Efeitos da radiação no mineral zircão: Caracterização via espectroscopia micro-Raman e espalhamento de raios X em pequenos ângulos (SAXS)

Abstract

Radiation damage in zircon (ZrSiO₄) fundamentally compromises its reliability as a geochronometer by altering the crystalline structure through accumulation of α-recoil damage from uranium and thorium decay. This thesis presents an integrated investigation of radiation damage assessment, thermal annealing behavior, and nanoscale track morphology in zircon, employing complementary analytical techniques and advanced statistical methodologies to establish robust protocols for Raman-based thermochronology applications. The first component addresses a critical methodological gap in Raman spectroscopy preprocessing for radiation damage quantification. Twelve preprocessing combinations were systematically evaluated across six vibrational modes using 80 zircon spectra from radiation-damaged samples. Performance assessment employed fitting quality metrics and statistical validation with outlier detection protocols, algorithms (polynomial, spline, and comparing three baseline correction iterative AirPLS) combined with four normalization methods (min-max, area, peak, and vector). Results demonstrate that preprocessing effectiveness is highly mode-dependent: the ν₃(SiO₄) symmetric stretching mode at 1008 cm⁻¹ achieved superior fitting quality and precision with Spline baseline correction and Min-Max normalization, while ν₁ and ν₂(SiO₄) vibrational modes performed optimally with Polynomial baseline correction and area normalization. These findings establish standardized preprocessing protocols that enhance inter-laboratory comparability and analytical precision for U-Pb geochronology applications. The second component investigates crystalline recovery dynamics during isothermal annealing of metamict zircon through Raman spectroscopy and advanced statistical modeling. A dataset of 6,225 valid spectral observations was analyzed using Generalized Additive Models (GAM), segmented regression with Differential Evolution optimization, and paired bootstrap methods to test six fundamental hypotheses. Results confirm non-linear model superiority and reveal a two-phase recovery characterized by an initial phase (Stages I-II, 473-840 K) with preferential external mode recovery and higher variability (CV ratio = 1.56), followed by a final phase (Stage III) where SiO₄ tetrahedral modes dominate and variability converges (CV ratio = 0.92). Empirical breakpoints at 607 K and 840 K validate Geisler’s three-stage phenomenological model, while bootstrap analysis reveals non-monotonic behavior in band area ratios with a peak at 622 K. These quantitative constraints provide a foundation for Raman-based low-temperature thermochronology. The third component integrates Small-Angle X-ray Scattering (SAXS) measurements to characterize nanoscale ion track morphology in swift heavy-ion irradiated zircon. SAXS analysis using the hard cylinder model with Gaussian polydispersity yielded track radii ranging from 3.32 to 4.99 nm across seven samples. The correlation between SAXS track parameters and Raman FWHM is not simply linear: the sample with the largest track radius (4.99 nm) exhibits the lowest FWHM (7.38 cm⁻¹), whereas the sample with the smallest radius (3.32 nm) presents a FWHM of 13.05 cm⁻¹, indicating that reduced track radii do not necessarily imply lower vibrational disorder. This behavior reflects the fact that the two techniques respond to structurally distinct aspects of radiation damage — short-range vibrational disorder and physical nanoscale track morphology, respectively — with direct implications for the calibration of Raman parameters in fission track thermochronology. Taken together, the standardized Raman preprocessing protocols, statistical modeling of recovery kinetics, and structural characterization by SAXS offer a starting point for the standardization of Raman-based zircon analysis in thermochronological contexts, although their generalization depends on validation with independent datasets and varied chemical compositions. It should be noted that Transmission Electron Microscopy (TEM), included in the original research scope and reflected in the title, could not be performed due to the non-return of samples sent to the Australian Synchrotron for SAXS characterization, which also precluded the TEM stage; this absence does not affect the results obtained by Raman spectroscopy and SAXS.