Compositional dependence of molar absorptivities for infrared absorption bands of H2O in rhyolitic to peridotitic glasses

Le résumé fourni par la source

Fourier transform infrared spectroscopy is widely used for H 2 O measurements in natural and synthetic silicate glasses. Previous calibrations focused on relatively polymerized compositions from rhyolitic to basanitic (non-bridging oxygen per tetrahedral cation ratios (NBO/T): 0–1.4), while we extend this range to highly depolymerized peridotitic glasses (NBO/T: 2.2–2.9). Bulk H 2 O contents of 0.3 to 4.7 wt% were determined by elastic recoil detection analysis, a reliable direct method for absolute hydrogen quantification; IR calibration used 10 samples with 1.8–4.7 wt%, excluding 0.3 wt%. We determined integrated molar absorption coefficients for the OH band at 4500 cm −1 ( ε * OH = 124 ± 10 L mol −1 cm −2 ) and derived those for the H 2 O band at 5200 cm −1 assuming ε * OH / ε * H2O = 0.83 from the literature ( ε * H2O = 103 ± 8 L mol −1 cm −2 ). Additionally, we determined linear and integrated molar absorption coefficients for the fundamental O H stretching vibration band at 3530 cm −1 , and its first overtone at 7030 cm −1 . We compiled and critically assessed over 350 published molar absorption coefficients, creating the largest available database. This database includes glass compositions, water contents with analytical techniques, and baseline subtraction procedures, enabling identification of methodological biases and consistency across studies. Combined with our measurements, these data reveal that molar absorption coefficients for all four infrared bands correlate systematically with compositional parameters such as SiO 2 mol.% and decrease with increasing depolymerization. These trends, supported by theoretical considerations, enable prediction of molar absorption coefficients as a function of glass composition and accurate water quantification across a wide range of silicate glasses.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.