Volume 17, Issue 7, 01 July 1946
Index of content:
17(1946); http://dx.doi.org/10.1063/1.1707751View Description Hide Description
Shadows of shock waves in water were photographed by means of a point‐spark of less than one‐tenth micro‐second duration; the waves were formed when high velocity spheres struck the water surface. The shock wave front had the form of an elliptical arc of small eccentricity; the ellipticity was caused by the greater strength and velocity of the wave at its center. The absence‐of‐light band had a varying width which depended on the strength of the wave. A large number of secondary waves behind the impact wave were observed; these were produced by the vibrating spherical missile as it traveled through the water. The period of vibration was determined from the spacing of the waves for 4/32″, 6/32″, and 8/32″ spheres; these agreed to within 19 percent of the periods calculated on the basis of Lamb's theory. Interference of the vibration waves was also observed when two spheres were shot into the water simultaneously. The cavity behind the missile was nearly conical in shape with the forward tip being distorted by pressure changed immediately ahead of the missile. A sphere produced a cusp at the forward tip of the cavity; cylinders and cones each produced their own characteristic shadow pattern. When a grid was placed in front of the tank, distortions in its shadowgram were observed. These were caused by the high pressures surrounding the missile and, from the shift in the grid pattern, the direction and magnitude of the pressure gradient could be inferred. When the shock waves were reflected with the wave front nearly normal to the surface there was an indication that the stem, which is characteristic of Mach reflections, was present. All other reflections were regular. Focusing of the wave by a brass mirror was demonstrated in the spark shadowgram. Waves reflected near normal incidence from substances having a lower acoustic impedance than water were reflected with a presence‐of‐light band ahead of the usual absence‐of‐light band. At glancing incidence this reversal occurred even when the reflecting substance had a higher acoustic impedance. Striations frequently appeared in the shadowgrams when waves were reflected. These striae were from 0.02 to 0.08 cm apart and represented a quasi‐stationary state. They were believed to be formed when the reflected vibration waves passed through the incident waves.
17(1946); http://dx.doi.org/10.1063/1.1707752View Description Hide Description
A lightning stroke to the propeller blades of an aeroplane in flight left a remarkably complete record of its own characteristics. Sputtering of the discharge on the rotating blades left a clear time scale of duration of the stroke. The lightning current direction was also recorded by magnetization of steel rings inherent in the propeller mechanism.
17(1946); http://dx.doi.org/10.1063/1.1707753View Description Hide Description
Dielectric loss is measured by Q determinations of a resonator partially filled with the sample. Double sample technique is used to eliminate dominant spurious losses. Metal loss is reduced by confining field mostly to sample. Detuning for Q determination is accomplished by large movement of small rod. Detuning is calculated by action theorem.
17(1946); http://dx.doi.org/10.1063/1.1707754View Description Hide Description
Discharge tubes are described in which a low pressure glow discharge is initiated by means of a magnetic field pulse. A tube has been developed in which a glow discharge so triggered leads to the formation of a cathode spot on a mercury pool; this tube will hold off 20 kv until fired and will then pass a current pulse of several hundred amperes. Its use in radar modulator circuits is envisaged.
17(1946); http://dx.doi.org/10.1063/1.1707756View Description Hide Description
Fluorescent lamps with thin and with partial coatings of phosphor on bulbs of ultraviolet transmitting glass were irradiated with Hg 2537 radiation. Comparison with similar lamps which were burned showed a decline in lumen output which was of the same general character in all experiments, except for different time factors. This indicates that the process of lamp depreciation is fundamentally of photochemical nature, involving excited mercury, but not requiring either Hg ions or neutral atoms. It is assumed that superficial oxygen ions of the phosphor crystal react with Hg* to form links of the type‐SiO‐O‐Hg‐Hg..., thus tying mercury to the crystal with a resulting loss of phosphorbrightness.
17(1946); http://dx.doi.org/10.1063/1.1707757View Description Hide Description
An elegant way of studying the geometrical meaning of a Harker synthesis is to transform it into a diagram here termed the implication diagram. This is done by transformation of polar coordinates, the characteristics of the transformation depending on the number of operations in the cyclical group of the axial symmetry for which the Harker synthesis is prepared. The implication diagram has the important property that it is a map of the location of atoms in the crystal structure plus additional locations (ambiguities) which would give rise to the same Harker synthesis, plus satellitic locations. Fortunately the satellitic locations can be identified as such. Satellitic peaks are caused by powers of rotation operations and by reflections. Satellitic peaks occur in specialized locations which are discussed for the several cases. On the other hand, it is not generally possible to decide between ambiguities. An implication diagram which is a map of the crystal structure, with ambiguities, but without satellitic locations, exists for the following 18 of the 21 possible parallel‐axial symmetries: P2, P21, C2, H3, H31, H32, R3, P4̄, I4̄, P41, P42, P43, I41, H61, H65, H62, H64, and H63. This is also true of certain axial symmetries combined with reflections, such as 2c and 3c. Only for the three axial symmetries P4, I4, and H6 do no implications exist without satellitic peaks. The implication diagrams of the Harker syntheses P(xy⅙) for the space groups H61, H65, H612, and H652 and for Harker synthesis P(xy0) for R3̄ lack both ambiguities and satellitic peaks. For these space groups the implication diagrams are true maps of the crystal structure. Harker syntheses greatly exaggerate the electron densities of atoms in the structure, and for this reason their greatest usefulness lies in the location of the more compact atoms. Diffuse atoms, and therefore especially anions, provide Harker peaks of such low magnitude as to be often lost in the background. Background due to non‐Harker interactions is discussed. An important characteristic of Harker syntheses, which is independent of any interpretation of their meaning with regard to the location of atoms in the crystal structure, is that they provide criteria for distinguishing space groups which cannot be distinguished in ordinary qualitative x‐ray crystallography. This is because the Harker synthesis provides definite criteria for identifying reflection planes and rotation axes. It fails only to distinguish pairs of space groups which differ by a group of inversions alone. Harker synthesis thus realizes the ultimate possibilities of x‐ray crystallography because it is quantitative.
Dissociation Energies of Surface Films of Various Oxides as Determined by Emission Measurements of Oxide Coated Cathodes17(1946); http://dx.doi.org/10.1063/1.1707758View Description Hide Description
A mechanism of falling emission in vacuum tubes is discussed. It is demonstrated that when an electron achieves a critical kinetic energy in moving from the cathode to the anode, and if the anode is an oxide, the electrons will cause a dissociation of the oxide. The liberated oxygen will return to the cathode and falling emission will result. The critical energy of the electron starting the dissociation is found to be equivalent to the heats of formation of the oxides bombarded in the case of five different oxides. Using this equivalence principle, the heats of formation of two compounds are found which have not been recorded in the literature, Ta2O4 and ZrO. In addition, the contact potential between barium oxide cathodes and seven various oxides are determined and the work function of the seven oxides computed.
An Experimental Investigation of Forced Vibrations in a Mechanical System Having a Non‐Linear Restoring Force17(1946); http://dx.doi.org/10.1063/1.1707759View Description Hide Description
This paper introduces a mechanical apparatus capable of generating and recording forced vibrations in a system having a non‐linear restoring force. The experimental wave forms are then compared with the theoretical results given by three graphical methods. These are the methods of Martienssen, Hartog, and Rauscher. Provided a suitable first guess for the amplitude can be made, one application of Rauscher's method gives better results than the other methods. A graph is made of the experimental amplitude as a function of the disturbing frequency. These results are compared with the theoretical results of the above three methods. Again one application of Rauscher's method gives satisfactory results. In all experimental results it is noted that even though the restoring force is distinctly non‐linear, the wave forms of the resulting motion are nearly sinusoidal as long as the frequency of the observed motion is the same as the frequency of the disturbing force. However, steady oscillations can be maintained for which the observed frequency is a sub‐multiple of the disturbing frequency. Two such subharmonics were recorded and the experimental wave forms are shown.
17(1946); http://dx.doi.org/10.1063/1.1707760View Description Hide Description
In 1940, dielectricmeasurements in the centimeter range were considered as difficult and not very accurate. The authors, therefore, developed a ``hollow‐pipe'' method which overcame these objections and required only a weak oscillator and small amounts of the dielectric material. The theory and its practical applications, as perfected by March 1941, are presented in this paper.
17(1946); http://dx.doi.org/10.1063/1.1707761View Description Hide Description
Given the values of y −1, y 0, y 1, and y 2 corresponding to x −1, x 0, x 1, and x 2, we first plot Δy/Δx, the ratio of first difference to abscissa interval, as a function of mean abscissa x̄, and then draw a smooth curve through the graph points which gives the values of a function S(x̄). To determine the value of yp for any value of xp between x 0 and x 1, we substitute values of S(x̄′) and S(x̄″), obtained from the curve, in the following equations: ; where With a carefully drawn smooth curve, the error in the result is only about one‐sixth of the fourth difference term in the Newton‐Bessel interpolation formula. In a simpler variation of the method, the function L(x̄), whose values are given by straight lines connecting the graph points, is substituted for S(x̄) in the equations; however, the accuracy of the result is only about one‐sixth. It is shown that the accuracy obtainable in any case can be estimated from the sagittas σ1 and σ0 for x̄ equal to x 1 and x 0, respectively, where σ1 = L(x̄ 1) − S(x̄ 1) and σ0 = L(x̄ 0) − S(x̄ 0), since the error is proportional to (y″ − y′) times (σ1 − σ0) / (σ1 + σ0). This is true for unequal datum point intervals as well as for equal intervals. When the sagittas are equal or approximately so, the results may be as accurate if the rectilinear variation is used. Taking advantage of this fact, a diagram is given which enables the logarithm of any number to be determined to eight places by substituting values of L(x̄), obtained by linear interpolation between given figures, in certain equations. The method has been found useful when a number of interpolations are required in a single set of data. An important advantage is that it can be used as easily when the abscissa intervals between datum points are unequal as when they are equal.
17(1946); http://dx.doi.org/10.1063/1.1707762View Description Hide Description
The fact that light absorption and emission depend on the environment of the ions and molecules which absorb or emit light can be used to draw conclusions about the structure of liquids,glasses, and crystals. This method supplements, to a certain extent, information gained from other sources, such as x‐ray diffraction data. The method is illustrated by several examples, like the influence of thermal history on the structure of glass and the constitution of the yellow fluorescing zinc orthosilicate.