Elimination of bandwidth effect in attenuation measurement with picosecond ultrasonics

Abstract

We study the broadening effect of probing pulse light on the apparent attenuation of the Brillouin oscillation measured with picosecond ultrasonics. We observe experimentally that the attenuation of the Brillouin oscillation is sensitive to the bandwidth, and the apparent attenuation coefficient increases as the bandwidth increases, being far from the intrinsic attenuation coefficient. Theoretical calculation is performed to reconstruct the observed oscillations, and it is confirmed that there are several factors affecting the apparent attenuation in addition to the bandwidth. We finally propose equations that deduce the contribution of the broadening to the apparent attenuation of the Brillouin oscillation.

Intrinsic coherent acoustic phonons in the indirect band gap semiconductors Si and GaP

“…The observation indicates that the reflectivity oscillation is damped not because the acoustic pulses move out of the probed region, since the probing depth (2α) −1 varies by a factor of 5 for Si and more than 1000 for GaP over the tuning range of λ. We therefore attribute the damping of the oscillations in the two-color measurements to the dephasing caused by the broad bandwidth of the probe light [43]. We will confirm this by varying the bandwidth in our theoretical modeling in Sect IV B.…”

“…This is because the above expressions are for monochromatic probe light, whereas in our experiments the femtosecond probe pulses have finite spectral bandwidths, 0.11 and 0.030−0.044 eV (half-width at half-maximum) for one-and two-color pump-probe schemes, respectively. For such broadband pulses, dephasing among the reflectivity oscillations at different probe wavelengths can no longer be neglected [43]. We therefore calculate ∆R/R by taking into account the intensity profile I pr (k) of the broadband probe light: Table II, for both the pump and probe interactions.…”

Time-domain Brillouin scattering assisted by diffraction gratings

“…This fact in combination with the opportunities to monitor multiple different BS processes/frequencies simultaneously would be advantageous in studying the dispersion of the sound velocity and attenuation in many materials. It is worth noting here that the applications of usual TDBS schemes for studying acoustic wave attenuation are documented for a variety of the media [36][37][38][39][40][41][42][43][44][45][46]. The opportunity to monitor in a single measurement the acoustic phonons propagating in different directions could be attractive for revealing the elastic/inelastic anisotropy of materials, including one which could be caused by nonisotropic loading or by the residual stress.…”

“…It is worth noting here that the applications of usual TDBS schemes for studying acoustic wave attenuation are documented for a variety of the media. [30][31][32][33][34][35][36][37][38][39][40] The opportunity to monitor in a single measurement the acoustic phonons propagating in different directions could be attractive for revealing the elastic/inelastic anisotropy of the materials, including one that could be caused by non-isotropic loading or by the residual stress. For the studies of the anisotropy it is also extremely advantageous that gratings, as demonstrated by our experiments, can simultaneously launch phonons, detectable by TDBS, in the complete diapason of the angles, i.e., from 0 degrees to 90 degrees relative to their surface.…”

Coherent optical and acoustic phonons generated at lattice-matched GaP/Si(0 0 1) heterointerfaces

“…This can be understood in of the larger probing depth of the visible probe light, which makes the two-color-scheme detection less sensitive to the acoustic dyamics in the GaP layer. The dephasing rate Γ, summarized in figure 7(b), does not systematically depend on E or d. The insensitivity of Γ to E implies the damping of interference patterns due to the dephasing among the oscillations at different wavelengths within the broadband probe light [58] rather than due to the acoustic pulses moving out of the probing depth (2α) −1 , which critically depends on E as shown in figure 1(a). Calculations of the strain-induced reflectivity changes of bulk Si by taking into account the finite probe spectral width (0.03-0.04 eV) reproduced the experimentally observed damping and the frequency quantitatively [40].…”