Volume 17, Issue 5, 01 May 1946
Index of content:
17(1946); http://dx.doi.org/10.1063/1.1707719View Description Hide Description
The stress relaxation of corky materials is of theoretical significance for the understanding of such two‐phase (corky material+air) systems, and is of great practical interest, since these materials are commonly used as gaskets, seals, etc. Stress relaxation was studied at various degrees of compression and at temperatures ranging from 30°C to 200°C. The experimental data were obtained from an automatic stress‐relaxation machine which employed the principle of a chainomatic balance. It was found that the stress‐time curves for various compressions could be obtained from each other by multiplication. This experimental fact is generalized by the statement that the stress is a product of two functions. The first factor is essentially identical with the S‐shaped load‐compression curve of cork and is independent of time. The second factor is shown to be a linear function of log t and is independent of compression. The linear stress‐log time curve is found to persist over a wide range of times and to continue until the stress has decayed to zero at the higher temperatures. The decay time as obtained either by actual experiment or by extrapolation is shown as a function of temperature. This gives a fairly complete picture of the stress‐time‐temperature behavior of cork under compression.
17(1946); http://dx.doi.org/10.1063/1.1707720View Description Hide Description
When a dielectric is placed in an atmosphere of 100 percent relative humidity, an ionized film of water forms, whose conductivity at the end of one minute is within a factor of ten of its equilibrium value, which is usually attained within an hour. This equilibrium conductivity ranges from essentially zero for certain hydrocarbon waxes, silicone resins and silicone‐treated glass to 100 micromhos for ordinary glass and quartz. The ionized waterfilm also produces interfacial polarization at its interface with the dielectric, which produces a marked increase in both capacitance and dissipation factor at audiofrequencies. This polarization builds up in the same manner as the conductivity. Its relaxation frequency appears to be in the audio‐range.
17(1946); http://dx.doi.org/10.1063/1.1707721View Description Hide Description
A statistical analysis of the thermal polymerization of drying oils has been made, and equations have been derived which predict the amount of a specific polymer present as a function of the extent of reaction and the original composition of the oil. Several samples of oils have been heat bodied to test the validity of this analysis. Certain discrepancies are apparent. The presence of an intrapolymer is postulated to explain these discrepancies. Indirect evidence for the presence of such a compound is presented. Additional studies of the heat polymerization of drying oils have been made by conducting the polymerization in dilute solution. After the reaction has been carried out, the solvent is removed and the properties of the oils compared with oils similarly treated without solvent. It has been found that iodine values of the oils are substantially the same regardless of the concentration of oil, and depend only on the time and temperature of heating, but that the viscosity decreases with lower concentration of the oils. This behavior substantiates the formation of an intrapolymer. By using the values at various concentrations it may be possible to estimate the extent of intrapolymer formation in heat bodied oils.
17(1946); http://dx.doi.org/10.1063/1.1707722View Description Hide Description
This paper describes a new apparatus for measuring electrostaticcontact potentials on various materials at elevated temperatures. In this apparatus, the electrostatic charge, acquired by rolling a steel ball down the surface of a rubber test specimen on a heated inclined plane, is measured when the ball drops into the cup of a suitable measuring device such as our electrostaticmodulator. With this instrument the contact potentials of both rubber and GR‐S was found to become highly negative at elevated temperatures. This apparent ``boiling off'' of electrons and resultant disruption of electrostatic attractive forces within the material is much greater for GR‐S than for rubber and probably accounts for the much greater decrease in tensile of GR‐S over rubber at elevated temperatures, and is further confirmation of the electrostaticcontact potentialtheory of reinforcement. By the further application of this theory suitably dispersed compounding materials, which are in effect highly positive at elevated temperatures such as certain proteins, finely divided silica and sodium silicate, have been found to increase substantially the hot tensile strengths of GR‐S compounds.
17(1946); http://dx.doi.org/10.1063/1.1707723View Description Hide Description
The physical significance of stress‐strain curves and of isometrics obtained by the relaxation method is discussed and clarified. Stress‐strain curves taken at various temperatures give the correct dependence of stress upon temperature if they are taken sufficiently fast so that stress relaxation does not mask the temperature dependence. Isometrics obtained after previous relaxation of the sample are shown to depend upon duration and temperature of the relaxation by a numerical factor only. The basis for this behavior is the factorization of the stress into a factor depending upon extension and temperature only which corresponds to the equation of state and another factor depending upon the temperature T* and the duration of the relaxation process. For simple stress relaxation, the same factorization holds with T* equal to T. A general theory is formulated for time dependent elastic phenomena by generalizing Boltzmann'stheory. The theory explains why factorization does not hold for creep, in agreement with experiment.
17(1946); http://dx.doi.org/10.1063/1.1707724View Description Hide Description
A molecular theory is developed to describe quantitatively the permanent set taking place in thin samples of vulcanized natural and synthetic rubbers held at constant extension at elevated temperatures. Permanent set is considered to be the result of the formation, through the action of molecular scission and cross‐linking reactions, of a dual molecular network in the rubber sample, in which the network chains are of two types: chains which are at equilibrium when the sample is at its unstretched length, and chains which are at equilibrium when the sample is at its stretched length. According to the theory the amount of permanent set in a rubber sample is a function of only two quantities: the relative ratio of the number of chains of the two types, and the elongation at which the sample was held. Experimental data on permanent set for various rubber types and under different conditions are presented and are shown to be in good agreement with the theory.
17(1946); http://dx.doi.org/10.1063/1.1707725View Description Hide Description
The crystallization and melting of unvulcanized natural rubber in the unstretched state have been investigated at different temperatures. Change of volume has been used as a quantitative measure of the extent of crystallization, and mercury‐filled dilatometers containing the rubber have been used for the volume measurements. Crystallization was observed to occur at temperatures between −50° and +15°C and to be most rapid at about −25°C. The final decrease of volume on crystallization was usually found to lie between 2.0 and 2.7 percent. The melting of the crystalline rubber was found to occur over a range of temperature and to be strongly dependent on the temperature at which the crystals were formed. The temperature at which the beginning of melting occurs is from 4° to 7° above the temperature of crystallization. The range of melting is about 35° at the lowest temperatures and decreases to about 10° at the highest. The same range of temperature of melting is obtained regardless of the extent of the crystallization.
17(1946); http://dx.doi.org/10.1063/1.1707726View Description Hide Description
The deficiency of GR‐S in ``tackiness'' resulted in a great deal of work in the rubber industry to improve this complex property, but practically all the work had to be evaluated by qualitative ``hand tests'' because quantitative tests were not available. An analysis of the factors involved in the ``tackiness'' of various materials showed that different factors were critical in the so‐called ``tackiness'' of rubber, paints, and varnishes, printing inks, adhesive tapes, etc. A quantitative test was developed to measure the effects of rate of removal on the adhesion or ``tackiness'' of pressure‐sensitive adhesive tapes over a 106‐fold range of rates. This showed why the results of standard ``adhesion tests'' of tapes do not predict the relative behavior of the tapes under small forces acting for long times. Another test was developed for the measurement of the ``tackiness'' of GR‐S rubber which correlates well with the judgment of tackiness by hand tests and it even works reasonably well with rough samples taken from a laboratory mill. With this test, the effects of mastication conditions, aging, and of the addition of softeners and tackifiers on ``tackiness'' of GR‐S (with and without carbon black) were studied. The German tackifier ``Koresin,'' which is now manufactured in this country, was found to be far more effective than rosin and similar materials formerly used, and gave a tackiness approaching that of natural rubber.
17(1946); http://dx.doi.org/10.1063/1.1707728View Description Hide Description
The physical properties of a synthetic polymer are associated with variations in molecular structure which are dependent on the conditions of the polymerizing reaction and the monomers used. In some cases, structural differences of significance for physical properties cannot be detected by infra‐red methods because the relative number of chemical linkages affected is too small. An important structural detail which can be followed by infra‐red analysis is the relative amount of 1,2 and 1,4 polymerization occurring in polymerization reactions of butadiene. Absorption curves are reproduced to show the wide range in the relative amounts of these two structures. A description is given of an attempt to obtain a quantitative measure of this ratio by means of a calibration curve derived from known mixtures of pure octene 1 and octene 2. For polyisoprene, variations in the proportions of 1,4 and 1,2 or 3,4 structure also occur depending upon the type of polymerization. Comparison with Hevea rubber and Balata fails to disclose definite evidence of trans‐isomerism in synthetic polyisoprene. Structural differences due to oxidation of polymers may be readily apparent in infra‐red spectra, hydroxyl and carbonyl groups being especially prominent. The effectiveness of anti‐oxidant in preventing structural changes caused by oxidation is shown in a series of absorption curves for samples of GR‐S with and without anti‐oxidant. When the samples were heat‐treated in air, pronounced structural changes occurred for the sample without anti‐oxidant, but no perceptible changes were evident for the sample with anti‐oxidant. On the other hand, the anti‐oxidant used (phenyl‐β‐naphthylamine) was ineffective for stabilizing the structure toward ultraviolet light.
17(1946); http://dx.doi.org/10.1063/1.1707729View Description Hide Description
The method of determining particle sizes and molecular weights by light scatteringmeasurements on solutions is finding increasing fields of application. In order to obtain the necessary data quickly and conveniently, an instrument has been developed which allows the required light scatteringmeasurements to be taken on a routine basis. This apparatus is described. It measures transmission and the scattering of light at 90°; two different schemes can be employed to obtain the angular intensity distribution of the scattered light. A description is also given of a compensating refractometer which is used to determine the difference in refractive index of solvent and solution. The instrument, which is simple in design, gives a direct reading of the refractive index difference and has an accuracy of better than 10−5. The light scattering instrument has been tested using polystyrene of known molecular weight, and has been applied to study the change of apparent molecular weight (or particle size) of a series of polyvinyl ethers as a function of concentration in some cases, also a number of solvents. For a series of polyvinyl n‐butyl ether samples which was studied, the relation between the intrinsic viscosity [ηi ] of the benzene solutions and the weight average molecular weight M determined by light scattering is ηi = .365×10−7 M 1.45.
17(1946); http://dx.doi.org/10.1063/1.1707730View Description Hide Description
It has been suggested that the so‐called second‐order transition in high polymers is not an equilibrium phenomenon, i.e., a true thermodynamic singularity, but rather a rate effect. Confirmation of this point of view has been obtained by determining the equilibrium volume‐temperature curve of polystyrene. Such equilibrium curves exhibit no transition within the range of from 20° to 140°C, whereas the same material shows a transition at about 82°C for rates of heating usual in thermal expansion measurements. Two facts appear in agreement with the viscous flow interpretation of the thermal expansion process: First, two mechanis ms were found to operate, at markedly different rates. One gave almost instantaneous expansion, even at room temperature, whereas the other was quite temperature‐dependent in this region, being extremely slow at room temperature. Second, the activation energy for the slower mechanism was found to be of the order of magnitude of that for viscous flow, considerably lower than the activation energy for rubber‐like elasticity. Volume‐temperature curves at finite rates of heating are discussed in light of these findings.