The Haynes-Shockley experiment allows students to measure the drift mobility of electrons and holes in semiconductors. It is an experiment with great educational value, because it allows direct investigation of the drift velocity, of the diffusion process and of the recombination of excess charge carriers.
In our new setup the excess carriers are optically injected (using internal photoelectric effect) avoiding the need of a reliable point-contact emitter.
Consider a P-doped semiconductor bar, of length l, with ohmic contacts soldered at both ends Inside the sample an electric field (named sweep field Es) is temporarily produced by a
Two point contacts (electrodes E and C) are made by two metal needled separated by a distance d. By applying to the electrode E (emitter) a short negative pulse voltage with an amplitude large enough to forward bias the diode DE, electrons will be injected into the crystal region underlying the emitter.
This electron pulse will drift, under the electric field action, with velocity vd, and after some time it will reach the region underlying the electrode C (collector).
When the excess electron pulse reaches the point contact C, the minority charge carrier density is locally increased, thus increasing the inverse current and producing a voltage drop
across the resistance R.
On the oscilloscope screen we may observe a first short negative pulse, with amplitude comparable to that of the injection pulse and, after some delay t a second negative pulse,
wider and much smaller than the first one. The first peak is simultaneous with the injection pulse: it is due to the electromagnetic signal propagating across the sample (at the light speed).
The second pulse corresponds to the excess electon distribution passing under the collector contact: its shape is approximately gaussian and its amplitude and width are
determined by diffusion and recombination processes.
The actual pulse shape depends on the drift time t, on the covered distance d, on the drift velocity: vd=μEs, where μ is the electron mobility, and also on the diffusion constant D. The measurement of the time of flight t and of the distance d between the fiber and the point contact gives the drift velocity vd: vd = d/t.