I studied the electric behaviour of little crystals of Cadmium Sulfide (typically 1 mm), a light sensitive semiconductor compound that can be used in photocells and similar electronic devices. After having attached metal electrodes I applied high voltages and measured the associated electric current that runs through the device. Simple as that
.
At the start, increasing the voltage leads to a proportial increase of the electric current (Ohm's law). At a certain voltage (up to one thousand Volts), however, the current simply refuses to increase any further: it saturates. It 's like pushing down your car's accelerator further and further while your car refuses to go any faster. Annoying as such.
Taking a closer look at the electric current, it appears to have become pretty unstable, not to say that it shows hugh and irregular fluctuations. Although noise phenomena are pretty normal in nature, the magnitude of my current noise is quite extreme: a many thousands times stronger than ordinary noise levels. Indeed, a kind of noise catastrophe. Note that these current variations are extremely fast: we're talking about radio frequencies here, which means up to a billion times per second. It 's really weird.
It turned out that these curious phenomena originates in the ordinary thermal vibrations that occur in the crystal lattice. I succeeded to find an explanation by framing a theory, which incorporates a very special interaction between electricity and mechanical distortions: this so-called piezo-electric coupling also exists in gas igniters (squeeze and spark).
In my crystals, which were known to be piezo-electric in kind, ordinary thermal vibrations of the crystal lattice highly interact with the electric current, which causes the vibrations to increase dramatically in magnitude.
These huge acoustic waves, in turn, force electrons to move only with the velocity of sound instead of higher drift velocities. Clearly, electrons get trapped in amplified acoustic waves and are forced to move only with the velocity of sound: therefore the electric current saturates.
In addition, the amplified thermal acoustic waves, which remain pretty chaotic as they reflect random motion of the crystal lattice, force the moving electrons to copy this random motion. When the acoustic waves display extreme large amplitudes, the electric current shows huge vibrations as well: large groups of electrons are suddenly captured (and slowed down) or are released again when the wave desintegrates.. There you are: a current noise catastrophe!
To investigate the amplified acoustic waves I tried to make them collide with laser light. And it worked! Why? To make light collide with sound you have to make sure that the mechanical structure (the sound wave) is of the same order of magnitude as the applied wave length of the light. For instance, radio waves (wavelength typical 10 meters or more) can easily pass the brick walls of your room, because the waves hardly notice such small objects. In contrast, light waves (up to 1/1000 of a millimeter) are reflected or absorbed by the wallpaper.
There is an alternative way to explain the ability of light to collide with sound. As you may know, light waves can also be described as particles (photons). Accordingly, acoustic waves are called phonons. Now, a collision between a photon and a phonon strongly ressembles the collision of billiard balls: due to the collision the velocities of the balls change in direction and magnitude. Similarly, in my crystal after the collision the outcoming laser light has a different wavelength and direction. Comparing the outcoming laser light with the original light yields the direction and wavelength (frequency) of the sound waves, which obviously is the cause og the change.
Checked and double checked: it worked. I succeeded to measure the soundwaves (unhearable though: up to 1.000.000.000 vibrations per second). I heard my crystal sing by measuring scattered light intensities!