Low‐temperature measurements of the Hall effect have been used to determine donor and acceptor concentrations and in some cases to identify the major impurity in silicon. A simple apparatus for making these measurements with liquid helium is described. Typical results are presented and correlated with lifetime data at room temperature. Interpretation of the Hall coefficient measurements is discussed, and a simplified analysis of the case of the two types of donor impurity is suggested.

Hall measurements at room temperature have been used to determine the average net carrier density in polycrystalline silicon. The method does not require cutting of samples and destruction of the rods. Current contacts are made to the ends of a rod with strips of metal foil; Hall contacts are made with two titanium blades which close on the rod. Hall measurements made this way on single-crystal zone-refined rods agree with measurements on samples cut from the rods in the conventional manner. The average carrier densities measured in the polycrystalline rods have been correlated with the resistivities of these same rods after one, two, three floating zone passes. Measurement of Whole IngotsThe conventional bridge-shaped samples used for Hall measurements require that an ingot be sawed up and thus partially destroyed. It would obviously be advantageous to make Hall measurements on an ingot of semiconductor material without cutting it up (1). This was done with the apparatus ~ shown in Fig. 1. The holder was made of Lucite and was clamped between the poles of a Varian 4-in. electromagnet. The sample (which was a round ingot) was placed in the V-shaped blocks of Lucite and rested on a knife-edge titanium blade which formed one Hall contact. A pivoted Lucite arm carrying the other titanium blade Hall contact was placed on top of the sample and was held down with a nonmagnetic (lead) weight. The pivoted arm could be moved laterally by an adjusting screw to align the top blade accurately above the lower blade; this adjustment eliminated the IR voltage between the Hall contacts, so that the measured Hall voltage was zero when the magnetic field was zero. Current contacts to the sample were made on the ends of the ingot either by ultrasonic soldering with indium, by