In any sound wave the pressure is oscillating up and down relative to the ambient pressure. If the amplitude of
the sound is large enough the pressure on the negative part of the oscillation will become below zero. In our
experiments we use ultrasound generated by a piezoelectric transducer to make negative pressures. To make a
transient negative pressure within a small volume of the liquid we can use a hemispherical transducer which
produces sound of frequency typically between 100 kHz and a few MHz. This is shown schematically in the figure.
The sound comes to a focus at the center of the hemisphere giving a large pressure oscillation with a region of
dimensions comparable the sound wavelength. If the pressure oscillation is large enough any electron within this
region will become unstable and grow rapidly. Light from a helium-neon laser is focused into the same region as
the sound. The light is scattered from the surface of the expanded bubble and is detected by the photomultiplier
By adjusting the sound frequency and the duration of the sound pulse in the right way, it is possible to make the
bubble large enough to be seen by eye. Here is a photograph of such a bubble taken by Clare Cramer.
The sound transducer described above produces a large negative pressure in just a very small volume of the liquid.
Only if an electron happens to be in this volume, will it be detected. In order to make a movie of an electron we
have to make a transducer which will generate a sound pulse which will explode an electron located anywhere inside
a large volume, i.e., a volume of several cc. This requires a planar non-focusing transducer. To make such a transducer
is challenging because of the high amplitude that is required; many of the transducers we have used have broken due
to the high stress. We have had best success with lithium niobate but are planning to test other materials which may