Titanite Spectroscopy and In Situ LA-ICP-MS U–Pb Geochronology of Mogok, Myanmar

: With the development of mineral testing technology and ore deposit geochemistry, titanite has become a hot topic in the study of accessory minerals. Two large-grained titanite crystals from Mogok, Myanmar, were used for a detailed study. In this study, the standard gemmological properties and spectral characteristics of titanite crystals were obtained by Fourier transform in-frared, micro ultraviolet-visible-near-infrared and Raman spectroscopy, respectively, which pro-vide a full set of data. Mineral major and trace elements were analysed using Electron-Probe Mi-croAnalysis (EPMA) and Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS). The purpose of this study is to report spectral characteristics and major and trace elements of Mogok, Myanmar, in order to ﬁnd new potential titanite standard samples. The two titanite crystals have similar major element compositions, and both grains have relatively low Al content (0.011–0.014 apfu) and Al/Fe ratios (0.157–0.222), but high Fe content (0.063–0.079 apfu). The two titanite crystals have similar chondrite-normalised rare earth element (REE) patterns with signiﬁcantly Light Rare Earth Element (LREE) (La–Gd) enrichment and deletion of Heavy Rare Earth Element (HREE) (Tb–Lu). The 238 U/ 206 Pb ages of the two titanite samples are 43.5 ± 5.8 Ma and 34.0 ± 4.2 Ma, respectively. Generally, magmatic titanite has a low Al/Fe ratio, metamorphic and hydrothermal titanite crystals have extremely low Th/U ratios close to zero, with ﬂat chondrite-normalised REE patterns or depletions in light REEs relative to heavy REEs. Different genetic types of titanite can be distinguished by the characteristics of major and trace elements. Combined chemical features such as REE differentiation, Al/Fe and Th/U ratios with formation temperature, the analysed titanite samples are considered magmatic-hydrothermal titanites. Their 238 U/ 206 Pb ages may indicate a potential stage of magmatic hydrothermal conversion.

In this study, in situ major trace element compositions of two titanite single crystals in Mogok, Myanmar, have been studied through electron probe microanalysis (EPMA) and LA-MC-ICP-MS. Spectral features are obtained and compared with the calibration spectra in the RRUFF database [30]. Our aims were to establish potential young titanite standard samples and reveal the genesis of titanite.

Geological Setting
The Mogok area in Myanmar is a world-famous gem-producing area (Figure 1a) [31][32][33], which, extending in the north-south direction, is primarily composed of metamorphic rocks, mixed rocks and late Mesozoic granitic intrusive rocks (Figure 1b) [34][35][36]. The Mogok metamorphic zone was formed by the northward subduction of the Indian plate in 71 Ma and the collision between the Eurasian and Indian plate at approximately 50 Ma [37]. Previous studies show that deep crustal melting occurred with ages ranging from 20-15 Ma in this region, and this process was accompanied by granite intrusion. Precious age determinations indicate that metamorphism along the MMB occurred during 68-21 Ma with peak between the Middle Eocene and the early Oligocene [38][39][40][41]. The Mogok titanite samples can meet the requirement of younger standard samples.

Materials and Methods
Two Mogok titanite crystals (MG-5 and MG-6) were examined using standard gemmological techniques. The specific gravity (SG) of the samples was obtained using a hydrostatic weighing method. Spectroscopy tests for the samples were conducted at the Gemmological Research Laboratory of China University of Geosciences (Beijing) to obtain their spectral properties.
The titanite grains were mounted in epoxy, polished and examined using BSE images to select suitable targets for in situ analysis. The TESCAN field emission scanning electron microscope (MIRA 3LMH) was used to capture BSE images with the following setting: acceleration voltage: 7 kV; absorption current: 1.2 Na; scan time: 80 s.
Electron-Probe MicroAnalysis (EPMA) was conducted in the Laboratory of EPMA, China University of Geosciences (Beijing) with the following setting: acceleration voltage: 15 kV; electic current: 10 Na; beam spot diameter: 1 µm.
An Agilent 7900 Q-ICP-MS instrument coupled to a 193-nm ArF excimer laser ablation system was used to determine trace element compositions and U-Pb ages in the Laboratory of Mineral Laser Microzone Analysis, China University of Geosciences (Beijing). The Ontario standard was used for calibration, and the MKED1 standard was used as a standard reference.

Visual Appearance and Gemmological Properties
The titanite samples are dark brown and translucent with medium cleavage and greasy lustre (Figure 2a,b). The MG-6 sample has crystal plane steps. Triangular etching on the crystal surface can also be observed. Some weak areas of MG-5 sample are

Materials and Methods
Two Mogok titanite crystals (MG-5 and MG-6) were examined using standard gemmological techniques. The specific gravity (SG) of the samples was obtained using a hydrostatic weighing method. Spectroscopy tests for the samples were conducted at the Gemmological Research Laboratory of China University of Geosciences (Beijing) to obtain their spectral properties.
The titanite grains were mounted in epoxy, polished and examined using BSE images to select suitable targets for in situ analysis. The TESCAN field emission scanning electron microscope (MIRA 3LMH) was used to capture BSE images with the following setting: acceleration voltage: 7 kV; absorption current: 1.2 Na; scan time: 80 s.
Electron-Probe MicroAnalysis (EPMA) was conducted in the Laboratory of EPMA, China University of Geosciences (Beijing) with the following setting: acceleration voltage: 15 kV; electic current: 10 Na; beam spot diameter: 1 µm.
An Agilent 7900 Q-ICP-MS instrument coupled to a 193-nm ArF excimer laser ablation system was used to determine trace element compositions and U-Pb ages in the Laboratory of Mineral Laser Microzone Analysis, China University of Geosciences (Beijing). The Ontario standard was used for calibration, and the MKED1 standard was used as a standard reference.

Visual Appearance and Gemmological Properties
The titanite samples are dark brown and translucent with medium cleavage and greasy lustre (Figure 2a,b). The MG-6 sample has crystal plane steps. Triangular etching on the crystal surface can also be observed. Some weak areas of MG-5 sample are dissolved into small pits by corrosion, forming regular shaped pits on the crystal surface ( Figure 3). dissolved into small pits by corrosion, forming regular shaped pits on the crystal surface ( Figure 3). The SG values of MG-5 and MG-6 are 3.51 and 3.56, respectively. Since the refractive index of titanite varied between 1.89 and 2.02, the index of refraction exceeded the refractometer's value and could not be measured. The titanite samples do not change colour under a Chelsea colour filter. The titanite samples have a rare earth spectrum, as seen through spectroscopes, with multiple fine absorption lines and 580 nm double lines.     The SG values of MG-5 and MG-6 are 3.51 and 3.56, respectively. Since the refractive index of titanite varied between 1.89 and 2.02, the index of refraction exceeded the refractometer's value and could not be measured. The titanite samples do not change colour under a Chelsea colour filter. The titanite samples have a rare earth spectrum, as seen through spectroscopes, with multiple fine absorption lines and 580 nm double lines.   The SG values of MG-5 and MG-6 are 3.51 and 3.56, respectively. Since the refractive index of titanite varied between 1.89 and 2.02, the index of refraction exceeded the refractometer's value and could not be measured. The titanite samples do not change colour under a Chelsea colour filter. The titanite samples have a rare earth spectrum, as seen through spectroscopes, with multiple fine absorption lines and 580 nm double lines.

UV-Vis Spectrum
The colour origin of the titanite samples was analysed by UV-Vis spectra. MG-5 and MG-6 are brown titanite samples. The UV-Vis spectrum of MG-6 has an obvious absorption peak centred at 565 nm in the range of 550-600 nm, weak absorption peaks at 700-800 nm and a wide and slow absorption band at 220-300 nm ( Figure 5). The yellowgreen absorption band with 565 nm as the centre in the range of 550-600 nm is caused by the electron transition between Ti 4+ and Fe 4+ .

UV-Vis Spectrum
The colour origin of the titanite samples was analysed by UV-Vis spectra. MG-5 and MG-6 are brown titanite samples. The UV-Vis spectrum of MG-6 has an obvious absorption peak centred at 565 nm in the range of 550-600 nm, weak absorption peaks at 700-800 nm and a wide and slow absorption band at 220-300 nm ( Figure 5). The yellow-green absorption band with 565 nm as the centre in the range of 550-600 nm is caused by the electron transition between Ti 4+ and Fe 4+ .

UV-Vis Spectrum
The colour origin of the titanite samples was analysed by UV-Vis spectra. MG-5 and MG-6 are brown titanite samples. The UV-Vis spectrum of MG-6 has an obvious absorption peak centred at 565 nm in the range of 550-600 nm, weak absorption peaks at 700-800 nm and a wide and slow absorption band at 220-300 nm ( Figure 5). The yellow-green absorption band with 565 nm as the centre in the range of 550-600 nm is caused by the electron transition between Ti 4+ and Fe 4+ .

Major and Trace Elements
Major element compositions of the titanite samples are presented in Table 1

Major and Trace Elements
Major element compositions of the titanite samples are presented in Table A1. Previous studies have demonstrated that REEs replace Ca and Zr in titanite and high field strength elements such as Nd will displace Ti (Al, Fe) in the titanite lattice. The positive correlation of Nd-Zr in the samples shows that the two elements exhibit the same substitution characteristics (Figure 8).

Major and Trace Elements
Major element compositions of the titanite samples are presented in Table A1. Previous studies have demonstrated that REEs replace Ca and Zr in titanite and high field strength elements such as Nd will displace Ti (Al, Fe) in the titanite lattice. The positive correlation of Nd-Zr in the samples shows that the two elements exhibit the same substitution characteristics (Figure 8). Previous studies have demonstrated that REEs replace Ca and Zr in titanite and high field strength elements such as Nd will displace Ti (Al, Fe) in the titanite lattice. The positive correlation of Nd-Zr in the samples shows that the two elements exhibit the same substitution characteristics (Figure 8).

Titanite U-Pb Ages
The titanite ages with common Pb [45] were calculated using the weighted mean of the 207 Pb-corrected ages and the Tera-Wasserberg (TW) Concordia intercept age anchored through common Pb. Titanite U-Pb isotope results and ages are listed in Tables 4 and 5. Overall 20 LA-ICP-MS analyses of each titanite sample were performed in different sections (Figure 9a,b). On the TW diagram, the common Pb-uncorrected data of MG-5 define a linear array, yielding a lower-intercept age of 44 ± 17 Ma (n = 20, 2σ, MSWD = 0.63) and a y-intercept of initial 207 Pb/ 206 Pb of 0.848 (Figure 10a,b). The common Pb-uncorrected data of MG-6 define a linear array, yielding a lower-intercept age of 34 ± 14 Ma (n = 20, 2σ, MSWD = 0.75) and a y-intercept of initial 207 Pb/ 206 Pb of 0.843 (Figure 10c,d). On the basis of this common Pb composition, a common Pb correction was performed using the method of fitting 207 Pb/ 206 Pb c . All analyses of MG-5 yielded a weighted average 206 Pb/ 238 U age of 43.5 ± 5.

Titanite U-Pb Ages
The titanite ages with common Pb [45] were calculated using the weighted mean of the 207 Pb-corrected ages and the Tera-Wasserberg (TW) Concordia intercept age anchored through common Pb. Titanite U-Pb isotope results and ages are listed in Tables A4 and  A5. Overall 20 LA-ICP-MS analyses of each titanite sample were performed in different sections (Figure 9a,b). On the TW diagram, the common Pb-uncorrected data of MG-5 define a linear array, yielding a lower-intercept age of 44 ± 17 Ma (n = 20, 2σ, MSWD = 0.63) and a y-intercept of initial 207 Pb/ 206 Pb of 0.848 (Figure 10a,b). The common Pb-uncorrected data of MG-6 define a linear array, yielding a lower-intercept age of 34 ± 14 Ma (n = 20, 2σ, MSWD = 0.75) and a y-intercept of initial 207 Pb/ 206 Pb of 0.843 (Figure 10c

Titanite U-Pb Ages
The titanite ages with common Pb [45] were calculated using the weighted mean of the 207 Pb-corrected ages and the Tera-Wasserberg (TW) Concordia intercept age anchored through common Pb. Titanite U-Pb isotope results and ages are listed in Tables A4 and  A5. Overall 20 LA-ICP-MS analyses of each titanite sample were performed in different sections (Figure 9a,b). On the TW diagram, the common Pb-uncorrected data of MG-5 define a linear array, yielding a lower-intercept age of 44 ± 17 Ma (n = 20, 2σ, MSWD = 0.63) and a y-intercept of initial 207 Pb/ 206 Pb of 0.848 (Figure 10a

Comparision with RRUFF Database
The RRUFF database contains eight titanite standard samples from Brazil, the USA, Pakistan, Canada and Mexico. R050114 with yellowish-brown fragments from Pakistan was chosen for comparison. Fourier transform infrared spectra of titanite R050114 have

Comparision with RRUFF Database
The RRUFF database contains eight titanite standard samples from Brazil, the USA, Pakistan, Canada and Mexico. R050114 with yellowish-brown fragments from Pakistan was chosen for comparison. Fourier transform infrared spectra of titanite R050114 have 416.5, 559.3 and 848.5 cm −1 characteristic absorption peaks in the fingerprint region ( Figure 11), indicating that these characteristic peaks occur in the range of peaks caused by different bonds; however, the positions of these characteristic peaks are different. Raman spectra of titanite R050114 have 252, 316, 423, 605, 873, 880 and 1177cm −1 characteristic peaks ( Figure 12). The characteristic peak at 605 cm −1 is consistent with that of the analysed samples MG-5 and MG-6.

Formation Temperature of Titanite
The Al2O3 content of titanite was used to estimate the formation pressure of titanite [47], and a pressure-dependent Zr in titanite geothermometer [48] was used to estimate the formation temperature of titanite. High temperatures enable more Zr to enter the structure of titanite. The Al2O3 content of titanite increases with pressure (P) according to

Formation Temperature of Titanite
The Al 2 O 3 content of titanite was used to estimate the formation pressure of titanite [47], and a pressure-dependent Zr in titanite geothermometer [48] was used to estimate the formation temperature of titanite. High temperatures enable more Zr to enter the structure of titanite. The Along with the weighted average 206 Pb/ 238 U ages of these two samples, the calculated temperature range is consistent with the geological background of their ages ( Figure 14). However, there is a significant difference in the calculation results of temperature, which may be caused by the inaccuracy of cogenetic minerals in the calculation formula.

Formation Temperature of Titanite
The Al2O3 content of titanite was used to estimate the formation pressure of titanite [47], and a pressure-dependent Zr in titanite geothermometer [48] was used to estimate the formation temperature of titanite. High temperatures enable more Zr to enter the structure of titanite. The Al2O3 content of titanite increases with pressure (P) according to the following: P (in MPa) = 101.66 × Al2O3 in titanite (in wt%) + 59.013 (R 2 = 0.83). The pressures of MG-5 were estimated to be 88. 39 Along with the weighted average 206 Pb/ 238 U ages of these two samples, the calculated temperature range is consistent with the geological background of their ages ( Figure 14). However, there is a significant difference in the calculation results of temperature, which may be caused by the inaccuracy of cogenetic minerals in the calculation formula.

Genesis of Analysed Titanite
The classification of the genetic types of titanite requires comprehensive consideration of the major and trace elements. In general, magmatic titanite has a low Al/Fe ratio, whereas metamorphic titanite, including hydrothermal titanite, has a high Al/Fe ratio. Both grains have been plotted in the igneous field of Kowallis et al. (1997) (Figure 15).

Genesis of Analysed Titanite
The classification of the genetic types of titanite requires comprehensive consideration of the major and trace elements. In general, magmatic titanite has a low Al/Fe ratio, whereas metamorphic titanite, including hydrothermal titanite, has a high Al/Fe ratio. Both grains have been plotted in the igneous field of Kowallis et al. (1997) (Figure 15). Metamorphic and hydrothermal titanite crystals commonly have extremely low Th/U ratios close to zero, along with flat chondrite-normalised REE patterns or depletions in light REEs relative to heavy REEs. In contrast, two analysed titanite grains show heavy light REE (La-Gd) to weak heavy REEs (Tb-Lu) differentiation [49].
The Th/U ratio is also an indicator of the origin of titanite. In general, hydrothermal titanite has a lower ratio of Th/U (mostly < 1) than magmatic titanite. In this study, the two samples yielded a similar ratio of Th/U (MG-5: 1.30-1.92; average: 1.61; MG-6: 1.08-1.57; average: 1.37). Thus, titanite that crystallises under high-temperature hydrothermal conditions would also have the characteristics of magmatic titanite Th/U > 1. To discrimi- Metamorphic and hydrothermal titanite crystals commonly have extremely low Th/U ratios close to zero, along with flat chondrite-normalised REE patterns or depletions in light REEs relative to heavy REEs. In contrast, two analysed titanite grains show heavy light REE (La-Gd) to weak heavy REEs (Tb-Lu) differentiation [49].
The Th/U ratio is also an indicator of the origin of titanite. In general, hydrothermal titanite has a lower ratio of Th/U (mostly < 1) than magmatic titanite. In this study, the two samples yielded a similar ratio of Th/U (MG-5: 1.30-1.92; average: 1.61; MG-6: 1.08-1.57; average: 1.37). Thus, titanite that crystallises under high-temperature hydrothermal conditions would also have the characteristics of magmatic titanite Th/U > 1. To discriminate the origin of titanite using a single ratio of Th/U indicator is inaccurate.
However, considering that the calculated formation temperatures were lower and consistent with that of the hydrothermally modified titanite, EMPA data have shown low concentrations of La in the analysed titanite samples (Table 5), which is commonly related to late hydrothermal activity [50], and our samples can be recognized as products of magmatic hydrothermal conversion [27,51,52].
Analysed titanite samples are interpreted as magmatic-hydrothermal.

Conclusions
We have investigated the chemical composition, structure and genesis of titanite crystals from Mogok, Myanmar, through Fourier infrared, UV-Vis and Raman spectroscopy, EPMA and LA-MC-ICP-MS. These results are used to assess the potential of titanite as reference material for micro-analytical dating. In this paper, the analysis and summary of mineralogical and spectral characteristics based on the samples provide characteristics of a new production area of titanite not found in the RURFF database. According to U-Pb dating analysis, age data have a high concordant degree with weighted average 206 Pb/ 238 U ages 43.5 ± 5.8 Ma and 34.0 ± 4.2 Ma, respectively, which indicates the potential stage of magmatic hydrothermal conversion. Thus, these two grains can be used as potential stand samples for U-Pb dating analysis. Major and trace element analyses of titanites can be used to discuss the genetic type and explore their geological background.

Data Availability Statement:
The data presented in this study are available within this article.

Acknowledgments:
We thank the editor and reviewers for constructive comments which helped in improving our paper. This is the 8th contribution of B.X. for the National Mineral Rock and Fossil Specimens Resource Center.

Conflicts of Interest:
The authors declare no conflict of interest.