Structural relaxation in sputter-deposited silica glass

Correlated evolution of structure and mechanical loss of a sputtered silica film

“…In summary, IBS deposition leads to a dense and internally-stressed silica film. This outcome is consistent with previous studies showing that IBS and ion-assisted deposition yield highly-densified silica coatings [28,[31][32][33][34]. Thanks to internal stress, IBS silica relaxes towards a less dense structure even if T ≪ T g .…”

“…ω MB shifts from 476.7 ± 0.5 cm −1 at t a = 0 h to 465.5 ± 0.5 cm −1 at t a = 300 h, without reaching a plateau. This corresponds to an increase ∆θ = 0.9 • , which is coherent with FTIR measurements of sputtered silica coatings showing that θ increases with annealing [28], and can be related to a less dense structure. HWHM MB increases along with t a , denoting a widening of the θ distribution: HWHM MB is 69 ± 5 cm −1 at t a = 0 and 85±5 cm −1 at t a = 300 h. For comparison, HWHM mb ∼ 120 ± 10 cm −1 for fused silica.…”

“…Thus the annealed film is less homogeneous in terms of θ values than the as-deposited film. The normalized D 2 area decreases monotonically, following a stretched exponential law [28] with τ = 6.3 ± 1.1 h and β = 0.36 ± 0.05. This relaxation time is relatively short: magnetron-sputtered silica deposited at 520 • C has the same relaxation time at about 800 • C, and follows the Arrhenius law with activation energy of 5.4±0.2 eV [28].…”

“…The normalized D 2 area decreases monotonically, following a stretched exponential law [28] with τ = 6.3 ± 1.1 h and β = 0.36 ± 0.05. This relaxation time is relatively short: magnetron-sputtered silica deposited at 520 • C has the same relaxation time at about 800 • C, and follows the Arrhenius law with activation energy of 5.4±0.2 eV [28]. Considering also that relaxation times of fused silica (with fictive temperature T f = 1100 • C) are 6 · 10 4 times longer [28], we might say that the deposition temperature is more effective than the annealing temperature in stabilizing sputtered silica.…”

“…Considering also that relaxation times of fused silica (with fictive temperature T f = 1100 • C) are 6 · 10 4 times longer [28], we might say that the deposition temperature is more effective than the annealing temperature in stabilizing sputtered silica. Unfortunately, those data are not conclusive since relaxations in [28] were observed in the infra-red reflectivity (stretching mode of Si-O-Si) whereas our data concern the D 2 evolution, and the fact that in [28] β < 0.3 indicates the presence of a broad spectrum of relaxation mechanisms.…”