Measurements of Sonoluminescence Temporal Pulse Shape

Sonoluminescence light emission

“…2͑b͔͒ reflects the characteristic asymmetry of the dynamics, i.e., the fall time of the emitted radiance is longer than the rise time. 40 The ''black'' bubble's pulse width, however, is well outside the range of experimental observations [39][40][41][42] -with Ϸ500 ps it is too long. While shorter pulses can be obtained with smaller driving pressures and radii, e.g., Ϸ275 ps FWHM with P a ϭ1.2 atm and R 0 ϭ3.0 m, the average pulse widths are clearly overestimated in the blackbody model.…”

“…A small difference like this, amounting to Ϸ16 ps here ͑at very different absolute intensities͒, can hardly be detected in today's experiments. [40][41][42] Moran and Sweider 42 do find longer pulses in the red, but only for lower water temperatures, where the bubbles can be driven at larger P a and R 0 ͑cf. Ref.…”

“…The most simple explanation, namely, thermal blackbody radiation, has been discarded on two accounts: ͑i͒ blackbody radiation would be coupled to the dynamics of the temperature in the bubble, i.e., it is expected to change on the time scales of bubble dynamics, which from theoretical calculations seem to be longer than the observed pulse widths. [39][40][41][42] ͑ii͒ Gompf et al, 39,40 and later Hiller et al, 41 experimentally found the pulse width to be independent of the wavelength of the radiation. Moran and Sweider 42 find only a rather weak dependence in some parameter regimes.…”

“…41 While Gompf et al 39 and Hiller et al 41 could not detect any spectral variation of the pulse width, a small increase toward the red wavelengths is reported by Moran and Sweider. 42 Within the present formalism, it is easy to integrate the spectral radiance over suitable wavelength intervals and to compare with the experimental results. Figure 2͑b͒ shows a simulation of the light pulses from a strongly collapsing bubble treated as a blackbody ͑param-eters P a , R 0 as in Fig.…”

“…In several experiments these pulse widths have been measured with filters of bandwidth ⌬ Ϸ100 nm 39 and ⌬ Ϸ40 nm, 42 or alternatively with a spectrographic procedure. 41 While Gompf et al 39 and Hiller et al 41 could not detect any spectral variation of the pulse width, a small increase toward the red wavelengths is reported by Moran and Sweider.…”