A new method for measuring the Hall mobility of semiconductor films is developed, the principle of which is based on the theory of the microwave Faraday rotation: that is, the total rotation angle θtotal equals θs+θƒ, where θs and θƒ are angles of Faraday rotation by a substrate and a film respectively. Then the Hall mobility μƒ of the film is calculated from a simple formula of θƒ. where μ0 is the permeability of free space, ω is the angular frequency, tf and σf are the thickness and the dc conductivity of the film, and B is the magnetic field. Hall mobilities of germanium wafers with n-type epitaxial films of about 10μ in thickness and of 0.2_??_0.5 Ω-cm in resistivity are estimated at 0.20m2/v-sec, with the error of about 10%, This value is in good agreement with dc data. The new technique has the advantage of requiring no electrode attachment.
An experimental study is made on alloyed silicon p-n junctions which have breakdown voltages in the range of 0.5 to 30 volts. Most of the experiments are carried out on the cross-section made across the junction. When the junctions are biased in the reverse direction to breakdown, two types of visible light are emitted from the cross-section of their junction planes. Spectrum of the light is analyzed by the use of a double monochrometer, and the results are found more in con-formity with the theory than those obtained by Chynoweth et al. Besides the above two types of light emission, a new type of emission is observed on a sample with rectangular cross-section lapped by #3000-mesh alundum. The spectra of all these emissions are almost the same as the spectrum observed on a sample made by evaporating aluminum on an n-type silicon wafer. In both cases, the decay time of emissions is less than 0.1μs. From these results it may be concluded that these emissions are not caused by the breakdown in p-n junction but by recombination of high energy carriers in silicon. The spectrum obtained by Chynoweth et al. seems to be in line with that of these newly found emissions; the emission by carrier recombination in Si other than in p-n junction seems likely to be involved in high energy components in their results.
The pulses the GM counter produce in intense pulsed photon beams are higher than those normally observed at the same applied counter voltage. These pulses are formed by superposition of the electrons generated in proportion to the beam intensity onto the electrons by characteristic discharge of the counter. Theoretical derivation made to express these composite pulses is verified experimentally by the use of an analogue computer and counters of several kinds exposed to photon beam from 4 MeV electron linear accelerator. Furthermore, some tubes were found to produce several output pulse patterns in an extremely intense field.
When a liquid drop strikes vertically upon a linear grid made of equally spaced steel needles, the drop splits into a number of droplets regularly distributed along the direction parallel to the row of needles. If the nature of liquid as well as experimental conditions are not adequate, however, the above mentioned regular splitting does not occur; in some condition, droplets themselves recombine immediately after splitting; in another one, the drop splits into a numerous fine dropletss irregularly scattered. Successive stages of such splitting phenomena are observed by means of instantaneous photography. Water, milk, benzine, ethyl alcohol, methyl alcohol and aqueous solustions of surface-active agent are used. By dimensional analysis, an attempt is made to find the critical condition for the occurrence of regular splitting. Denoting two characteristic numerics by R=dvρ/η and W=lv2_??_/γ, where, d: diameter of the grid needle, v: striking velocity, ρ: density of the liquid, l: diameter of the drop, η: viscosity, γ: surface tension, we obtain the said critical condition expressed approximately by the relation R·S=Constant, and when the magnitude of this constant in C. G. S. units lies between 1/2·105_??_1/2·106, regular splitting of the drop occurs. Furthermore, by collating to the characteristic features of splitting drop revealed by instantaneous photography, the relation W/R>1 or <1 seems to characterize whether the major factor affecting the phenomenon is viscosity or surface tension.
Any automatic Device for plotting physical quantities of fields-electrostatic or electromagnetic field and others-can be studied as a time-variable gain control system, for the over-all gain of the system varies with the distributed intensity of gradient of the field at the point where the detector is placed impairing the stability and constancy of the measurement. To cope with this shortcoming, application of automatic gain control system is tried. As an example, the auto-matic equipotential line plotter provided with an electrolytic tank is taken up, and a variable gain servo-amplifier of amplification factor devised to vary in inverse proportion to the over-all gain of the system is applied. Another example is the electron trajectory plotter to which a servo-mecha-nism that controls the scale factor is applied. These measures may be applied to other kinds of plotter if the detecting arrangement is appropriately chosen. Significance of the problem, details of the apparatus, and results of the experiments are discussed.