sn, Mo, and Cu mineralization are commonly found within various types of the rapakivi granites and granite gneisses of the domes. Tin was discovered in PMD skarns in 1834 [53,54]. Brief descriptions of thẽ 1.54 Ga Salmi batholith and Pitkäranta ore field and previous geochronology of granites and ores were published [55][56][57][58]. The granites are enriched in a number of rare metals (including Sn) and F. Geochemically they belong to the within-plate A-type granites. Salmi granitoids are strongly differentiated and their later intrusive phases are represented by small bodies and dikes of topaz-bearing granites. A cassiterite sample, SPG, from the Pitkäranta mining district was dated previously [13,19,20] and was also used as a matrix-matched reference material in LA-ICPMS U-Pb dating to correct for U/Pb instrumental bias [13,15,16,59]. During this study we analyzed five more cassiterite samples from the Pitkäranta region collected at the Old Mine Field (sample of pyroxene-garnet (Px-Gar) skarn 31 and sample LA-1 of quartz-carbonate metasomatite developed over a Px-Gar skarn with coarse-grained cassiterite) and at the Kitilä deposit (samples of Px-Gar-Mt skarns with cassiterite O-4-11, X-146, and VSH-5).   [55] and Larin [58]. Bt-biotite; Qz-quartz; Tpz-topaz.
A cassiterite sample, SPG, from the Pitkäranta mining district was dated previous [13,19,20] and was also used as a matrix-matched reference material in LA-ICPMS U-P dating to correct for U/Pb instrumental bias [13,15,16,59]. During this study we analyze five more cassiterite samples from the Pitkäranta region collected at the Old Mine Fie (sample of pyroxene-garnet (Px-Gar) skarn 31 and sample LA-1 of quartz-carbonate me asomatite developed over a Px-Gar skarn with coarse-grained cassiterite) and at the Kiti deposit (samples of Px-Gar-Mt skarns with cassiterite O-4-11, X-146, and VSH-5).

Mokhovoe Porphyry Tin Deposit
The tin-bearing metasomatites of the Mokhovoe tin deposit are located in wester Transbaikalia (in the Muya District of the Republic of Buryatia), in the Baikal-Muya be of the East-Baikal segment of the Central Asian fold belt ( Figure 5), and the Mokhovoe t deposit is ~370 km south of the tin deposits in the Tonod uplift. This 20 km × 30 km stru ture is mainly composed of volcanics within the Zhanok suite and by intrusive bodies Bambukoy granites. The ID-TIMS zircon ages 834 ± 23 Ma and 818 ± 7 Ma [60] for volcani and granites, respectively, coincide within error. The deposit consists of 10 tubular an two lenticular bodies of magnetite-feldspar metasomatites with cassiterite-chalcopyri and magnetite-hematite mineralization. The main ore minerals are magnetite, hematit mushketovite, cassiterite, chalcopyrite, bornite, pyrite, and rare stannite. The ore bodi are hosted by felsic rocks of the Bambukoy igneous complex that have geochemical fe tures similar to those of S-type granites. Tin concentrations in the rocks are elevated (2.7 7.2 ppm Sn) but are not high enough to be classified as tin granites [60].  [55] and Larin [58]. Bt-biotite; Qz-quartz; Tpz-topaz.

Mokhovoe Porphyry Tin Deposit
The tin-bearing metasomatites of the Mokhovoe tin deposit are located in western Transbaikalia (in the Muya District of the Republic of Buryatia), in the Baikal-Muya belt of the East-Baikal segment of the Central Asian fold belt ( Figure 5), and the Mokhovoe tin deposit is~370 km south of the tin deposits in the Tonod uplift. This 20 km × 30 km structure is mainly composed of volcanics within the Zhanok suite and by intrusive bodies of Bambukoy granites. The ID-TIMS zircon ages 834 ± 23 Ma and 818 ± 7 Ma [60] for volcanics and granites, respectively, coincide within error. The deposit consists of 10 tubular and two lenticular bodies of magnetite-feldspar metasomatites with cassiterite-chalcopyrite and magnetite-hematite mineralization. The main ore minerals are magnetite, hematite, mushketovite, cassiterite, chalcopyrite, bornite, pyrite, and rare stannite. The ore bodies are hosted by felsic rocks of the Bambukoy igneous complex that have geochemical features similar to those of S-type granites. Tin concentrations in the rocks are elevated (2.7-7.2 ppm Sn) but are not high enough to be classified as tin granites [60].
The age of the tin mineralization is constrained by the presence of magnetite/hematite pebbles with minor cassiterite in basal conglomerates of the Amatkansk suite of the Ediacarian (635-540 Ma) age [60]. During this study, we analyzed two cassiterite samples (S-19-7 and S-23-39) separated from magnetite-feldspar metasomatites of the Mokhovoe deposit.  The age of the tin mineralization is constrained by the presence of magnetite/hematite pebbles with minor cassiterite in basal conglomerates of the Amatkansk suite of the Ediacarian (635-540 Ma) age [60]. During this study, we analyzed two cassiterite samples (S-19-7 and S-23-39) separated from magnetite-feldspar metasomatites of the Mokhovoe deposit.

Valkumei Silicate-Sulfide Vein Tin Deposit
The significant deposits in Chukotka, Russian Northeast, are at Valkumei (presently being mined), Pyrkakay, Ekug, Telekai, Kukenei, Mramornoe, and Dioritovoe ( Figure 6). Most of the known Sn reserves in the region occur in stockworks of the porphyry Sn Pyrkakay deposit [62][63][64]. The Valkumei Sn silicate-sulfide vein deposit [65] used to serve as the principal source of tin ore for the Soviet Union. It consists of simple and complex veins, mineralized zones, and less common linear stockworks. The deposit occurs mainly within the marginal zone of the Late Cretaceous (108-105 Ma) small post orogenic granitic pluton ( Figure 6) and to a lesser degree in Cretaceous sandstone and shale that host the pluton [30,66]. This pluton belongs to the Chaun igneous complex, composed of quartz monzonites, monzogranites, granites, zinnwaldite granites and ongonites, and topazbearing albite-rich microgranites. Mineralization occurs in a north-northwest-trending zone alone the contact of the pluton.
ttps://doi.org/10.3390/xxxxx www.mdpi.com/journal/minerals Most of the known Sn reserves in the region occur in stockworks of the porphyry Sn Pyrkakay deposit [62][63][64]. The Valkumei Sn silicate-sulfide vein deposit [65] used to serve as the principal source of tin ore for the Soviet Union. It consists of simple and complex veins, mineralized zones, and less common linear stockworks. The deposit occurs mainly within the marginal zone of the Late Cretaceous (108-105 Ma) small post orogenic granitic pluton ( Figure 6) and to a lesser degree in Cretaceous sandstone and shale that host the pluton [30,66]. This pluton belongs to the Chaun igneous complex, composed of quartz monzonites, monzogranites, granites, zinnwaldite granites and ongonites, and topazbearing albite-rich microgranites. Mineralization occurs in a north-northwest-trending zone alone the contact of the pluton. Figure 6. Simplified geologic map showing the distribution of late Mesozoic magmatic bodies and tin ore deposits in Chaun ore district, Chukotka, the Russian Northeast (modified from Alekseev and Alekseev [35]).
The ore bodies commonly consist of a conjugate system of (1) major north-south veins and feathered veinlets, and (2) a zone of approximately east-west-and northwest-trending veins. The deposit hosts 70 minerals; the majority of the veins are composed dominantly of tourmaline with quartz, chlorite, albite, arsenopyrite, cassiterite, pyrrhotite, chalcopyrite, stannite, sphalerite, stibnite, fluorite, and various carbonates. The ore bodies are vertically extensive. The cassiterite-quartz-tourmaline veins are replaced by sulfide veins at depth. The deposit is large, was discovered in 1935, and has been mined from 1941 to the mid-1990s. The average grade is 0.4 to 1.2 percent Sn.
Here, we include data for coarse-grained cassiterite sample W-1 collected from this deposit for in situ U-Pb dating previously reported by Neymark et al. [13]. The ore bodies commonly consist of a conjugate system of (1) major north-south veins and feathered veinlets, and (2) a zone of approximately east-west-and northwesttrending veins. The deposit hosts 70 minerals; the majority of the veins are composed dominantly of tourmaline with quartz, chlorite, albite, arsenopyrite, cassiterite, pyrrhotite, chalcopyrite, stannite, sphalerite, stibnite, fluorite, and various carbonates. The ore bodies are vertically extensive. The cassiterite-quartz-tourmaline veins are replaced by sulfide veins at depth. The deposit is large, was discovered in 1935, and has been mined from 1941 to the mid-1990s. The average grade is 0.4 to 1.2 percent Sn.
Here, we include data for coarse-grained cassiterite sample W-1 collected from this deposit for in situ U-Pb dating previously reported by Neymark et al. [13].

Merek Greisen Tin Ore Deposit
The Russian Far East is a part of the Eastern Asia tin belt that extends from Indonesia in the south to the Chukchi Peninsula in the north. The northern part of the Eastern Asia tin belt is represented by five tin ore regions Chukotka, Kolyma, Yana-Indigirka, Khingan-Okhotsk, and Sikhote-Alin [67].
Most deposits are related to Cretaceous and Paleogene tin-bearing volcanic-plutonic complexes localized at an Andean-type transform or active continental margin [68]. The Sn deposits occurring in this belt are typical deposits related to ilmenite-series granitic activities [26,69].
The Merekski ore district (or Dusse-Alin ore district) is located at the eastern margin of the Khanka-Burea massif ( Figure 7) and occurs in the eastern contact zone of biotite and/or biotite-muscovite granites of the Dusse-Alin intrusive complex. It is a cassiteritequartz type containing quartz, muscovite and tourmaline, besides cassiterite, wolframite, molybdenite, and beryl. The Russian Far East is a part of the Eastern Asia tin belt that extends from Indonesia in the south to the Chukchi Peninsula in the north. The northern part of the Eastern Asia tin belt is represented by five tin ore regions Chukotka, Kolyma, Yana-Indigirka, Khingan-Okhotsk, and Sikhote-Alin [67].
Most deposits are related to Cretaceous and Paleogene tin-bearing volcanic-plutonic complexes localized at an Andean-type transform or active continental margin [68]. The Sn deposits occurring in this belt are typical deposits related to ilmenite-series granitic activities [26,69].
The Merekski ore district (or Dusse-Alin ore district) is located at the eastern margin of the Khanka-Burea massif ( Figure 7) and occurs in the eastern contact zone of biotite and/or biotite-muscovite granites of the Dusse-Alin intrusive complex. It is a cassiteritequartz type containing quartz, muscovite and tourmaline, besides cassiterite, wolframite, molybdenite, and beryl. About 320 km northeast of the Khinganskoe deposit, muscovite from the mineralized greisen in the biotite and/or biotite-muscovite granite of the Dusse-Alin intrusive complex (90 Ma [69]) within the Merek tin ore deposit was dated by the K-Ar method at 85.9 Ma [26], which is ~4 Ma younger than the host granite itself. We analyzed cassiterite sample MK-1 separated from a mineralized greisen at the Merek deposit. About 320 km northeast of the Khinganskoe deposit, muscovite from the mineralized greisen in the biotite and/or biotite-muscovite granite of the Dusse-Alin intrusive complex (90 Ma [69]) within the Merek tin ore deposit was dated by the K-Ar method at 85.9 Ma [26], which is~4 Ma younger than the host granite itself. We analyzed cassiterite sample MK-1 separated from a mineralized greisen at the Merek deposit.

Methods
The rocks were crushed to liberate the cassiterite fragments and crystals prior to handpicking at the Institute of Precambrian Geology and Geochronology (IPGG), St Petersburg, Russia. At the U.S. Geological Survey (USGS), handpicked cassiterite fragments or single large crystals were mounted in epoxy and imaged in transmitted and/or reflected light on a petrographic microscope. Cassiterite is opaque to translucent reddish brown in transmitted light depending on crystallinity and its chemical composition. Internal textures and some inclusions in opaque grains are difficult to identify by light microscope techniques so cathodoluminescence (CL) and backscatter electron (BSE) imaging was used to examine internal characteristics of individual grains. The textural features revealed by CL primarily originate from the substitution of Ti, Fe, and W for Sn (e.g., [70,71]). Titanium and W behave as luminescence activators and result in light CL features, in contrast to Fe, which quenches luminescence and results in dark features. The images were used to guide our selection of spots for in situ analyses to target inclusion-and fracture/vein-free parts of crystal fragments. Selected CL images of analyzed cassiterite samples are shown in Figure 8.
Minerals 2021, 11, x. https://doi.org/10.3390/xxxxx www.mdpi.com/journal/mine or single large crystals were mounted in epoxy and imaged in transmitted and/or reflec light on a petrographic microscope. Cassiterite is opaque to translucent reddish brown transmitted light depending on crystallinity and its chemical composition. Internal t tures and some inclusions in opaque grains are difficult to identify by light microsc techniques so cathodoluminescence (CL) and backscatter electron (BSE) imaging was u to examine internal characteristics of individual grains. The textural features revealed CL primarily originate from the substitution of Ti, Fe, and W for Sn (e.g., [70,71]). Titani and W behave as luminescence activators and result in light CL features, in contrast to which quenches luminescence and results in dark features. The images were used to gu our selection of spots for in situ analyses to target inclusion-and fracture/vein-free pa of crystal fragments. Selected CL images of analyzed cassiterite samples are shown in F ure 8. . This is a small portion of a~1 cm 2 grain exhibiting sector zoning towards the interior of the grain (at left), followed by fine-scaled oscillatory zoning, and then truncated by a zone of dark or diffuse CL activation (at right). Using BSE-EDS spot analyses, no mineral inclusions within the cassiterite were found.
U-Pb compositions of cassiterite were analyzed at the U.S. Geological Survey (USGS) in Denver, Colorado on a Nu Instruments AttoM ES™ sector field inductively coupled mass spectrometer (SF-ICPMS) coupled to a Teledyne Photon Machines Analyte™ 193 nm ArF excimer laser ablation system with a two-volume HelEx sample cell. Gas flows were optimized in line scan mode on NIST610 for maximum intensity, optimal peak shape, and low oxide production and Th/U fractionation.
Approximate U and Th concentrations (Tables S1 and S2) were obtained using the Iolite™ 2.5 [72] U-Pb geochronology Data Reduction Scheme (U_Pb_Geochronology3) using NIST 612 glass as the primary standard reference material for which U and Th concentrations of~37.4 and 37.8 ppm, respectively, were certified by the National Institute of Standards (NIST) [73]. NIST glasses can exhibit substantially different elemental fractionation behavior than many common geological matrices, resulting in degraded precision and accuracy for some elements [74,75]. Because of the matrix difference between the NIST 612 soda-lime glass and cassiterite, we assumed the U and Th cassiterite data to be not better than ±20% accurate based on results by Jochum and Stoll [76].
All data were collected with laser spot diameters of between 85-135 µm, a laser repetition rate of 5Hz and laser energy density of between 4.0 and 5.0 J/cm 2 . The NIST 612 glass standard was used as a primary non-matrix-matched standard to correct for instrumental Pb-isotope fractionation. A~1.54 Ga, low-Th, cassiterite sample SPG from a skarn deposit in Russian Karelia was used as a primary matrix-matched reference material to correct for the instrumental U/Pb bias. A~155 Ma Jian-1 cassiterite from the Jiangxi W-Sn district in the eastern Nanling Range, South China, was used as a secondary matrix-matched reference material to estimate external reproducibility of the LA-ICPMS U-Pb analyses. U-Pb ages for both SPG and Jian-1 samples have been measured by ID-TIMS [19], and our LA-ICPMS data for the SPG sample, presented in Table S1, did not show any statistically significant (beyond two standard deviations, 2 SD uncertainty) deviation from the ID-TIMS results (no systematic error). Replicate analyses (n = 12) of the secondary matrix-matched reference cassiterite Jian-1 show an external 2 SD reproducibility error of the SPG-normalized 206 Pb*/ 238 U lower intercept T-W isochron ages of 1.6% (Table S2).
Methods of the data collection and reduction used in this study were described in detail in Neymark et al. [13]. Our method assumes that there is no matrix effect on the Pb isotope ratios and therefore sample/standard bracketing using matrix-mismatched, well-characterized homogeneous NIST glass standard can be used to correct for the instrumental Pb isotope fractionation. The absence of an appreciable matrix effect on Pb isotopes during the LA-ICPMS analyses was shown in several publications (e.g., [77,78]). We used the Iolite 2.5 output for fractionation-corrected Pb isotope ratios in further offline data reduction. In contrast, the U/Pb ratios for the matrix-matched reference material from the Iolite output are biased from the "true" U/Pb values due to the matrix effect, which is assumed to be the same for the cassiterite reference and unknowns. Then, these biased 238 U/ 206 Pb ratios and the corrected NIST-normalized 207 Pb/ 206 Pb ratios from the Iolite output were used to calculate a biased U-Pb age of the matrix-matched reference material using Isoplot V4.15 software [79]. The difference between the biased and known "true" age of the matrix-matched material is used to derive correction factors F for the measured 238 U/ 206 Pb ratios in unknowns in each analytical session. Finally, these bias-corrected 238  The use of the age ratio to determine the correction factor "F" (F = T biased /T true , Figure 9A) is not accurate because the age is an exponential function of Pb/U ratios. A proper approach would be to calculate apparent radiogenic 238 U/ 206 Pb in the standard using the measured LA-ICPMS data and divide it by the 238 U/ 206 Pb ratio corresponding to the "true" ID-TIMS age of the standard. Chew et al. ( [82], Appendix A) proposed to correct within-session bias in the U-Pb ratios by calculating a ratio of the measured lower concordia 238 U/ 206 Pb* intercept to that of the true 238 U/ 206 Pb* ratio of the standard (F = R b1 /R True , Figure 9B). However, this method is fully accurate only for geologically young reference materials (<500 Ma) falling in the age range where T-W concordia is reasonably approximated by a horizontal straight line ( Figure 9A) and may give biased correction factor estimates for older references where the T-W concordia has a significant curvature ( Figure 9B). A more accurate approach was proposed by Roberts et al. [83] who suggested using F = X Rm /X Rc ( Figure 9B). This approach, similar to that of Chew et al. [82], requires an assumption about initial 207 Pb/ 206 Pb in the matrix-matched reference material. If we only use the parameters of its "raw" T-W isochron and the "true" age, the correction factor can be calculated as F = R b2 /R True ( Figure 9B).
LA-ICPMS results for unknowns are presented in Table S3. All age uncertainties in this paper are given as 2 SD (standard deviations) or 2 SE (standard error = SD/sqrt of the number of values). Uncertainties of U-Pb ages given in parentheses include an added in quadrature external 2 SD reproducibility error of 1.6%. This uncertainty estimate is based on 12 replicates of SPG-normalized U-Pb ages for the Jian-1 secondary matrix-matched reference cassiterite. Uncertainties of Pb-Pb inverse isochron ages for unknowns reported in this paper include an external 2 SD uncertainty of 0.6 percent that was derived from 103 replicate 208 Pb/ 206 Pb vs. 207 Pb/ 206 Pb inverse isochron ages of SPG reference cassiterite and added in quadrature to within-run age uncertainty of an unknown. The inverse Pb-Pb isochron approach was also applied to spot analyses in geologically "old" (>500 Ma) unknown samples with spot analyses with Th/ 206 Pb < 0.01. The details and limitations of this approach were described in Neymark et al. [13]. Figure 9. Procedure to determine the correction factor "F" of sample 238 U/ 206 Pb using the measured array of SPG cassiterite as a reference and assuming no matrix effect for Pb-isotope ratios (A). The difference between the true U/Pb bias and the bias calculated from the ratio of ages is shown in (B). This figure also shows data (open ellipses) for the SPG reference cassiterite measured in an analytical session. Table S3. Two plots showing ranges of U and Th concentrations in the samples are also included in that table. The concentrations in cassiterite are highly variable and range from ~0.1 to ~100 ppm and 0 to ~5 ppm for U and Th, respectively.

Replicate Analyses of the SPG Matrix-Matched Reference Cassiterite
Cassiterite sample SPG from a Px-Gar skarn in Pitkäranta ore district of the Russian Karelia was introduced by Neymark et al. [13] as a matrix-matched reference material for in situ LA-ICPMS U-Pb dating of cassiterite. This coarse-grained cassiterite has elevated U (tens of ppm) and very low Th (<0.01 ppm) contents and is suitable for inverse Pb-Pb isochron dating using 208 Pb as a Pbc proxy. Since then, we have conducted a total of 103 analytical sessions where this cassiterite was used as a matrix-matched reference; a summary of the results is presented in Table S1. An average 207 Pb*/ 206 Pb* (the asterisk denotes radiogenic Pb component) age value for 103 replicates is 1541.2 ± 9.3 Ma (2 SD). A Figure 9. Procedure to determine the correction factor "F" of sample 238 U/ 206 Pb using the measured array of SPG cassiterite as a reference and assuming no matrix effect for Pb-isotope ratios (A). The difference between the true U/Pb bias and the bias calculated from the ratio of ages is shown in (B). This figure also shows data (open ellipses) for the SPG reference cassiterite measured in an analytical session. Table S3. Two plots showing ranges of U and Th concentrations in the samples are also included in that table. The concentrations in cassiterite are highly variable and range from~0.1 to~100 ppm and 0 to~5 ppm for U and Th, respectively.

Replicate Analyses of the SPG Matrix-Matched Reference Cassiterite
Cassiterite sample SPG from a Px-Gar skarn in Pitkäranta ore district of the Russian Karelia was introduced by Neymark et al. [13] as a matrix-matched reference material for in situ LA-ICPMS U-Pb dating of cassiterite. This coarse-grained cassiterite has elevated U (tens of ppm) and very low Th (<0.01 ppm) contents and is suitable for inverse Pb-Pb isochron dating using 208 Pb as a Pb c proxy. Since then, we have conducted a total of 103 analytical sessions where this cassiterite was used as a matrix-matched reference; a summary of the results is presented in Table S1. An average 207 Pb*/ 206 Pb* (the asterisk denotes radiogenic Pb component) age value for 103 replicates is 1541.2 ± 9.3 Ma (2 SD). A weighted average age after rejecting five outliers is 1542.05 ± 0.71 Ma (2 SE, MSWD = 1.3). An average initial 207 Pb/ 206 Pb value from the Y-axis intercepts of T-W isochrons for 82 replicates after rejecting 4 outliers (for sessions with sufficient spread of data points that allowed calculating parameters of T-W isochrons) is 0.92 ± 0.15 (2 SD). A weighted average initial 207 Pb/ 206 Pb c value is 0.891 ± 0.011 (2 SE, MSWD = 2.7) in this cassiterite sample.

Vishnyakovskoe Deposit
Uranium concentrations in 60 spot analyses of cassiterite sample 2622 range from~0.2 to 39 ppm (median value is 5.8 ppm, Table S3). The U-Pb isotope data are scattered in the T-W Concordia diagram ( Figure 10A) thus indicating an open-system behavior and variable initial 207 Pb/ 206 Pb compositions (from~0.7 to~0.95). Selected 17 low Th/ 206 Pb (<0.01) spot analyses define the inverse Pb-Pb isochron age of 1849 ± 20 Ma ( Figure 10B). Analyses with radiogenic Pb (measured 207 Pb/ 206 Pb < 0.16) yield an anchored to Pb c 207 Pb/ 206 Pb = 0.95 ± 0.05 T-W isochron lower concordia intercept age of 1833 ± 31 Ma ( Figure 10C). A weighted average 207 Pb-corrected 206 Pb*/ 238 U age (using the S-K Pb c correction) for 47 spot analyses is 1835 ± 33 Ma ( Figure 10D). All three age estimates are within uncertainties of each other. The weighted average U/Pb age was calculated after eliminating nine outliers with younger apparent U-Pb ages which may indicate either radiogenic Pb loss or formation of younger cassiterite generation(s).
weighted average age after rejecting five outliers is 1542.05 ± 0.71 Ma (2 SE, MSWD = 1.3). An average initial 207 Pb/ 206 Pb value from the Y-axis intercepts of T-W isochrons for 82 replicates after rejecting 4 outliers (for sessions with sufficient spread of data points that allowed calculating parameters of T-W isochrons) is 0.92 ± 0.15 (2 SD). A weighted average initial 207 Pb/ 206 Pbc value is 0.891 ± 0.011 (2 SE, MSWD = 2.7) in this cassiterite sample.

Vishnyakovskoe Deposit
Uranium concentrations in 60 spot analyses of cassiterite sample 2622 range from ~0.2 to 39 ppm (median value is 5.8 ppm, Table S3). The U-Pb isotope data are scattered in the T-W Concordia diagram ( Figure 10A) thus indicating an open-system behavior and variable initial 207 Pb/ 206 Pb compositions (from ~0.7 to ~0.95). Selected 17 low Th/ 206 Pb (<0.01) spot analyses define the inverse Pb-Pb isochron age of 1849 ± 20 Ma ( Figure 10B). Analyses with radiogenic Pb (measured 207 Pb/ 206 Pb < 0.16) yield an anchored to Pbc 207 Pb/ 206 Pb = 0.95 ± 0.05 T-W isochron lower concordia intercept age of 1833 ± 31 Ma ( Figure  10C). A weighted average 207 Pb-corrected 206 Pb*/ 238 U age (using the S-K Pbc correction) for 47 spot analyses is 1835 ± 33 Ma ( Figure 10D). All three age estimates are within uncertainties of each other. The weighted average U/Pb age was calculated after eliminating nine outliers with younger apparent U-Pb ages which may indicate either radiogenic Pb loss or formation of younger cassiterite generation(s).

Tonod Uplift Tin Deposits
Isotope data for the Tonod uplift tin deposits were published in Neymark et al. [16]

Tonod Uplift Tin Deposits
Isotope data for the Tonod uplift tin deposits were published in Neymark et al. [16] and combined data for four cassiterite samples (418 spot analyses) are presented here. Uranium concentrations in spot analyses of cassiterite samples L-304, L-316, L-490, and L-528 range from~0.5 to 57 ppm (median value is 4.4 ppm, Table EA4 in Neymark et al. [16]. Like in the Vishnyakovskoe deposit the U-Pb data are scattered in the T-W diagram ( Figure 11A) indicating both potential variability in Pb c compositions and open U-Pb system behavior. Analyses with lower Th/ 206 Pb ratios (<0.01, n = 93) define an inverse Pb-Pb isochron age of 1861 ± 12 Ma (rounded off age value with an external uncertainty here and hereafter shown in parentheses, Figure 11B). Analyses with radiogenic Pb (measured 207 Pb/ 206 Pb < 0.13) yield an anchored to Pb c 207 Pb/ 206 Pb = 0.95 ± 0.05 T-W isochron lower concordia intercept age of 1846 ± 31 Ma ( Figure 11C). The weighted average 207 Pb-corrected 206 Pb*/ 238 U age (using S-K 1.85 Ga Pb c correction) for 395 spot analyses is 1631 ± 32 Ma ( Figure 11D). This weighted average U/Pb age is younger than the Pb-Pb and lower concordia intercept T-W isochron ages and indicates younger apparent ages for the significant portion of spot analyses due to either radiogenic Pb loss or formation of younger cassiterite generation(s).  Figure 11B). Analyses with radiogenic Pb (measured 207 Pb/ 206 Pb < 0.13) yield an anchored to Pbc 207 Pb/ 206 Pb = 0.95 ± 0.05 T-W isochron lower concordia intercept age of 1846 ± 31 Ma ( Figure 11C). The weighted average 207 Pb-corrected 206 Pb*/ 238 U age (using S-K 1.85 Ga Pbc correction) for 395 spot analyses is 1631 ± 32 Ma ( Figure 11D). This weighted average U/Pb age is younger than the Pb-Pb and lower concordia intercept T-W isochron ages and indicates younger apparent ages for the significant portion of spot analyses due to either radiogenic Pb loss or formation of younger cassiterite generation(s).

Pitkäranta Mining District Skarn Deposits
LA-ICPMS U-Pb data for the coarse-grained cassiterite sample SPG collected from a Px-Gar skarn were presented in Neymark et al. [13]; here we report data for five more cassiterite samples from that ore district (Table S3).

Pitkäranta Mining District Skarn Deposits
LA-ICPMS U-Pb data for the coarse-grained cassiterite sample SPG collected from a Px-Gar skarn were presented in Neymark et al. [13]; here we report data for five more cassiterite samples from that ore district (Table S3).
A coarse-grained cassiterite sample LA-1 separated from a quartz-carbonate metasomatite developed over a Px-Gar skarn (Old Mine Field) has much lower U concentration (median value 2 ppm U) and moderate Th (median 0.21 ppm Th). Lower U content in this sample results in most Th/ 206 Pb ratios of > 0.01. These values are too high and do not allow the usage of 208 Pb as a Pb c proxy in 208 Pb/ 206 Pb-207 Pb/ 206 Pb inverse Pb-Pb isochrons (see Neymark et al. [13] for the discussion). This sample yielded T-W isochron and weighted average 207 Pb-corrected ages of 1551 ± 28 Ma and 1546 ± 26 Ma, respectively Figure 14A,B).

Mokhovoe Deposit
Two samples of cassiterite separated from tin-bearing metasomatites have moderate U concentrations (median U 11.8 and 20.3 ppm in samples S-13-39 and S-19-7, respectively, Table S3). Data for both samples are scattered on the T-W diagrams ( Figure 15A,E) which may indicate open-system behavior and variable Pbc compositions.
Spot analyses of sample S-23-39 with most radiogenic Pb (n = 14, Figure 15B) define an anchored T-W isochron with lower-intercept age of 809 ± 13 Ma. A similar age of 813 ± 14 Ma is obtained for this sample from an inverse Pb-Pb isochron for selected 28 spot analyses with lower Th/ 206 Pb ratios ( Figure 15C). A weighted average 207 Pb-corrected U-Pb age is 811 ± 17 Ma ( Figure 15D).
Spot analyses of sample S-19-7 with most radiogenic Pb (n = 16, Figure 15G) define an anchored T-W isochron with lower-intercept age of 813 ± 21 Ma. A similar, within error age of 821 ± 27 Ma is obtained for this sample from an inverse Pb-Pb isochron for selected 28 spot analyses with lower Th/ 206 Pb ratios of <0.01 ( Figure 15F). A weighted average 207 Pbcorrected U-Pb age of this sample is 798 ± 19 Ma ( Figure 15H).

Mokhovoe Deposit
Two samples of cassiterite separated from tin-bearing metasomatites have moderate U concentrations (median U 11.8 and 20.3 ppm in samples S-13-39 and S-19-7, respectively, Table S3). Data for both samples are scattered on the T-W diagrams ( Figure 15A,E) which may indicate open-system behavior and variable Pb c compositions.
Spot analyses of sample S-23-39 with most radiogenic Pb (n = 14, Figure 15B) define an anchored T-W isochron with lower-intercept age of 809 ± 13 Ma. A similar age of 813 ± 14 Ma is obtained for this sample from an inverse Pb-Pb isochron for selected 28 spot analyses with lower Th/ 206 Pb ratios ( Figure 15C). A weighted average 207 Pb-corrected U-Pb age is 811 ± 17 Ma ( Figure 15D).
Spot analyses of sample S-19-7 with most radiogenic Pb (n = 16, Figure 15G) define an anchored T-W isochron with lower-intercept age of 813 ± 21 Ma. A similar, within error age of 821 ± 27 Ma is obtained for this sample from an inverse Pb-Pb isochron for selected 28 spot analyses with lower Th/ 206 Pb ratios of <0.01 ( Figure 15F). A weighted average 207 Pb-corrected U-Pb age of this sample is 798 ± 19 Ma ( Figure 15H).

Valkumei Deposit
Spot analyses of sample W-1 (n = 119, median U 17 ppm, data from Neymark et al. [13] form a T-W isochron ( Figure 16A) defining a lower-intercept age of 108.4 ± 2.2 Ma and a Y-axis intercept initial 207 Pb/ 206 Pb value of 0.840 ± 0.017. This initial 207 Pb/ 206 Pb value is within error of the S-K model value and the anchored T-W isochron age is 108.5 ± 1.7 Ma ( Figure 16B). A 207 Pb-corrected age is 108.3 ± 2.2 Ma ( Figure 16C). The large number of analyses for this sample was needed because very low U of 0.37 ppm in some spots resulted in elevated errors of the measured U/Pb ratios. The cassiterite age is similar to the age of the Cretaceous (108-105 Ma) Pevek granite-adamellite-granodiorite pluton [30,84] that is spatially associated with the deposit. An attempt to apply an inverse 208 Pb/ 206 Pb-207 Pb/ 206 Pb isochron to 45 spot analyses with Th/ 206 Pb < 0.01 (not shown) resulted in an imprecise age estimate of 117 ± 39 Ma. This result confirms a limited applicability of Pb-Pb dating to Mesozoic cassiterite samples.

Merek Deposit
Spot analyses of cassiterite sample MK-1 separated from a tin-mineralized greisen (n = 63, median U of 4.2 ppm) form a T-W isochron corresponding to a lower intercept age

Merek Deposit
Spot analyses of cassiterite sample MK-1 separated from a tin-mineralized greisen (n = 63, median U of 4.2 ppm) form a T-W isochron corresponding to a lower intercept age of 93.8 ± 2.1 Ma and initial 207 Pb/ 206 Pb ratio of 0.867 ± 0.027 ( Figure 16D). An anchored T-W isochron yields a similar age of 93.2 ± 2.0 Ma and a weighted average 207 Pb-corrected U-Pb age of this sample is 92.6 ± 1.9Ma ( Figure 16E,F). No attempt was undertaken to apply an inverse 208 Pb/ 206 Pb-207 Pb/ 206 Pb isochron to the data because only one spot analysis yielded Th/ 206 Pb < 0.01.

A Problem of Heterogeneous Isotopic Composition of Initial Pb in Cassiterite
Tapster and Bright [19] reported a high-precision ID-TIMS~285 Ma U-Pb age for cassiterite from the Cligga Head granite (Cornwall, UK). They interpreted their results as strongly suggesting variable Pb c isotopic composition and concluded that a simple binary mixture between a radiogenic and Pb c may not be expected within and between crystals. These authors postulated that some of the overdispersion on T-W isochron diagrams presented in Neymark et al. [13] could be attributed to variable Pb c isotopic compositions within cassiterite that are inherited from the hydrothermal systems. They also claimed that this complexity in the U-Pb systematics represents a potential limitation on the accuracy of the resulting age interpretation of cassiterite. This over-dispersion on T-W isochron diagrams based on LA-ICPMS data may be masked by low precision of individual data points but may bias the regression of a dataset away from the accurate lower intercept with concordia.
We tested this suggestion using replicate analyses of our SPG reference cassiterite (Table S1) and assumed that a variable isotopic composition of Pb c should cause some excess scatter of the results obtained in different analytical sessions. We indeed observed an excess scatter (MSWD = 2.7, Table S1) which is statistically significant beyond uncertainty for n = 82. Also, the SPG initial 207 Pb/ 206 Pb weighted average value of 0.891 ± 0.011 (2 SE) is significantly lower than the 207 Pb/ 206 Pb value of 1.004 ± 0.014 (2 SD, n = 30) in acid-leached feldspars from the Salmi granites that are considered to be genetically linked with the skarn mineralization [56]. It is also lower than the average value of 1.0104 ± 0.0062 (2 SD, n = 11) in galena from the skarns [25]. Additional cassiterite analyses for five samples from skarns in the Pitkäranta mining district also yielded variable estimates of 207 Pb/ 206 Pb c values of 0.854 ± 0.15, 0.961 ± 0.031, 1.023 ± 0.039, 1.031 ± 0.021, and 1.09 ± 0.21 derived from Y-axis intercepts of unconstrained T-W isochrons (Figures 12-14). Scattered data points on the T-W concordia diagrams for cassiterite samples from the Vishnyakovskoe deposit ( Figure 10A), deposits within the Tonod uplift ( Figure 11A), and the Mokhovoe deposit ( Figure 15A) may also indicate variable Pb c compositions in these samples.
The data indicate that Pb c in low-Pb cassiterite may be a mixture of Pb from a granitederived fluid and Pb present in situ in the host rocks. Incomplete mixing between the two Pb components during cassiterite formation may potentially cause excess scatter of U-Pb data on T-W isochron diagrams and result in biased lower concordia intercepts as suggested in [19]. One way to minimize the impact of the Pb c variability for age calculations is to use anchored T-W isochrons based on spot analyses showing the most radiogenic Pb isotope compositions. In this case, anchoring to the S-K model Pb c composition would not produce any appreciable age bias because the data points plot close to those of concordia.
The ID-TIMS data from Tapster and Bright [19] also allow calculation of a three-point inverse 204 Pb/ 206 Pb-207 Pb/ 206 Pb isochron age of 1540.9 ± 3.6 Ma age for the SPG cassiterite, which is within uncertainty with a weighted average of 1542.13 ± 0.76 Ma (2 SE) based on our replicate LA-ICPMS 208 Pb/ 206 Pb-207 Pb/ 206 Pb inverse isochron ages (Table S1). Although our LA-ICPMS Pb-Pb age is within error with the ID-TIMS results, it is older than the reported ID-TIMS U-Pb ages of 1536.6 ± 1.0 Ma [19] and 1539.5 ± 0.9 Ma [20]. These two reported U-Pb age values are based on three and two data points, respectively, and are different beyond the error limits. The small number of high-precision ID-TIMS analyses does not allow us to estimate a potential "geological" scatter of the data points and emphasizes the necessity for better characterization of potential matrix-matched reference materials for in situ cassiterite U-Pb analyses.

Summary of Cassiterite LA-ICPMS U-Pb Ages
A compilation of U and Th concentrations and U-Pb ages of cassiterite from several ore deposits in Russia, determined by in situ LA-ICPMS, is presented in Table 1. During this work we analyzed nine new samples from five ore deposits located in different parts of Russia from the Baltic Shield in the west to the Far East. All these new samples showed a wide range of U concentrations (ppm to tens of ppm U, similar to previously published data also included in Table 1) and variable degrees of Pb c concentrations; nonetheless, all were datable by the in situ LA-ICPMS method. Older Paleoproterozoic samples (the deposits of Vishnyakovskoe, East Sayan pegmatite belt and within the Tuyukan ore region, Baikal fold belt) showed open U-Pb system behavior, but selected analyses with the most radiogenic Pb isotopic compositions allowed for reliable U-Pb and Pb-Pb age determinations. The in situ LA-ICPMS ages of 1848 ± 17 Ma (Pb-Pb) to 1835 ± 33 Ma (U-Pb) for cassiterite from the Vishnyakovskoe deposit are within uncertainty of the 1838 ± 3 Ma ID-TIMS age of manganotantalite from this deposit [48]. Published Rb-Sr ages of~1.5 Ga for pegmatites of the Vishnyakovskoe deposit [49] are~350 Ma younger than the U-Pb age of 1838 ± 3 Ma of manganotantalite separated from the ore [48] and of the 1.83-1.85 Ga age of cassiterite (this work). The observed disturbance of the Rb-Sr isotope system may be caused by processes that also caused the observed partial disturbance of U-Pb isotope systems in cassiterite.
Similar Paleoproterozoic ages of 1861 ± 12 Ma (Pb-Pb) and 1846 ± 31 Ma (U-Pb) are obtained for combined cassiterite data for four samples from tin deposits in the Tuyukan ore region located in the Baikal-Patom fold-thrust belt. These ore deposits also occur in a region with protracted geologic history and observed disturbance of U-Pb isotope systems in cassiterite from these deposits may be caused by superimposed metamorphic processes [16]. Some degrees of U-Pb systems disturbance were also found in cassiterite samples from the Neoproterozoic Mokhovoe deposit in the Muya district where younger superimposed magmatic processes were dated by U-Pb ID-TIMS method at~280 Ma [85].
Cassiterite samples from Mesoproterozoic skarn deposits in the Pitkäranta Mining District show no obvious signs of open-system behavior and essentially all the spot analyses were usable for age calculations. Similarly, much younger cassiterite from Cretaceous deposits Valkumei and Merek in the Russian North East and Far East, respectively, shows well-behaved closed U-Pb systematics.
Recalculated published data for cassiterite spot analyses of sample SPG from the Old Mine Field of the Pitkäranta Mining district yield an average age value of 1541.2 ± 9.3 Ma (2 SD) for 208 Pb/ 206 Pb-207 Pb/ 206 Pb inverse isochron ages determined in 103 analytical sessions. Similar ages were obtained during this study for two additional cassiterite samples from this ore field (Samples 31 and LA-1, Table 1) and for three cassiterite samples from the ore deposit Kitelä (samples O-4-11, VSH-5, and X-146, Table 1). The age for sample LA-1, separated from a quartz-carbonate metasomatite developed over a Px-Gar skarn (Old Mine Field), is within uncertainty of the other samples from the skarn deposits in this mining district. The large age uncertainties do not allow determining a potential time gap between the skarn formation and the superimposed metasomatic process and indicate that it would not exceed~20-30 Ma. Sample LA-1 has much lower U concentration compared to other samples from the Pitkäranta skarns, which agrees with previously described lower U content in recrystallized (younger) cassiterite that was observed in the Sullivan SEDEX deposit in Canada [59].
The data discussed in this paper show that the inverse 208 Pb/ 206 Pb-207 Pb/ 206 Pb isochron approach can be successfully applied to dating older (Proterozoic) cassiterite using spot analyses with low (<0.01) Th/ 206 Pb ratios. Considering the highly variable Th concentrations in cassiterite, a large number of spot analyses was needed to discover the ones with suitable low Th/ 206 Pb. The T-W isochron approach also required screening of the data to select analyses with the most radiogenic Pb isotopic compositions. The spot analyses plotting close to the T-W Concordia lower intercept are less sensitive to potential Pb c heterogeneity and can be used to calculate more accurate anchored T-W isochron ages and 207 Pb-corrected 206 Pb*/ 238 U ages.
The deposits studied in this paper differ in age, type (greisen, skarn, vein, etc.), composition, in the nature of the associated igneous rocks, and in their geodynamic settings. All of the deposits are spatially and genetically related to granitic plutons. The cassiterite age of 1834-1848 Ma from the Vishnyakovskoe deposit is very close to the age of the associated granite of the Sayan complex (1869 ± 6 to 1855 ± 5 Ma [48]). This deposit also contains economically significant amounts of Ta and Nb [46]. Cassiterite from tin deposits of the Tyukan ore region is dated at 1.85-1.86 Ga. The deposits are hosted by thẽ 1.85 Ga Chuya-Kodar tin-bearing S-type granitic rocks [16,40]. Tin-bearing metasomatites of the Mokhovoe deposit are within 820-830 Ma host rocks of the Bambukoy igneous complex which are geochemically similar to S-type granites [60]. Tin-bearing skarns of the Pitkäranta mining district are closely spatially and genetically related to rapakivi granites of the Salmi batholith [25,58]. Tin ores in the Mesozoic Valkumei and Merek deposits are spatially and genetically associated with granitic rocks of the Pevek pluton and Dusse-Alin intrusive complex, respectively. In all the studied deposits, LA-ICPMS U-Pb ages of cassiterite are within error of the available ages of associated granitic rocks; however, the large LA-ICPMS age uncertainties do not allow determination of a potential time gap between the ore formation and magmatic processes. If this gap existed, it would not exceed 30 Ma for the Vishnyakovskoe deposit and deposits in the Tuykan ore region, 10-15 Ma for the Pitkäranta Mining District and Mokhovoe deposits, and 2-3 Ma for Cenozoic deposits Valkumei and Merek.

Conclusions
LA-ICPMS U-Pb and Pb-Pb ages of cassiterite from six ore deposits from Russia located in the Baltic Shield, Eastern Siberia, Russian North East, and the Russian far east range from~1.85 Ga to~93 Ma. Regardless of highly variable U concentrations and common Pb abundances, cassiterite produced reliable geochronological information in all the studied ore deposits. The Tera-Wasserburg isochron approach was applied to samples in the whole age range, while inverse 208 Pb/ 206 Pb-207 Pb/ 206 Pb isochrons were used for deposits older than~0.8 Ga deposits. In several cases, excess scatter of data points was observed on the Tera-Wasserburg diagrams, indicating open system behavior (radiogenic Pb loss), formations of younger cassiterite generations, and variable isotopic composition of the common Pb. The best examples of closed U-Pb systems were found in cassiterite from skarn deposits in the Pitkäranta Mining district. These cassiterites are very low in Th and moderately enriched in U which makes them potentially good candidates as matrix-matched references for in situ U-Pb dating of cassiterite. The inverse Pb-Pb isochron approach that uses 208 Pb as a common Pb proxy is applicable to spot analyses with Th/ 206 Pb < 0.01 for which an input of radiogenic 208 Pb is negligible. In many cases, a large number (>50) of spot analyses were needed to select the data with the lowest Th/ 206 Pb values needed for the inverse Pb-Pb isochrons and to select the most radiogenic measured 207 Pb/ 206 Pb for Tera-Wasserburg isochrons to minimize potential impacts of heterogeneous Pb c isotopic composition.
All studied ore deposits are spatially associated with granitic igneous rocks. The cassiterite ages are within error of the ages of the dated granites; however, the errors of the in situ LA-ICPMS U-Pb and Pb-Pb age estimates are too large to determine potential age gaps between the intrusions and the ore-forming metasomatism. In many cases the main source of the elevated uncertainties in the LA-ICPMS age determinations is an external reproducibility of secondary matrix-matched standards, which is not better than~1.6% for U-Pb ages and~0.6% for Pb-Pb ages. This is the main limitation of LA-ICPMS U-Pb dating for studies requiring precise geochronological results. However, the low cost and high throughput of the method make it an attractive choice for solving many geological problems that do not require very high precision of age determinations.