Index of content:
Volume 25, Issue 7, 01 July 1954
- SPECIAL ISSUE ON HIGH POLYMER PHYSICS
25(1954); http://dx.doi.org/10.1063/1.1721751View Description Hide Description
The molecular characteristics which determine whether a long‐chain polymer will crystallize at all—and if so, how readily—are considered. The criterion for crystallization is geometrical regularity of molecular structure; but absolute regularity is not essential, for certain copolymers, as well as certain homopolymers with geometrically irregular structures, do crystallize; the requirement is that the different units which indiscriminately occupy corresponding positions in the structure must be not too different in size and shape (as in monomeric ``mixed crystals''); an approximation to regularity is sufficient.
It is unlikely that quite regular molecules can be incapable of crystallizing; when such substances are reported as noncrystalline, it is more likely that they crystallize only with difficulty, and the right conditions have not been found. Reluctance to crystallize is associated with a small interval between relaxation transition (``second‐order transition'') and melting points; this interval appears to be influenced by the general shape of the molecule (though shape is not the only factor). The temperature range within which crystallization proceeds at an appreciable rate may be still further restricted if the chain repeat distance of the molecules is long; and this same factor may make crystallization at any temperature very slow.
25(1954); http://dx.doi.org/10.1063/1.1721752View Description Hide Description
This paper concerns an empirical linear relationship between the melting temperature T 1 and the second‐order or glassytransition temperature T 2 for partially crystalline polymers. Two lines can be drawn, one for symmetrical polymers, the other for unsymmetrical polymers, the latter being of lower slope. These families are only approximate. It is shown for a limited group of polymers, that T 2 is proportional to the cohesive energy density of the polymer. Since T 1=ΔH/ΔS, it is suggested that the two families of curves arise because of differences in ΔS which depends on symmetry and chain stiffness. It is shown that T 1 and T 2 are linearly related with diluent as a common parameter.
25(1954); http://dx.doi.org/10.1063/1.1721753View Description Hide Description
The rates of crystallization of five polymers, poly‐(ethylene oxide), poly‐(decamethylene adipate), and three poly‐(N,N′‐sebacoyl piperazine)'s differing in molecular weight, have been observed dilatometrically at various temperatures ranging from 5.5 to 24° below the respective melting points Tm . The crystallization isotherms are remarkably reproducible, and the previous thermal history above the melting point is without effect provided that melting is completed before the crystallization experiment is begun. The crystallization isotherms display induction periods. The very large temperature coefficients observed are uniform throughout the crystallization process, i.e., isotherms at different temperatures are superimposable by rescaling the time. It is inferred from this and other evidence that steady‐state nucleation and growth occur simultaneously throughout the process. The results are satisfactorily interpreted by applying conventional nucleation theory and assuming that nuclei grow at rates proportional to their surface areas. The treatment resembles that applied to crystallization in metals. Spherical and disk‐shaped nuclei are considered, but the data do not discriminate between them. An extension of the theory accounts for the maximum commonly observed in the crystallization rate as the temperature is decreased below the range of the present measurements.
25(1954); http://dx.doi.org/10.1063/1.1721754View Description Hide Description
The current x‐ray and infrared techniques for estimation of the crystalline‐amorphous ratio in polymers are surveyed. Representative data are assembled for some of the common crystallizingpolymers. Advantages and limitations of the methods are pointed out.
25(1954); http://dx.doi.org/10.1063/1.1721755View Description Hide Description
The melting point, temperature for the onset of crystallization, and the isothermal rate of crystallization of polyethylene all decrease on increasing the extent of branching of the polymer chains. In the molecular weight range studied the melting point is independent of the molecular weight. The temperature for the onset of crystallization at any given rate of cooling appears to increase as the molecular weight is decreased. The isothermal rate of crystallization becomes much greater as the crystallization temperature is lowered, however, at a given temperature the rate is affected only slightly by a change in molecular weight and is primarily a function of the degree of branching.
25(1954); http://dx.doi.org/10.1063/1.1721756View Description Hide Description
Butadiene‐styrene copolymers contain the following units: trans‐1,4‐polybutadiene, cis‐1,4‐polybutadiene, 1,2‐polybutadiene, and bound styrene. Of these the first is the only type which has been shown to form crystals. By the use of published data on volume change arising from crystallization (largely due to Meyer) and on structure as a function of polymerization temperature (largely due to Hampton) it is found that crystallization in emulsionpolymers has been observed only when the trans‐1,4‐polybutadiene content has been greater than about 58 percent, becoming more complete the more this value is exceeded. From this observation one can then draw the following conclusions, all of which are in accordance with direct experimentation:
(1) Crystallization is not observed if the polymerization temperature is above 60°C.
(2) For polymerizations at 50°C a small amount (2 to 6 percent) of bound styrene inhibits crystallization completely.
(3) For polymerizations at 5°C the limit is at about 15 to 18 percent bound styrene content.
(4) At −40°C (the lowest polymerization temperature normally utilized) this limit is at about 30 percent bound styrene.
- REGULAR CONTRIBUTED ORIGINAL RESEARCH
25(1954); http://dx.doi.org/10.1063/1.1721757View Description Hide Description
Spherically symmetric optical instruments, and their desirability for microwaves, are discussed. Application of Hamiltonian optics in analyzing the rays in the Luneberg lens has led to formula of a new lens that has spherical symmetry and a focus near its center. A cylindrical version of this new lens has been designed for operation in the TE 10 mode.
25(1954); http://dx.doi.org/10.1063/1.1721758View Description Hide Description
The well‐known relation between the total energy cross section of a scatterer and its forward scattered amplitude is extended to obtain an approximate transmission coefficient for a uniform planar distribution of parallel cylinders. An analogous reflection theorem for an arbitrary cylindrical boss on a perfectly reflecting plane is then derived; here the total cross ssction of the boss is related to the scattering amplitude in the specular direction of reflection of the plane. This is extended to obtain an approximate reflection coefficient for a uniform distribution of cylindrical bosses on a plane.
25(1954); http://dx.doi.org/10.1063/1.1721759View Description Hide Description
The diffusion equation is solved for transistors for (a) linear flow of minority carriers and (b) non‐uniform flow of minority carriers due to a metal contact applied to the base region. The current transport ratio β is calculated for both of these cases as a function of frequency.
Frequency response characteristics of a transistor, as predicted by the one‐dimensional diffusion equation, give excellent results for junction transistors having base length‐to‐base width ratios greater than ten and having negligible surface recombination rates. For transistors where these conditions are not applicable, the two‐dimensional diffusion equation must be used in the calculation of β.
25(1954); http://dx.doi.org/10.1063/1.1721760View Description Hide Description
A method is described for simultaneously measuring the diffusion coefficient, solubility, and permeability for a given gas in a single sample of solid material. The method utilizes solutions of the diffusion equation (for both plane and cylindrical geometry) which make it easily possible to calculate the diffusion parameters from data taken either in the period of steady‐state gas flow or in the early part of the transient period. Using ultra‐high vacuum techniques, it is possible to make the measurements with extremely small amounts of gas and without errors due to leaks or contamination in the gas handling system.
Values for the diffusion parameters have been determined for helium in Pyrex glass (Corning 7740), and comparisons with previous measurements are made. Directly measured values of the diffusion coefficient are given for the first time. The diffusion coefficient and permeability both exhibit a temperature dependence of the form exp (−E/kT) both having the same activation energy (6.4×103 cal/mole) within the experimental error. The solubility exhibits little or no variation with temperature over the range 25°C to 300°C, and is in good agreement with the only previous measurement, carried out at 515°C.
25(1954); http://dx.doi.org/10.1063/1.1721761View Description Hide Description
The asymptotic expansion of wave fields in powers of 1/k for large k, where k is the wave number, has as its lowest‐order term what is commonly known as the geometrical optics field. The caustics of geometrical optics are those point sets on which the zero‐order terms become infinite. It is well known that caustics may exist even where the exact wave field is perfectly regular. An investigation of reflection from cylindrical walls of arbitrary cross section shows that the occurrence of caustic points means a change in character of the asymptotic expansion of the true field such that the lowest‐order term is no longer independent of k but actually contains a factor k raised to a positive power. There also occurs a jump in phase along a ray passing through a caustic which, as is well known, equals π/2. As an application of the general method, the geometrical optics field is worked out in detail for the case of a plane wave incident on a parabolic cylinder, and the field is obtained in its lowest order at the focus and in the neighborhood of the focus. A similar problem for a reflector consisting of a segment of a circular cylinder is also considered in detail.
25(1954); http://dx.doi.org/10.1063/1.1721762View Description Hide Description
A mass spectrometer ion collector system has been devised for the investigation of secondary electron emission by impact of ions on surfaces. Results are reported for various ions incident on clean, baked (but not atomically clean) surfaces of AgMg, CuBe, and Nichrome V targets. For rare gas isotopic ions it is found that for a given energy ion the value of γ varies approximately inversely with the square root of the mass.
25(1954); http://dx.doi.org/10.1063/1.1721763View Description Hide Description
The methods of information theory are used in the discussion of observations under a microscope. Conclusions are similar to those obtained in a previous paper and exemplify again the difficulty of measuring extremely small distances. A comparison with the discussion of uncertainty relations shows that quantum uncertainties can be very much higher than the theoretical limit, especially when material screens or shutters are not being used and the observation is made on a free particle. This correction to uncertainty relations, however, does not introduce any change in the theory of quantized fields of Bohr and Rosenfeld.
25(1954); http://dx.doi.org/10.1063/1.1721764View Description Hide Description
Several inorganic substances have been examined by electron microscope and selected area electron diffraction techniques. Large crystals of sodium chloride and potassium chloride break up into many smaller crystals of the same substances. Ammonium chloride evaporates, and nitron nitrate appears merely to melt. Reduction to metal was found for silver chloride, silver nitrate, lead carbonate, and cupric oxide. Only morphological changes were observed on bombardment of silver sulfate, lead oxide, lead chloride, and gold chloride. Hydrous ferric oxide and aluminum oxide were converted to crystalline forms which could not be identified by electron diffraction.
25(1954); http://dx.doi.org/10.1063/1.1721765View Description Hide Description
The available experimental data on the short arcs which occur on closure of electrical contacts with potential differences less than the minimum sparking potential are shown to be consistent with the hypothesis that the arcs are initiated and sustained by field emission electrons. Positive ions then carry less than 7 percent of the current at the cathode. The primary function of the ions is to supply sufficient positive space charge to maintain the high cathode field, but the action of the discrete ions near the cathode in creating a temporary pass in the potential barrier may result in the attainment of sufficient electron emission to sustain the arc at values of the cathode field somewhat lower than that required by the Fowler‐Nordheim equation for the same current density.
It is shown that the supply of volatile impurities on the electrode surfaces is critical in determining the breakdown field and the minimum arc current.
Shadowing Technique for Electron Microscopy—A Possible Substitute for the High‐Vacuum Evaporation Technique25(1954); http://dx.doi.org/10.1063/1.1721766View Description Hide Description
A method of shadowing samples directly within the electron microscope has been developed in this Laboratory. This method, which uses thallium chloride immersed in distilled water as the shadowing material, is much less time consuming than is the usual method that involves high‐vacuum metal evaporation; the results obtained are not of the same high quality as those of the usual method, but the time saving often justifies the method. Polystyrene Latex (batch No. 580) is introduced on the specimen screen, to provide an internal calibration standard; this standard makes it possible to measure the angle of shadowing.
25(1954); http://dx.doi.org/10.1063/1.1721767View Description Hide Description
A method for the measurement of the complex conductivity of a rocket exhaust flame is described. Typical results are shown for an acid‐aniline rocket motor and a frequency of 200 mc/sec. The importance of sodium impurities is pointed out and illustrated experimentally.
25(1954); http://dx.doi.org/10.1063/1.1721768View Description Hide Description
An investigation has been made of the charging of fused quartz and borosilicate glass spheres rolling on a clean nickelsurface. The results indicate that the process of acquiring charge depends only upon the nature and surface condition of the materials in contact and is independent of the pressure of an atmosphere of dry N2. The total charge, however, is limited by gaseous discharge to the metal and therefore does depend upon the pressure. Such gaseous discharge has been verified by the observation of individual discharge pulses. The rate of acquiring charge and maximum charge in vacuum has been found to depend upon rolling speed, both reaching a maximum at an intermediate speed, and to decrease with increasing surface conductivity. Since both fused quartz and borosilicate glass charge negatively in contact with clean nickel, it is inferred that both have higher effective work functions than nickel. The surface conductivity of borosilicate glass has been found to be three of four orders of magnitude higher than that of quartz. Therefore, since the charging of the two materials is of the same order of magnitude, the effective work function of the glass appears to be considerably higher than that of fused quartz.
25(1954); http://dx.doi.org/10.1063/1.1721769View Description Hide Description
A calculation is made which relates the physical properties of a junction diode to the troublesome spike current arising from hole storage in the germanium. This spike current results when the diode is suddenly switched from the conducting to the nonconducting state. For small germanium pellets, this current decay is exponential with a decay considerably shorter than that indicated by the hole lifetime in the material. For larger pellets the decay is a sum of exponentials giving a considerably longer decay time, still shorter, however, than the hole lifetime.