Index of content:
Volume 18, Issue 3, 01 March 1947
18(1947); http://dx.doi.org/10.1063/1.1697648View Description Hide Description
18(1947); http://dx.doi.org/10.1063/1.1697649View Description Hide Description
Knowledge of the radar response of wires or thin metallic strips, as a function of their length and thickness, and of the radar frequency is important in the design of reflectors for radar. In view of the difficulty of this theoretical problem and the necessity of making approximations, as well as the dearth of adequate experimental data, two independent procedures for solution are presented. Detailed quantitative results are obtained for the angular dependence of the cross section, and also for the mean cross section, of randomly‐oriented wires or, more generally, of metallic strips, which behave electromagnetically like cylindrical wires of a certain ``equivalent radius.'' When expressed in terms of a unit of area equal to the square of the wave‐length, these cross sections depend on the dimensions of the wire only through the two ratios.>The mean cross section is shown to take on maximum values when 4l/λ is slightly less than an integer (n = 1, 2, etc.). The shift of these ``resonances'' from integral values depends on the ratio 2l/a, becoming greater as 2l/a decreases. The value of at resonance increases slowly with the order n of the resonance; it depends only very slightly on the ratio 2l/a, increasing as 2l/a decreases. For values of 4l/λ away from resonance, decreases rapidly, reaching minimum values near 4l/λ = 3/2, 5/2, etc. The value of at these minima is strongly dependent on 2l/a, increasing as 2l/a decreases. Also as 4l/λ increases, the heights of the minima increase and approach the height of the resonance peaks. A brief comparison with preliminary experimental results is given.
X‐Ray Scattering at Small Angles by Finely‐Divided Solids. I. General Approximate Theory and Applications18(1947); http://dx.doi.org/10.1063/1.1697650View Description Hide Description
The general approximate theory of x‐ray scattering at small angles by finely divided solids is reviewed. It is assumed that the x‐ray scattering data can be interpreted in terms of a particle size distribution, with particle‐to‐particle scattering negligible. Scattered intensity curves for Maxwellian, Gaussian, and rectangular size distributions are calculated, and several procedures for obtaining a size distribution from the experimental scattering data are described. Details of the experimental technique are given, and evidence is presented to show that appreciable error can result if crystal‐monochromated radiation is not used. General procedures are given for applying corrections to the experimental data for the slit geometry used. Scattering data, mass distribution curves, and average particle sizes are presented for amorphoussilicagels and for crystalline oxides of alumina, nickel and iron. These data are shown to correlate well with the results of crystal size measurements by x‐ray diffraction line broadening and with specific surface measurements. The average particle size values determined from the experimental data are shown to have considerable significance, at least for the type of materials considered here.
X‐Ray Scattering at Small Angles by Finely‐Divided Solids. II. Exact Theory for Random Distributions of Spheroidal Particles18(1947); http://dx.doi.org/10.1063/1.1697651View Description Hide Description
The theory of the scattering of x‐rays at small angles is given for a continuous distribution in size of randomly spaced and oriented spheroidal particles having arbitrary, but fixed, shape. Families of scattering curves are presented for spheres, and for spheroids ranging in shape from flat disks to long rods. Both Maxwellian and rectangular types of particle mass distribution are used. A fit between the experimental and a calculated scattering curve enables one, under favorable circumstances, to determine the mass distribution in the test sample. However, an unambiguous interpretation of the experimental scattering curve on the basis of the small angle scattering theory is not possible without additional evidence from independent investigations of such quantities as particle shape and sample specific surface.
18(1947); http://dx.doi.org/10.1063/1.1697652View Description Hide Description
The magnetron oscillator of some frequency‐modulation radars also furnishes the ``local oscillator'' excitation for the crystal mixer of the receiver. Excess noise generation by the magnetron was observed to reduce greatly the receiver sensitivity. This noise exhibited a strong dependence upon anode voltage and current, and changed with time in a perplexing manner. After many experiments, a hypothesis of the cause of excess noise was developed, and further experiments confirmed this hypothesis. The noise is thought to be caused by ionization of atoms of the cathode oxide coating, which atoms are removed from the cathode by electron bombardment. In order to reduce the generation of excess noise, and still preserve the advantages of an oxide‐coated cathode, a special shape of cathode has been developed. The coated regions of this cathode are sheltered from electron bombardment, and the noise is much reduced. Use of cathodes of this general type should also produce magnetrons with longer operating lives than present tubes.
18(1947); http://dx.doi.org/10.1063/1.1697653View Description Hide Description
A new form of Fourier series for crystal structureanalysis is developed and a graphical method of summation described. The procedure is designed to permit simple and fairly rapid computation and recording of the density contributions of all planes at all points in the unit cell. Its particular advantage is in the case of centrosymmetrical projections where phases must be assigned experimentally. Adjustment of phases involves only re‐addition of numbers rather than a complete new summation. The method also requires a minimum of equipment which is easily constructed at negligible expense.
18(1947); http://dx.doi.org/10.1063/1.1697654View Description Hide Description
Breakdown studies have been made between electrodes in high vacuum at constant voltages from 50 to 700 kv. These further demonstrate the inadequacy of the field emissiontheory to account generally for high voltage breakdown in vacuum. Experiments are described which investigate some of the ``total voltage'' breakdown mechanisms, including positive‐ion emission by electron impact, electron emission by positive‐ion impact and by photons. In the d.c. case these processes contribute to a steady interchange of charged particles between cathode and anode which increases with voltage until breakdown ensues. At higher breakdown voltages the cathode gradient has diminished far below the value for field emission. Measurements of electron emission by electrons with energies up to 300 kv for tungsten, steel, aluminum, and graphite are reported. The possibilities of predicting and of improving the insulating strength of electrode gaps in high vacuum by the study of the coefficients of the electrode materials are discussed.