Relation between emission and grid cathode contact potential of oxide cathode tube is studied. The contact potential, the difference between average work functions of grid and cathode, chiefly depends on the work function of the grid, which is influenced by barium and barium oxide evaporated on the grid and by gases adsorbed on the surface of the grid. When this BaO layer on the grid is bombarded by electrons from the cathode during activation, the variation of work function of the grid becomes larger than that of cathode and contact potential decreases. So it was ascertained that the contact potential depends considerably on the activation schedule applied. In twin triode the dissociative gas which is given off from the electrodes of one unit under electron bombardment poisons the deposited BaO on grid of the other unit severely and the contact potential of that unit increases. The variation of grid work function or contact potential was obserbed by changing grid wire material, also by changing anode material and cathode core metal. The effect of the grid side rod on contact potential is not so evident.
Microcontent of various halide gases in air is detected by means of the following process. A mixture of air and halide gas is introduced into an ionization chamber in which a RaD source for ionization is installed and the decrease of ionization current, which corresponds to the content of the halide, is measured. A method which resembles the above was adopted by Deisler et al. to the analysis of nonelectron-attaching gases by making use of the difference in specific ionization among gases. On the other hand, the combined processes of the electron attaching to the halide molecules and the recombination of the negative ions are found to be the main cause of the decrease of ionization current, and these processes are expressed by an equation. From this equation and experimental data, the electron attachment coefficient 7) for the next three electronegative gases is obtained, the value of which are 30cm-1 TORR-1 for SF6, 0.2 for CCl2F2, 0.018 for air at E/P=0.12 v cm-1 TORR-1. These values are concluded to be appropriate as compared with the values obtained by other authors or reported in Part 1. Various halide gases are compared with one another concerning their detection sensitivity: SF6 and Freon 11 were both found to be of very high sensitivity, better than 10-5 TORR lit sec-1. A brief explanation of a portable leak detector devised on the above principle is given.
Several electrodeless lamps with Cd-114 isotope (isotope reduced from Oak Ridge's Cd114O by the use of hydrogen, known to be of 98.2% isotope purity), and with natural Cd for comparison, are prepared. These lamps are heated at 250 to 360°C by a small heater and excited by a 60 MHz, 100 W high frequency oscillator. The temperature dependences of spectral profile and intensity of visible triplet lines of Cd, λ5086, 4800 and 4678, are measured and shown in a figure. The halfintensity width of each line of Cd-114 lamp is smaller than that of natural Cd lamp, but this difference is not so remarkable as in the case of Hg-198. The broadening of each line is caused mainly by the Doppler effect when the temperature is low (250°C) and, as the temperature rises, self-absorption furthers the broadening. The temperature dependence of spectral profile can be explained by theorizing the self-absorption which will be reported later. Self-reversal temperature and optimum condition of Cd lamp for interferometry use are given.
For dehydrated powdered food, hygroscopicity should be considered important for preventing the powder from caking, the food from becoming insoluble or off-flavour and for deciding proper mode of packing. Measurements are made on sorption isotherm of many kinds of powered food by the spring balance method with the following results. 1. Sorption isotherm of most of the food is of sigmoid type. Water content per molecular layer calculated by B. E. T equation seems to be the limit for safe storage. But for those food for which B. E. T. plots are not linear, water content can be higher. 2. B. E. T. plot of dry skim milk of 35% protein, 53% lactose and others deviates from linearity above 35°C. From Harkins-Jura equation, it is found that, as the temperature rises, the water molecules, even when the water content is low, remain condensed state, probably because of particular hygroscopicity of lactose.
Crystalline orientations of close-packed hexagonal and face-centered cobalt films made by vacuum deposition at above 200° on rock salt cleavage surface are studied by means of electron diffraction. It is found that close-packed hexagonal cobalt takes up the orientation with (00.1) plane parallel to (111) plane of the rock salt and also  direction parallel to  direction of the rock salt. From these orientation it is presumed that the orientated hexagonal cobalt grows by depostition on orientated cubic cobalt which is actually in contact with the cleavage surface of rock salt. Influence of magnetic field on the orientation of cobalt crystal during the deposition on the rock salt surface is studied. Magnetic field of 4000 gauss did not cause the orientation.
In alloyed germanium p+-n junctions, anomalous low voltage; breakdown phenomenon is observed when certain single crystals are used as base material. Both sharp and soft voltage current characteristics are found in the range between anomalous breakdown voltage and normal breakdown voltage which is determined by the impurity content of the base crystal. This considered to be caused by the difference in number of localized micro breakdown spots. Although current multiplication and microplasma are not observed, mechanism of breakdown is still regarded as effected by carrier avalanche because of the positive temperature coefficient of breakdown voltage. When surface of these particular crystals are chemically etched, many circular shallow pits are found besides usual pits corresponding to the edge dislocation. At the same time, experiment of abnormal X-ray absorption (Borrmann effect) indicates the existence of many micro defects which can not be found in normal crystal. Good correspondence is obtained between number of these defects and that of localized breakdown spots in the junction. The origin of these defects is as yet not perfectly clear, however, the defects themselves are considered to be localized low resistive spots caused by segregation of donor impurity in dislocation of some type. It is proved that lowering of breakdown voltage occurs when these structures exist at n-type side close to p-n boundary of junctions.
Origin of internal stress of evaporated gold films deposited in vacuo (10-5 mmHg) on glass at room temperatures and structure of the films are studied by X-ray diffraction method using a Norelco Spectrometer. Diffraction patterns other than those from (111), (200), (220), (311) planes are too weak to be observed. 1) The values of Bragg angles observed are larger than the normal values given in ASTM cards. This suggests that a tensile stress parallel to the substrate surface is present in the film. The relation between the stress calculated from the deviation in diffraction angles and the film thickness shows that the stress increases with the thickness until it becomes about 1000Å, beyond which the stress gradually approaches a constant value. 2) The diffraction intensities from films of various thicknesses are found to be very different from those from bulk gold. The intensity from (111) is unusually strong in comparison with that from any other plane. The intensity from each plane increases with film thickness but the rate of increase becomes smaller above 1000A. 3) The diffraction lines from the film are usually broadend than those from bulk gold. If it is assumed that this is exclusively due to grain size effect, the grain size can becalculated. When the film thickness is less than 1000Å, the thicker the film, the larger the mean grain size. The maximum calculated grain size is about 400Å. When the thickness is more than 1000A, the mean grain size becomes smaller and tends to about 200Å From these results, it is concluded that 1) in most crystallites, (111) planes are parallel to the substrate, 2) grains near the substrate, the maximun grain size being about 400Å, and 3) the grain size of crystallites which are apart more than 1000Å from the substrate is independent of the film thickness.
A mass spectrometer for study on appearance potential is assembled. R. P, D. method is used for ion source of this mass spectrometer so that ionization may essentially be effected by monoenergetic electron beam. An adequate operating condition is found by a study on characteristics of the ion source. The efficiency of this ion source for difference ion current is three times as good as that obtained by other workers. A secondary electron multiplier is made by way of experiment as an ion current detector. This is an Inghram type and consists of dinodes of Ag-Mg (Mg 1.7%), which are activated in oxigen gas of 1×10-3mm Hg for ten minutes at 500°C. The total gain is 5×105 for total voltage of 4.5 kV, and the noise level at no-signal is found to be 2×10-17A. By the use of this multiplier, sensitivity of the spectrometer increased to approximately a hundred times as good as that obtained by a conventional D. C. amplifier. The appearance potential of nitrogen moleculer ion measured with this mass spectrometer is almost in agreement with spectroscopic value. In addition, structures of the ionization curve that cannot be observed by the ordinary method are also observed.
Contraction process of plasma in azimuthal pinch experiment is investigated and over-all picture of the process is examined. On the assumption that a plasma column with a well-defined radius is compressed as a whole by external magnetic field, relevant parameters of the process are considered to be the following: B=the rate of field increase, a0=the tube radius, and p0=the initial gas pressure. Characteristic quantities of the process, such as the time-scale of contraction, the kinetic temperature of plasma at the maximum contraction, etc., are estimated in terms of these parameters, It is also inferred that the column radius at the maximum contraction is independent of B and p0. The snow-plow model is used to find the change of column radii with time. The experiment is performed with B=several×108 gausses/sec, a0=6cm, and p0=(1-250)μHg, using hydrogen, nitrogen, air, and argon gases. Current and voltage oscillograms and streak camera photographs are used to obtain and analyse the results. The rather lower values of B than those used in experiments at other places are to be stressed, and the condition for the induced breakdown is briefly described. Although the range of parameters is not quite wide, it is found that the over-all picture represents the actual process with a reasonable correctness. It it indicated, however, that some other phenomena not taken into account in the picture do exist, such as the reverse magnetic field remaining trapped in the plasma, which could have important effects on the contraction in certain conditions. This point as well as the radial oscillation of the compressed plasma column will be dealt with in other papers.
Measurement on electric impedance of unicellular organisms suspensions in dilute salt solution will provide useful information concerning membrane resistance of the cell and the physicochemical structure of protoplasm. However, few papers are available about this problem. An impendance bridge, covering a fairly wide range of impedance, simple in design and inexpensive in production cost, has been devised. With this bridge, frequency characteristics of electric impedance of yeast cell suspensions in dilute NaCI solution in a frequency range of 100 KC-2 MC are determined and electric conductivity and dielectric constant of the cells in the above frequency range are calculated by analysis based on Maxwell-Wagner theory. Obtained results are approximately 0.1×10-3Ω-1cm-1 for electric conductivity and 1800 for dielectric constant.
The experimentally known d E effect of polycrystalline ferromagnets is classified into three types. The third-type ΔE effect, in which Young's modulus, E, shows a maximum and then a minimum, is observed with cobalt and h.c.p. Ni-Co alloys and may be interpreted as associated with the process of reversible rotation of magnetization vectors. The first type in which E increases monotonously with magnetization and the second type in which E decreases at first and then increases to a value larger than that in unmagnetized state are observed with cubic metals and alloys. The second-type dE effect is especially conspicuous in f. c. c. Ni-Co alloys. It is shown that the initial decrease of E in the second-type JE effect is associated with the fact the non-180° domain wall displacements due to stress occurs more and more easily as the specimen is magnetized, and that the first-tpye ΔE effect is merely an extreme case in which non-180° domain wall displacements occur very easily at weak fields. Internal stress or inclusions distributed randomly in the specimen may certainly be a cause of the second-type ΔE effect, but such a cause can not explain remarakble extreme values of the ΔE effect and their composition dependence observed with Ni-Co alloys. It is, than, shown, based on the calculation of ΔE effect due to the displacements of (100) 90° domain walls in an ideal simple cubic solid solution, that the stabilization of domain walls by the induced uniaxial ferromagnetic anisotropy is the primary cause of the second-type ΔE effect observed with Ni-Co and other cubic sold solution alloys.