Library

Notes: This investigation considers a hypothetical three-phase ordered composite to permit the design of a new material both stiffer and tougher than the matrix phase alone. The required elastic solution is greatly simplified and was based on a single layered inclusion in an otherwise infinite matrix under uniaxial tension. The solution presented is exact and generalized to accommodate any number of concentric layers of different thickness and composition.

Notes: Attempts to use low-cycle fatigue techniques derived from the work of Coffin and Manson to predict the fatigue performance of six diverse structural plastics are discussed. Through an analysis of dynamic property and thermal behavior studies of polymer samples in reversed-tension fatigue, an improved model for the polymer fatigue process has been proposed. Laboratory test data are presented illustrating the relationship between the various models.

Notes: It is postulated that property gradients exist at the interface forming an “interphase” between the insert and the matrix. Such gradients might be caused by partial solubility and diffusivity of the two phases comprising a composite. For simplicity, the “interphase” is replaced by a shell exhibiting averaged properties (Eshell = 1/2 Einsert + 1/2 Ematrix, etc.). An exact linear elastic solution was employed to evaluate stress fields throughout all regions. The possible effect of the “interphase” and its thickness on several physical properties of the composite are evaluated in terms of the particular stress fields.

Notes: An investigation of the dynamic mechanical properties of several molten polymers was performed using the Maxwell Orthogonal Rheometer. Relaxation spectra derived from experimental data for the terminal region of viscoelastic response indicate that as molecular weight distribution broadens, terminal relaxation phenomena associated with molecular disentanglement and translation extend over a corresponding wider frequency range. The same data indicate that a true maximum relaxation time beyond which no elastic response is observed exists for the materials studied. Moreover, the maximum relaxation time corresponds to the reciprocal of the frequency where the dynamic viscosity deviates from its zero-shear value. Thus an estimate of the time necessary for complete elastic recovery in polymer melts is readily obtained experimentally.

Notes: Tensile creep measurements were carried out on a commercial ABS polymer over a temperature range from 40 to 100°C at stress levels from 0.6 to 1.8 × 108 dynes/cm2. Experiments were conducted in a prototype of an apparatus designed to be compatible with digital acquisition systems. Analysis of the data indicated that application of the time-temperature super-position principle was of limited value due to the use of test temperatures near and below the effective glass transition temperature of the acrylonitrile-styrene component of the polymer. A strong stress dependence of the compliance was observed, even at relatively short times after loading. This was analyzed in terms of a model in which the height of the potential energy barrier to motion of the molecular flow unit is lowered by the application of stress. Analysis of the temperature dependence of the compliance at low stress levels indicated that the effective Tg of the acrylonitrile-styrene phase is about 85°C. The temperature dependence of the magnitude of the activation energy is considered as is the stress dependence of the glass transition temperature. Implications of short-time response in creep with regard to response under impact loading are pointed out. Practical application of results to the prediction of dimensional stability of molded parts is discussed as well as the limitations involved in extrapolation of experimental data to long times and high stress levels.

Notes: Fine structures of several plastic/rubber two -phase polymer systems were studied by means of direct observations of ultrathin sections under the electron microscope using osmium tetroxide staining and a hardening procedure developed recently by Kato. Samples used are several types of both ABS polymers and high -impact polystyrenes, and several PVC/rubber blends and the results were discussed in relation to their dynamic viscoelastic properties. It is suggested that these studies may fruitfully be extended to clarify the structure -to property relationships by use of this method.

Notes: The purpose of this paper is to describe a new deformation calorimeter, which is based on differential thermometry and uses a flowing gas stream as a heat transfer medium which allows it to operate under near-isothermal conditions. Also presented are some preliminary test measurements with a crosslinked polyurethane elastomer and crystalline polybutene-1.

Notes: Antiplasticizers are considered to be diluents which when added to polymers result in mechanical property behavior opposite to that of plasticization. The addition of antiplasticizers to certain polymers such as Bisphenol A polycarbonate, polysulfone, and polyvinyl chloride results in the elimination of the secondary loss transitions of these polymers. As a drop in modulus accompanies these transitions, their elimination results in higher tensile strength and tensile modulus. As secondary transitions are commonly associated with ductility and impact strength, their elimination also results in the observed embrittlement characteristics. The addition of anti-plasticizers to polymers also restricts the diffusion of penetrants resulting from the decrease of molecular flexibility in the polymer matrix.

Notes: A useful concept in polymer science is the degree of crystallinity - the fraction of the polymer that exists in a relatively ordered state. Methods of determination of the degree of crystallinity using density, infrared, thermal, N.M.R. and X-ray measurements are examined in light of modern notions of the structure of semi-crystalline polymers.

Notes: The changes occurring when polythene melts are subjected to shear, have been investigated using a rotating bob fitted into a ram extruder. In addition to chain scission, low density polythene exhibits reversible changes in mechanical and viscoelastic properties. The magnitude of the changes is controlled by total shear strain, and is related to molecular weight and long chain branching. The mechanism appears to be one of reversible disentanglement operating mainly on the large highly branched molecules. The work is relevant to various extrusion processes and to other long-branched polymers.

Notes: This investigation deals with the fatigue behavior of a group of thermoplastics fortified with discontinuous glass fibers dispersed by an injection molding process. The thermoplastics included nylon, polystyrene and polyethylene reinforced with short (1/8 in.) and long (1/2 in.) glass fibers.Several aspects of the fatigue behavior are included in the study. First, classical S-N curves were generated under fluctuating tension with R = 0.05 to show the loss of strength due to cyclic load application. Next, the extent of progressive fatigue damage was established by measuring the residual strength after cyclic loading. Finally, hypotheses pertaining to the fatigue mechanisms operative in all four materials were made based on microscopic examinations of sections removed from fatiguedamaged specimens.

Notes: Twelve polymers and copolymers reinforced with random short glass fibers are used for the study of the strength -composition relationship. Six of these reinforced systems are new and have not been reported elsewhere. The effect of fiber volume fraction on tensile and flexural strengths is related to the Kelly and Tyson equation. For each composition the strengthening factor, F, is calculated and discussed with respect to the structure of the polymer. One factor pertaining to the actual performance of fiber glass, fiber efficiency, K, has been extrapolated for the fiber glass used for this study. The effects of fibers on toughness and the Izod notched impact strength are discussed. It is the impact strength ratio and not the toughness which is used to describe the net result of reinforcement.The deviations between the wet and the dry strengths are used to illustrate the effect of the fiber -matrix interfacial bond. A new parameter, Δθ, is introduced to describe the effect of structure on the fiber -matrix interface. The effect of water at the fiber -matrix interface is further demonstrated through the determination of dielectric constant and dissipation factor of the composite before and after water immersion. The rule of mixtures was found to apply to dry electrical properties of composites.

Notes: The effects of shear stress on the crystallization kinetics and morphology of linear polyethylene and polybutene-1 were studied with the aid of a specially designed apparatus. With this equipment, it was possible to heat a thin polymer sample between glass slides to a melt temperature, quench the sample to a crystallization temperature, and then deform the sample in shear by applying a constant load to one of the glass slides. During the deformation, the crystallization process was observed and photographed under a polarizing microscope. Also, the displacement of the glass slide was simultaneously recorded which made possible a determination of the shear strain as a function of time.The results demonstrate that two phenomena may occur in the initially supercooled polymer samples in response to the applied shear stress. In one case, the sample deformed until it fractured, generally exhibiting no evidene of crystallization; in the other, the sample deformed until an inflection point was reached after which the sample became rigid. This latter phenomenon was attributed to crystallization.At low shear stresses, the inflection point was associated with the growth of spherulites which simply became large enough to bridge the glass slides and prevent further deformation of the sample. This generally occurred prior to the completion of the radial growth of the lamellae.At high shear stresses, however, no evidence of crystallization was seen in the microscope until the inflection point was reached. At this point, birefringence was observed in the sample. The resulting structure generally could not be resolved in the microscope, thereby indicating very profuse nucleation.The results obtained clearly demonstrate that the application of a sufficiently high shear stress to an initially supercooled melt has a substantial effect on the rates of crystallization of both polyethylene and polybutene-1. This was shown most dramatically at temperatures close to the melting point, e.g., both polyethylene at 130°C and polybutene-1 at 113°C, which require over 104 sec to crystallize under quiescent conditions, crystallized at approximately 0.05 seconds.The application of a shear stress to a polymer melt is envisaged as resulting in molecular orientation. In accord with the theories of Flory, and Krigbaum and Roe, the associated decrease in entropy of the melt may be considered to increase the supercooling. Under high stresses at which large increases in supercooling result, crystallization occurs more rapidy at the high temperatures and with polymers of lower molecular weight. At low shear stresses, the influence of temperature and molecular weight on the crystallization kinetics is essentially the same as that obseved under quiescent conditions.Observations through the microscope have shown that the application of a shear stress to a polymer melt leads to large increases in the number of crystalline structures formed and to the formation of oriented morphologies. This latter phenomenon arises due to nucleation lines formed by impurities and spherulites in the deforming melt. The impurities and spherulites apparently cause a disturbance which is thought to result in a local increase in stress of the melt and, hence, a local increase in supercooling. Lamellae then nucleate on these lines and grow out radially.

Notes: Detailed crazing behavior in several plastic/rubber twophase polymer systems was studied by means of direct observation of ultrathin sections under the electron microscope by employing osmium tetroxide staining and hardening procedure. Samples used are ABS polymer, high -impact polystyrene and several PVC/rubber blends. All of the systems investigated showed evidence of stress -crazing when under flexural stress. Relationships between the dispersed rubber particles and the crazing behaviors were studied, and the role of rubber particles in the toughening mechanism of plastics was discussed based on these observations.

Notes: The effects of temperature and penetrant activity on the sorption kinetics and equilibria of a series of alkanes in glassy, biaxially oriented polystyrene were studied. Normal isomers of pentane, hexane, and heptane cause crazing of polystyrene film samples at high penetrant activities (〉 0.85). Crazing kinetics are identical to the kinetics of Case II transport. Transport of these normal hydrocarbons in glassy polystyrene in the temperature range 25 to 50°C is markedly non-Fickian; limiting Case II transport is observed at activities in exces of 0.6. Sorption appears to be controlled by highly activated relaxation processes including primary bond breakage at these high penetrant activities. Fickian diffusion behavior is approached, however, as penetrant activity is reduced. Sorption of the branched isomers of these compounds does not result in polymer microfailure. The sorption kinetics of the branched isomers, although time dependent, appear to be controlled primarily by thermally activated diffusion rather than large scale polymer relaxations which control Case II transport.

Notes: Some of the thermodynamic and physical properties of crystals of linear polyethylene, formed from dilute solution are discussed. These properties include density, enthalpy of fusion, infrared absorption spectrum, selective oxidation and dynamic mechanical measurements. Despite the fact that electron microscopic examination reveals an apparent geometric regularity to these crystals, all these physical measurements yield a consistent interpretation in that they require about 15-20% of the chain units to be in a non-random conformation, and to form a disordered amorphous overlayer.Detailed studies of the experimentally observed relations between the crystallite thickness, crystallization temperature, and dissolution temperature (melting temperature in a dilute solution) are also presented. A thermodynamic analysis of the dissolution temperature-size relations, based on independently determined equilibrium solubility temperatures, indicates that the interfacial free-energy characteristic of the mature crystallites is significantly different from that involved in nucleation. The interfacial structures of the nucleus and the crystallites must, therefore, also be different. Size control by a nucleation process can be demonstrated from very general theoretical considerations. However, the detailed molecular structure characteristic of the nucleus cannot be specified solely from the fact that this process is size controlling.

Notes: Work at Stanford Research Institute on the deterioration of polymers is based on the assumption that aging in the light is the most serious problem and that it has at least three components that must be separated if we are to understand the overall process. We have started by measuring the production of free radicals on the polymer when pairs of t-butoxy radicals are generated from thermal decomposition of di-t-butylperoxy oxalate (DBPO) in bulk atactic polypropylene at temperatures approximating service conditions, 35-50°C. Our preliminary experiments have brought out experimental difficulties and have raised more theoretical questions than they have answered. Nevertheless, they show that the efficiency of production of polymer radicals is low and sensitive to the mobilities of the polymer chains. They also suggest that two kinetically different oxidation processes, with different dependencies on rates of initiation and with different susceptibilities to inhibitors, are proceeding simultaneously. The implications of these findings for results of some previous work and for design of improved accelerated tests is considered.

Notes: Starting with specific constitutive equations, methods of evaluating material properties from experimental data are outlined and then illustrated for some polymeric materials; these equations have been derived from thermodynamic principles, and are very similar to the Boltzmann superposition integral form of linear theory. The experimental basis for two equations under uniaxial loading and the influence of environmental factors on the properties are first examined. It is then shown that creep and recovery data can be conveiently used to evaluate properties in one equation, while two-step relaxation data serve the same purpose for the second equation. Methods of reducing data to accomplish this characterization and to determine the accuracy of the theory are illustrated using existing data on nitrocellulose film, fiber-reinforced phenolic resin, and polyisobutylene. Finally, a set of three-dimensional constitutive equations is proposed which is consistent with nonlinear behavior of some metals and plastics, and which enables all properties to be evaluated from uniaxial creep and recovery data.

Notes: Equations have been derived to convert Brabender flow curves to Instron flow curves for practical application purposes. The technique involved is simple and has been described and discussed in detail. Nine pairs of Brabender and Instron flow curves of three different high polymers, each at three operating temperatures, have been found to fit each other quite well. The converted Brabender flow curves overlap a part of the Instron curves and extend the shear range toward the Newtonian flow region. An important consequence of this work is that the Brabender Plastograph can now be regarded as a formal viscometer while still functioning as a miniature of the Banbury mixers used in industry. More work is being done to take advantage of this consequence.

Notes: Constant stress creep and recovery experiments were performed in the nonlinear range under combined tension and torsion on a tubular specimen of polycarbonate. Nonlinear viscoelasticity theory was employed to describe the test results. It was found that the first three orders of stress terms and a power function of time adequately described the experimental results. The modified superposition principle satisfactorily predicted the recovery following creep. Recovery was almost complete in several days following two hours of creep.

Notes: The physical optics of iridescent multilayered plastic films are analyzed theoretically relating the layer thickness, layer arrangement, and refractive index difference between layers to the reflection spectra for light at normal incidence. Very vivid iridescent and metallic-appearing films may be made by proper control of layer distribution.

Notes: Salts of polyacrylic acid can be fabricated in heated molds using mixed powders of polyacrylic acid and metal oxides. Because of the strong intermolecular forces in these polyelectrolyte salts, they have elastic moduli several times greater than the moduli of conventional plastics. They also have very low coefficients of expansion and high compression strengths. Composites containing metallic and inorganic fillers have also been made and their mechanical properties measured.

Notes: The dynamic thermal stability of rigid PVC resin compositions as determined from Brabender torque rheometer tests has a linear relationship of stability time and torque from which a practical measure of processability at a fixed torque can be derived. The stability time at 1500 m-g torque, t1500, has been used to characterize differences between PVC homopolymer and copolymer resins in tin, calcium-zinc, lead, and bariumcadmium stabilizer systems. In every case, the vinyl chloride-propylene copolymers are unique in giving substantial process-ability advantages.The logarithm of stability time at a fixed torque, t1500, is a linear function of intrinsic viscosity, [η], for PVC homopolymers, giving a single line function corresponding to the expression log t1500 = A + B [η]. The constants, A and B, vary with each of the four stabilizer systems, and are related to the effectiveness of the particular formulations used.The ratio of t1500 for copolymers to the basic t1500 value for PVC homopolymers at the same intrinsic viscosity (molecular weight) is a practical measure of relative processability, trel, which is characteristic of PVC resin types and independent of stabilizer systems.

Notes: The melt flow behavior of straight emulsion-polymerized PVC in a capillary extrusion rheometer has been found to depend upon both the molecular weight and the particle size of the sample. Observations of flow-rate, post-extrusion swell, and extrudate appearance, as functions of extrusion temperature and pressure, suggest that both molecular deformation and particle slippage are involved in the flow mechanism. The relative importance of these two modes of flow varies with extrusion conditions and with the PVC molecular weight and particle size. Particle slippage is favored by large particle size, high molecular weight, and low temperature and by a shear stress above a critical yield value. Apparent melt viscosity, swelling, and roughness are minimized under conditions corresponding to the maximum contribution of particle slippage. In the proper range of temperature and shear rate, straight emulsion PVC yields smooth, low-swell extrudates of excellent physical properties.

Notes: Re-examination of our work on internal plasticization of polyvinyl chloride leads us to suggest a number of guidelines in the synthesis of internally plasticized PVC with optimum combination of strength and low temperature properties: (1) use of comonomers of solubility parameter around 8.5 or 10.5, (2) use of a high molecular weight (long chain) comonomers in a low molar ratio, and (3) their non-uniform incorporation in the PVC chain.

Notes: A number of polymers have been “cold” extruded (i.e. at temperatures well below their normal melting temperatures) by the application of relatively high pressures. In all cases extrudates of relatively large cross-sectional area have been produced, and non-circular cross-sections have also been extruded. The extruded products has been evaluated and the results discussed. In general it has been found that whilst extrusion gives an oriented product, the increase in tensile modulus is much less than for a comparably drawn fibre. This phenomenon has been discussed in terms of the changes in crystalline structure produced by extrusion.

Notes: Polymer pellets are frequently melted on a melting grid. The melting capacity of the gird depends on the force with which the bed of pellets is pressed against the grid bars and on the difference between the temperature of the grid bars and the melt temperature. In this investigation, this dependence has been examined, theortically as well as experimentally, for the case where the pellets melt on the top of the grid. Agreement with the theory was found.

Notes: This study was undertaken in order to understand the effect of low molecular weight additives on the visco-elastic properties of polymers and, in particular, Diamond Shamrock Corporation's PVC-40 resin. The viscous response has been reported in Part I of this series (11). This report describes the elastic properties of these mixtures.A new theoretical analysis of elastic response has been formulated. The direct calculation of stored elastic energy from capillary rheometer measurements is now possible.The conclusion drawn from both the viscous and elastic response study is that a change in viscosity is not a necessary nor sufficient condition for improved processability. Increased flow is obtained by lowering the viscous losses, but the melt fracture phenomena may be unchanged. In order to completely characterize the compound, both viscous and elastic response must be measured.

Notes: A previously proposed but further modified theoretical model for melting in plasticating extruders, in the form of a computer program, which predicts the amount of unmelted polymer at any point in the extruder, was used to simulate the effect of geometrical and operating variables on the melting performance of the extruder. The results indicate in increasing screw length required to complete melting with increasing throughput and a decreasing length of melting with increasing frequency of screw rotation. They further indicate the existence of an optimum barrel temperature for a maximum rate of melting, an optimum number of threads for a maximum melting rate, and a significant decrease in the rate of melting with increasing flight clearance. The effect of other geometrical variables and of operating conditions on the rate of melting and power consumption are also discussed.

Notes: The validity of the Busen-Roscoe reciprocity law in flash exposed dichromate sensitized polyvinyl alcohol (PVA) films was investigated. It was found that the flashlamp operating conditions were an important parameter in an exposure system, since a spectral distribution shift of the lamp output could be obtained by changing the pulse duration from 0.3 to 2.0 msec. The apparent reciprocity failure obtained by varying the pulse duration while maintaining the energy stored in the capacitor bank fixed was consistent with the spectral distribution shift in the lamp output. When this shift was accounted for there was only a negligible reciprocity failure found over the range of these flash exposures; however, comparison with a steady mercury are lamp showed that the flash exposure cross-linked the PVA about 200 times more efficiently.

Notes: An investigation of the viscous flow properties of mixtures of low molecular weight materials, such as lubricants and plasticizer, with PVC has been carried out. In all examples studied the PVC exhibited a discontinuity in flow with increasing temperature. This was interpreted as a phase change in the melt. A diffusion fractionation of the low molecular weight species was used to rationalize the shear rate dependent viscous behavior of these melts.

Notes: The effect of the room-temperature first-order transition on the plastic yield behavior of polytetrafluoroethylene (PTFE) has been investigated. Stress-strain curves were measured at different strain rates and temperatures. Tensile creep under constant dead load was also measured as a function of temperature and stress level. The effect of degree of crystallinity was investigated by using both a rapidly quenched and slow-cooled polymer. Observations were extended to large deformations, so that the phenomenon primarily observed was plastic yield rather than linear viscoelastic behavior.The curve of yield stress vs. temperature in the temperature range from -50 to +68°C was found to be almost identical with the curve of elastic modulus vs. temperature; the yield stress shows a marked local decrease at the first-order transition. The yield elongation was almost constant (at about 5%) over this same range, which is in accord with the above result. The more highly crystalline polymer is always more rigid than the less crystalline polymer at small deformations, but above 19°C its stress-strain curve shows a “cross-over” in stress level with the curve of the less crystalline polymer as extension increases. That is, above 19°C the less crystalline polymer shows a more rapid rate of “strain hardening”, even though the strain-hardening effect is pronounced in both polymers. Attempts to apply time-temperature superposition to creep data at different temperatures were partially successful; the lateral shifts required corresponded to an activation energy of approximately 80 kcal.The experimental observations suggest a model of the solid-state structure of PTFE which could be described as an “elastic-plastic network”, in which crystalline domains are connected by elastic amorphous regions, and in which the crystalline domains can flow plastically at sufficiently high stress or temperature.

Notes: Rheo-optical techniques for the study of crystalline polymers, developed primarily in the United States and Japan, include dynamic birefringence, X-ray diffraction, dichroism, light scattering, and fluorescence. These permit the characterization of the rates of orientation of crystals, amorphous regions, and crystalline superstructures. The application of these methods has permitted the formulation of a model for the deformation of crystalline polymers.

Notes: The fatigue failure of thermoplastics and glass-reinforced thermoplastics is a function of cyclic stress or deflection level, test frequency, viscoelastic polymer parameters and matrix-to-fiber stress transfer. This paper reports an investigation of the modes of high frequency fatigue failure in glass-reinforced and unreinforced thermoplastics. In particular the effects and control of dissipative heating of a working specimen and the role of efficient matrix-to-fiber stress transfer are considered. A test procedure with which the severity of dissipative heating can be continuously followed and controlled during an accelerated fatigue test is described.

Notes: Empirical test data on low-density polyethylenes exposed to the elements for very long times are compared with results on samples of the same materials subjected to artificial “weathering” in a modified carbon arc machine.The test results conform to the expression \documentclass{article}\pagestyle{empty}\begin{document}$$ t_a = B + t_n^k $$\end{document} where ta is hours of accelerated weathering, tn years of natural exposure, and the constants B and k have values of 150 and 2.4, respectively, for the materials and test conditions involved.Earlier results are confirmed in that optimum weather resistance is found to depend on good dispersion of a fine-particle black. Protection improves with concentration up to the limit of about 5% by weight imposed by degenerative effects on mechanical properties from higher loadings of the fine channel blacks. Dependence of weatherability on molecular weight (melt index) is noted.

Notes: Existing theory of polymer flow has been applied to a definition of shift factors which reduce the widely different melt viscosity/shear rate diagrams of plasticized PVC compounds to well-defined master curves. The master curves are temperature dependent and also define the flow properties of the unplasticized polymer on which a group of plasticized compounds is based. For given plasticizers, the value of the shift factor was found to depend on melt temperature and plasticizer volume fraction. Explicit relationships have been generated for three plasticizer systems; for these, melt viscosity/shear rate data can be precalculated over several decades of shear, and in the temperature range of 150-200°C. Absolute values of the shift factors depend on the type of plasticizer, and a correlation with polymer/diluent interaction parameters has been attempted. Initial results, valid only at high plasticizer volumes and near the reduced melt temperature of a polymer/plasticizer mixture, support the existence of such a correlation.

Notes: This discussion will illustrate some of the factors which go into the design of experiments required to solve problems in the field of plastics. Two quite diverse problems were considered. In the first case a series of experiments was described which provided a very general technique for predicting polyethylene flow behavior as a function of shear rate over a wide range of temperatures from a minimum of new data. In this case the experimentation concerned only the viscous aspect of flow and was not limited by the availability of special samples. In the second case it was desired to learn what the effect of a specific molecular structural variable was on the general melt rheological properties of polyethylene. This required the polymerization of special samples followed by detailed characterization by both solution property and melt rheological techniques.

Notes: Derivative stress-strain curves at constant strain rate for polyacrylonitrile copolymer fibers have been obtained by simple modification of conventional tensile testing apparatus. These derivative curves, considerably more complex and eventful than stress-strain curves, have been interpreted as direct records of changes undergone by the apparent Young's modulus during experiments which extend from infinitesimal strain to failure. Analysis of the data shows that these changes in the apparent Young's modulus cannot be accounted for to any significant extent by viscoelastic time effects and must therefore be consequences of the deformation process itself. In the light of this analysis, it is concluded that polyacrylonitrile fibers “soften” considerably in the strain range 0.5-3% and that they exhibit moderate “hardening” in the strain range which extends from ca 3% to the breaking strain (10%).

Notes: Some properties of multilayered films are summarized in relation to the properties of the component materials. Polymers may be selected for co-extrusion into layered films to obtain various combinations of properties, such as barrier, stiffness, heat sealability, optics, and toughness. Properly chosen combinations of high and low modulus materials exhibit mutual interlayer reinforcement. Permeabilities of multilayer films can be calculated from layer geometry and individual component properties.

Notes: Spherulitic films of polybutene-1 have been stressed under biaxial tension. Direct microscopic observations indicate that the deformation is localized at the spherulite centers and at the intersections of boundaries in agreement with previous results obtained with linear polyethylene. The results demonstrate an increasing tendency for brittle failure with decreasing molecular weight. Effects of surfactants and solvents are also presented.

Notes: Molten polymer was pumped through an isothermal-wall heat exchanger and the temperature at four radial positions was measured with a hypodermic thermocouple. Temperature profiles were obtained for heating, cooling, and isothermal flow of polyethylene, polystyrene and polypropylene at Graetz numbers from 1.7 to 9.2. The results show large viscous heating effects, particularly for cooling.

Notes: An analog computer-polymerizer system is described which polts continually the rate of polymerization in percent per hour and the total percent conversion. Computer components and their functions are briefly described and the polymerization vessel with its control instruments are shown. Methods of calibration and accuracy of the system are discussed.Polymerization rate curves for various initiators at various temperatures are presented. These curves show the typical autoacceleration of precipitating systems and the rate increase due to the gel effect. Experimental evidence is presented to explain the pressure drop seen in these curves at 70-75% conversion.

Notes: The influence of molecular weight and molecular weight distribution on the melt rheological behavior of two polystrenes of approximately the same weight average molecular weight, but of widely different molecular weight distribution, was determined. Then, using a series of capillaries with different length-to-diameter ratios in an Instron Capillary Rheometer, the entrance correction methods of E. B. Bagley and the relationships of W. Philippoff and F. H. Gaskins, the recoverable shear strain (SR) in the melt at the capillary wall for these mono- and polydisperse polystyrenes was determined. Shear modulus (G) and normal stress (PN) were calculated using the relationships: G = τRC/SR and PN = 2τRC SR, where τRC is the corrected shear stress at the capillary wall. These are compared to values obtained using a Weissenberg Rheogonimeter. These two polystyrenes were also injection molded into an ASTM specimen mold over a wide range of stock temperature, using a 12 OZ. in-line reciprocating screw injection press, and evaluated for mechanical property values. The effects of the elasticity parameters (SR G & PN) and their magnitude on the rheology, processability and mechanical properties of these polystyrenes are discussed.

Notes: The viscosity of low-density crosslinked polyethylene was studied as a function of gel content and shear rate (3.75-33 sec-1). A simple model relating viscosity with gel content is suggested. It is shown that sol viscosity decreases with cross-linking propagation. The rheological parameters of the sol fraction are changed as a result, and it is natural that this effect should be utilized for flow calculations. Experimental data indicate a high degree of interaction between the sol molecules and the gel, and an experimental technique is presented for measuring it.The rheological parameters of crosslinked polyethylene are closely dependent on the gel content, viscosity and pseudoplasticity increasing with the latter.

Notes: As discussed in this paper, melt treatment prior to crystallization, apparently proceeds by the following stages: partial melting; an insensitive region; and at higher temperatures, deactivation of nucleation sites. The deactivation-region onset temperature is quite dependent on the subsequent crystallization conditions and may not be observed in some polymers during crystallization from the melt. In addition, those polymers capable of being quenched directly from the melt to a non-crystalline glassy state and subsequently crystallized by reheating to above the glass transition, do not exhibit any more than a partial melting type melt treatment effect. The deactivation regime absence is a result of the homogeneous nucleation that occurs during crystallization from the quenched glassy state at temperatures slightly above the glass transition.Use of a metal- or glass-constraining medium does mask (at least partially) the effect of melt history upon crystallization from the melt. In addition to the masking effect of a constraining medium, some of the controversy in the literature pertaining to the existence of a melt treatment phenomenon may arise due to degredation of some polymers prior to the onset of the deactivation regime. The crystallization conditions employed are also quite influential on the possible effect melt treatment can have since the melt history phenomenon is noted by its effect on subsequent crystallization.

Notes: The flexural fatigue characteristics of boron fiber reinforced epoxy were shown to be strongly dependent on the aspect ratio of the reinforcing fibers. These composites were evaluated in a stiffness limited, constant maximum stress mode. The failure mechanism was found to be a combination of interfacial failure and brittle rupture of the matrix at 45 deg to the axis of the fibers. These fatigue cracks result in debonding of the outer fibers, thereby reducing the effective moment of inertia of the specimen.

Notes: The stress concentration and effective flexure stiffness of an ASTM V-notch Izod specimen is excessive for most cases of plastics use. A more typical stiffness is that of a thin plate in two dimensional bending. Such can be approached with a larger radius and deeper V-notch. Instead of a V-notch, however, a double edge notch with the same radius and thickness at the root is recommended. The special advantage of the double edge notch is that the notch at the back edge reduces the energy loss due to the clamp.

Notes: Silane coupling agents containing organo-functional and alkoxy groups are applied to glass fiber reinforcements from an acid solution commercially. It is found that finely divided soda glass suspended in a methanol-water solution of various silanes promotes hydrolysis without the presence of an acid catalyst. When acid is present in addition to the glass, as in commercial application baths, rate of hydrolysis is lower than with either acid or glass alone. There is evidence that basic glass components catalyze hydrolysis. Further work under commercial application conditions is indicated.

Notes: The extraordinary physical properties exhibited by block polymers have been ascribed to the presence of a multiphase microstructure in which the higher modulus phase acts as a quasi-crosslink or filler particle. Recently, the optical examination of these materials combined with mechanical testing has given investigatiors new insight into this complex morphology. In this investigation, simultaneous stress, strain, and birefringence data have been collected on a block polyester urethane elastomer. Stress-softening is observed in cyclic stress-strain experiments, giving rise to significant hysteresis in the stressstrain and birefringence-stress curves. In these tests, prestrain causes a large increase in the stress-optical coefficient, but has little effect on the strain-optical coeffcient. As the temperature is increased, the strain-optical coefficent decreases while the stress-optical coefficient increases with the latter exhibiting a larger temperature dependence than predicted by the kinetic theory of rubber elasticity. The good agreement between the mechanical-optical response of polyester urethanes and that of other block polymer systems provides further evidence of their morphological similarity.

Notes: A machine to measure the creep deformation of plastics under uniaxial compressive loads is described. The problems associated with accurate creep testing in compression, primarily the application of a uniform stress to the specimen and the measurement of the resultant strain, receive particular attention. For the specimen geometries used, the effect on the measured strain of frictional restraints at the specimen ends is negligible provided the strain measurement is made with an extensometer attached to the specimen.The effect of fabrication techniques on the deformation behavior of polytetrafluoroethylene (PTFE) has been examined. Sintering time and temperature are found to be the most significant variables in the processing of PTFE. A comparison of uniaxial tensile and compressive creep data has shown that the non-linear viscoelastic behavior of the material extends into the low strain region.

Notes: A useful concept in polymer science is the degree of crystallinity - the fraction of the polymer that exists in a relatively ordered state. Methods of determination of the degree of crystallinity using density, infrared, thermal, N.M.R. and X -ray measurements are examined in light of modern notions of the structure of semi -crystalline polymers.

Notes: A fundamental mathematical model of the plasticating extruder was assembled by combining a previously reported melting model with an improved melt pumping model. The validity of the model was tested with numerous experiments on 2 1/2 in. and 8 in. diameter extruders. The experiments were performed with low and high density polyethylene, plasticized polyvinyl chloride, rigid polyvinyl chloride powder, polypropylene, ABS and nylon 66. The mathematical model proved to be excellent in predicting pressure and plastic temperature profiles in the extruder channel, temperature fluctuation, solids content, and power consumption.

Notes: A new photoredox catalyst is described wherein a phenothiazine dye in its excited state oxidizes a catalyst to a free radical. This free radical efficiently initiates the polymerization of metal acrylates giving rise to polymers which are insoluble in the monomer-containing medium. The polymers precipitate as insoluble colloidal particles. Optical transmission measurements are used to follow the polymerization and to calculate the number and size of the polymer particles. The effect of various factors such as the nature of the catalyst and the photooxidant, the pH and the temperature of the medium on photopolymerization are described. It is indicated that this new system may be useful for rapid photographic imaging and display applications.

Notes: This paper is the third in a series describing work done under Air Force sponsorship to develop high temperature polyimide laminates for radomes and other parts for supersonic aircraft.Twelve resin compositions are described and evaluated. Most were prepared from benzophenonetetracarboxylic dianhydride and one or more aromatic diamines. Modifying linkages such as amide, ether, benzimidazole, and oxadiazole were present in some cases.A variety of precure and pressing conditions were studied, best results being obtained with a fully cured prepreg pressed at about 700°F and 200-1000 psi. Most of the laminates were made with E glass cloth, but some work was done also with S glass, Refrasil, and carbon cloth.Initial flexural strengths on E glass of 35-60,000 psi at room temperature, and 20-40,000 psi at 600°F were observed for the better resins. On S glass, slightly higher strengths were observed together with values of about 10,000 psi at 700-1000°F. Aging data on E glass laminates show retention of at least 10,000 psi flexural strength at temperature for about 250 hrs. at 650°F, 1000 hrs. at 600°F, 5000 hrs. at 527°F, and 30,000 hrs. at 482°F.Dielectric constant and dissipation factor at 10 Gc were found to be nearly constant at 3.5-4.4 and 0.005-0.01, respectively, for temperatures up to 662°F and for aging up to 1000 hrs. at 600°F. At frequencies of 60 c and 1 Kc, an increase of dissipation factor with temperature was found and was used as a measure of Tg.

Notes: Biaxial creep tests have been performed on thin-walled tubes of polymethyl methacrylate at 116°F and 50% R.H. The tubes were loaded radially up to a specified stress state, which was then maintained for a finite time. Additional tests involved a constant stress state for a period Δt1, after which the specimen was rapidly subjected to a second stress state which was maintained for a period Δt2. The experimental apparatus and procedure are described. The results for the range of stress, temperature, and time investigated show: 1) the creep behavior can be described by the form ∊ = ∊0 + mtn, where ∊ is the strain, ∊0 is the time-independent portion of the strain, and m and n are constants; 2) n appears to be independent of stress state in the fourth quadrant (-σz axial stress, +σθ tangential stress), and m is a linear function (log-log) of effective stress. In the first quadrant (+σz +σθ), the values of m and n are dependent on stress state, the constant m being also dependent on effective stress. Since two measurements are made for tangential strain ∊0, a description of the deformation can be obtained. Some results of pre-strain history and fracture are also presented.

Notes: Epoxy-Versamid specimens were loaded in tension up to failure at different constant strain-rates and temperatures. Results revealed three modes of behavior prevailing at different temperature-strain-rate regions and associated with brittle, ductile and rubbery failure modes. The ductile region was found to be confined within a narrow band on the temperature-strain-rate plane, and is characterized by a yield plateau in the stress-strain curve and by linear dependence of yield stress on log strain rate and temperature. Yield strain seems to be almost unaffected by strain-rate, but decreases slightly with temperature rise.Analysis indicated that experimental data within the ductile region are consistent with Eyring's formulation for non-Newtonian viscoplastic flows. It leads to the evaluation of the “apparent activation energy” and activation volume for the two epoxy systems tested.Comparison with previous work indicates that the above parameters as well as yield stress and elastic modulus tend to increase with the decrease of the Versamid content in the resin.

Notes: Neither the fringed-micelle, fringed-fibril or extreme folded-chain models explain the morphological and mechanical properties of highly drawn polyethylene. The modified folded-chain model, assuming that a substantial fraction (between 5 and 30 percent) of the molecules do not fold back at the crystal surface but go through the “amorphous” surface layer and enter the next crystal, avoids the insufficiencies of the above mentioned models.The elastic modulus and tensile strength of drawn polyethylene, both increasing with draw ratio, are to a large extent the consequence of the molecules bridging the quasi-amorphous layers and interconnecting the folded-chain lamellae oriented more-or-less perpendicular to the machine direction. The folds create enough space for the accommodation of more-or-less extended tie molecules in the quasi-amorphous layers between the lamellae. Electron microscopy and calorimetry of samples as drawn, annealed and/or etched with fuming nitric acid support the model.

Notes: The permeation of carbon dioxide through polyethylene membranes has been studied at pressures up to 54.4 atm. and at temperatures above and below the critical temperature of the gas (31.0°C.). The permeability coefficient is independent of pressure at the highest experimental temperature (61.0°C.), but becomes increasingly pressure-dependent as the temperature is lowered. The principle of corresponding states can be used to correlate the solubility of both gases and vapors in polyethylene over a wide range of temperatures. This principle can also be invoked to obtain an upper limit for the penetrant pressure above which the permeability coefficient becomes pressure-dependent. The effect of pressure on the permeability, solubility, and diffusivity of gases and vapors in polyethylene is discussed in some detail.

Notes: Data are presented on the solubility in and permeation through polyolefin films of a number of gases and liquids. The permeation rates are found to be dependent on the type of film, the concentration and the temperature. A classical diffusion model explains the permeation mechanism and other relations serve to correlate the data.

Notes: Entanglement theories for polymer solutions resemble those developed for solid rubbers. These rubberlike theories are extremely successful qualitatively; they give a good indication of the type of response observed experimentally in concentrated solutions. Quantitatively the theories are not so useful; in general they predict constant viscosities in simple shearing motions and ever-increasing tensile stress in steady elongational flow.If it is supposed that the lifetime of the entanglements is limited partly by a maximum allowable strain magnitude, and that the network ruptures locally whenever this magnitude is exceeded, greatly improved quantitative predictions are observed. For polyisobutylene-cetane solutions, where the critical strain magnitude appears to be about 3, excellent prediction of the steady shearing viscosity curve is available starting from the measured dynamical response to small sinusoidal strains and the critical strain magnitude. Normal stress effects are also well represented; in elongational flow the tensile stress shows a slight maximum. It thus appears that the notion of network rupture is useful in guiding the selection of continuum theories for polymer fluid description.

Notes: Calculations of the molar heat and entropy of adsorption for pure methane and pure ethane on silica gel at 25°C. show the heterogeneity of the adsorbent when plotted as functions of the spreading pressure. Plots are given for the activity coefficient of methane and ethane in the adsorbed phase based on experimental data for the entire range of compositions at 25°C. up to a maximum pressure of 1,400 lb./sq.in.abs. Similar results were obtained at 5, 15, and 35°C. The activity coefficients deviate from unity at intermediate compositions.

Notes: Differently shaped bubbles were observed growing during nucleate boiling of water at atomspheric pressure. The surface temperature beneath the bubbles was measured simultaneously with a fast response surface thermocouple. The evaporation from the base necessary to account for the observed cooling was calculated. A comparison of the ratio of vapor volume formed at the base to the total volume of the bubble shows a dependency on bubble shape. The ratio is smallest for spherical bubbles, largest for hemispherical and intermediate for oblate bubbles.

Notes: The nonuniform distribution of vapor across a distillation plate causes a reduction in the Murphree plate efficiency. The magnitude of the reduction has been evaluated and the maximum effect occurs when the liquid mixing is intermediate between the perfectly mixed and the plug flow cases.

Notes: The free volume equation for liquid viscosities is generalized to multicomponent mixtures. No experimental mixture data are needed in order to use the correlation. Calculated viscosities for liquid mixtures composed of molecules which approximately obey the Lennard-Jones 6, 12 potential law show very good agreement with experimental data for eight binary systems.

Notes: A finite-difference method is developed for numerical solution of parabolic partial differential equations. This technique is explicit and stable. It is shown that the present method is more accurate and faster, in terms of computer time, than the Crank-Nicholson method. A method of handling nonlinear problems is also presented. Two examples are given to illustrate the present technique. The first problem is a linear diffusion equation. The second problem deals with two simultaneous nonlinear parabolic partial differential equations with Neumann boundary conditions describing the steady state of a packed-bed catalytic reactor with radial mixing.

Notes: A rigorous and simple thermodynamic equation relating equilibrium vapor-liquid compositions and the phase enthalpy differences for a binary, isobaric system is extended to multicomponent systems. An analysis is made to indicate the potential applications of computing the latent heat of vaporization directly from the isobaric vapor-liquid equilibrum data and testing the consistency of phase composition and enthalpy data.

Notes: The laminar boundary layers on a moving continuous flat surface in non-Newtonian fluids characterized by the power law model are investigated using exact and approximate methods. Both pseudoplastic and dilatant fluids are considered. Numerical solutions of the boundary-layer equations are obtained for values of the parameter n in the power law model ranging from 0.1 to 2.0. An integral solution of the momentum equation, which can be used to obtain values of the dimensionless shearing stress that are in good agreement with the exact values, is developed. An integral solution to the energy equation is also presented.

Notes: Forced convection heat transfer from vertical cylinders normal to an air-water spray flow stream was measured over an air velocity range from 60 to 140 ft./sec. and a water spray density range from 0.03 to 0.50 lb.m/(min.) (sq. in.). Local heat transfer coefficients were determined at 15 deg. intervals around the circumference of both a 1.5 and a 1.0 in. diam. cylinder. It was found that the addition of 0.426 lb.m/(min.) (sq. in.) of water spray to a 133 ft./sec. air stream raised the stagnation point heat transfer coefficient from 45 to 1,650 B.t.u./(hr.) (sq. ft.) (°F.). Similar intensification was found for other angles around the cylinder circumference; however, the magnitude decreased with increasing distance from the stagnation point. Local heat transfer coefficients were normalized with respect to their corresponding stognation point values and plotted parametrically as a function of angle and air velocity. These profiles showed that the normalized heat transfer coefficients decreased with increasing air velocity at angles other than the stagnation point. Average cylinder heat transfer coefficients were calculated from the air-water data and two correlations were obtained relating these coefficients to the air and the water spray Reynolds number.

Notes: A study was made of the radiolytic oxidation of cyclohexane in aqueous solution by using cobalt 60 gamma radiation. In the presence of dissolved oxygen gas the reaction proceeds irreversibly with the formation of a number of oxidation products. Those identified were cyclohexanol, cyclohexanone, and hydrogen peroxide. Trace quantities of carbon dioxide were also detected in the liquid phase. Using a constant dose rate of 15 × 104 rad./hr. in all runs, radiation yields were G(C6H11OH) = 0.49, and G(C6H10O) = 1.08. When molecular oxygen was excluded from the system, neither cyclohexanol nor cyclohexanone was formed on irradiation. A simplified kinetic model based on competition by the various solutes in the system for a single primary radical species produced by radiolysis of the solvent molecules is developed. Results generated using the mathematical model are compared with the experimental data. Excellent agreement between the observed and computed trends is noted at radiation doses up to 50 × 104 rad.

Notes: The turbulent transport of mass, energy, and momentum was studied in a 4 in. diameter cylindrical column packed with 0.3 in. diameter stoneware spheres. Helium was used as a tracer material, with air as the mainstream fluid. The concentration of tracer present, the temperature, and the velocity of the gas were measured at several axial and radial increments. These data were analyzed numerically to obtain the radial component of the effective thermal conductivity and the radial component of the mass-dispersion coefficient as functions of radial position. Experimental conditions covered isothermal determinations at room temperature and nonisothermal determinations in which a temperature gradient was established in the radial direction. Significant differences were found between the isothermal and strongly nonisothermal results, primarily for the velocity profiles and thermal conductivities. Correlations were developed for the local values of the parameters.

Notes: No Abstracts.

Notes: Manning, Wolf, and Keairns' model for continuous flow-stirred tank reactors has been extended to include adiabatic as well as isothermal operation. The adiabatic, steady state yield of a continuous flow-stirred tank reactor was measured experimentally using the second-order, homogeneous, exothermic reaction between sodium thiosulfate and hydrogen peroxide. Model predictions of overall tank yield agreed closely with the data, thus verifying the applicability of this model to explain the effects of operating variables such as impeller size, impeller type, agitator speed, feed location, feed concentration, and flow rates.

Notes: The photodecomposition of acetone vapor was studied in a tubular flow, differential reactor at atmospheric pressure and from 82 to 112°C. The products included significant amounts of biacetyl along with carbon monoxide and ethane. A rate equation was developed from a sequence of elementary reactions believed to explain the photolysis of acetone. The effects of light intensity and acetone concentration, indicated in the derived rate equation, were confirmed by the data.The rate is less than first-order with respect to acetone, and this is believed to be due to reformation of stable acetone molecules by deactivation of CH3COCH3*.

Notes: Metal surfaces of titanium, aluminum, copper, lead, and zinc were irradiated by a 25 joules pulsed laser with an energy flux ranging from 106 to 108 cal./(sq.cm.) (sec.). The resulted cavities were sectioned, polished and measured to compare with the isotherms computed from several heat conduction models. Reasonable agreement was found between the experimental cavity diameters at the interface and those calculated from the disk source model. Also, a graphical correlation between the cavity depth and its interface diameter is presented.

Notes: A theoretical analysis of the operation of a Schildknecht type of column crystallizer is presented and simplified to three special cases. Experimental determinations were made of concentration profiles established during the purification of benzene with cyclohexane as impurity under steady state, batch conditions of operation. Benzene containing 34 to 500 ppm. cyclohexane was obtained by processing feed material with 5,000 to 30,000 ppm. cyclohexane. It was found that experimental data obtained under conditions of normal operation are well correlated by theoretical expressions if it is assumed that the crystals are not pure but contain cyclohexane at the level of 34 to 100 ppm. It was further established that the resistance to mass transfer between the liquid which adheres to the crystals and the wash liquid is negligible in comparison to backmixing effects as represented in the model by an eddy diffusion term. Smooth operation of the column was obtained at crystal rates below a limiting upper value and at oscillation frequencies above a lower limit.

Notes: No Abstracts.

Notes: No Abstracts.

Notes: A generalized approach to reverse osmosis process design is presented for solution-membrane-operating systems characterized by the dimensionless parameters γ, θ, and λ defined in terms of the pure water permeability constant A, solute transport parameter (DAM/Kδ), mass transfer coefficient k on the high pressure side of the membrane, and the properties of the solution. Analytical expressions are derived, in terms of dimensionless quantities, for the change of volume of solution, concentration of the bulk solution and that of the concentrated boundary solution on the high pressure side of the membrane, the change in the permeating velocity of solvent water through the membrane, solute separation, and the other related quantities, at any instance, as a function of concentration of the product solution on the atmospheric pressure side of the membrane, or time from the start of the operation for reverse osmosis systems specified by γ, θ, and λ. The equations are applicable to membranes for which (DAM/Kδ) is independent of solute concentration and feed flow rate, and for aqueous feed solutions whose molar density can be assumed constant and whose osmatic pressure is proportional to mole fraction. The equations are developed first for the case of batch-by-batch operation, and their applicability to the flow case is then indicated.

Notes: The final stages of condensation polymerization are characterized by a rapid rise in molecular weight, as the condensation product is formed and diffuses out of the polymer. The process occurring is one of desorption accompanied by a chemical reaction. The penetration theory equations for a generalized condensation polymerization reaction have been solved and some solutions are presented. The penetration theory solution, obtained by finite-difference computations, is compared with an analytical solution for the special case of no diffusional resistance.

Notes: No Abstracts.

Notes: No Abstracts.

Notes: Positive feedback control systems sometimes can be used to make stable chemical reactors generate periodic outputs. For one set of system parameters the time average conversion obtained from the oscillating reactor was approximately 20% higher than the economic optimum steady state value; for another case the time average operating cost was approximately 5% lower than that corresponding to the optimum steady state design. A general method for predicting when an improvement can be obtained is presented.

Notes: A method utilizing a diffusion controlled electrochemical reaction was used to measure average and instantaneous mass transfer coefficients at the solid-liquid interface in upward gasliquid climbing film flow in a vertical annular duct. These measurements give some indication of the mechanics of flow of the film, the extent of turbulence at the inner wall, and the effect of film thickness and wave motion on the mass transfer process at the inner wall. Predictions of incipient downflow of the film, shear stress at the inner wall, and interfacial shear stress were obtained from these measurements. Fluctuations in the velocity gradient at the inner wall were also studied. Results of this study are in good agreement with previous work and with theoretical predictions based on simplified momentum balance concepts.

Notes: The design of an experimental cocurrent spray dryer and its method of operation are described. The rationale for the development of some new measurement techniques and modifications to existing ones is discussed. These techniques were used to obtain profiles at several dryer levels of the air velocity, air humidity, and spray mass velocity for the case of a 30% sodium nitrate solution sprayed into 100°C. air. Somewhat less detailed measurements of the spray moisture and radial turbulent diffusivity were also made.The humidity and spray moisture measurements gave very nearly identical results for the progress of the evaporation along the length of the dryer. The axial injection of the spray was found to result in steep gradients for the air velocity, air humidity, and salt mass velocity but these were found to decay under the influence of turbulent mixing which is also generated by the spray injection.

Notes: The local volume average of the equation of motion is taken for an incompressible fluid flowing through a porous structure under conditions such that inertial effects may be neglected. The result has two terms beyond a pressure gradient: g, the force per unit volume which a flowing fluid exerts on a porous structure, and the divergence of the local volume-averaged extra stress tensor (viscous portion of the stress tensor).Constitutive equations for g are examined with the aid of the principle of material indifference. When g is assumed to be a function of the velocity of the fluid relative to the solid as well as various scalars, the usual results for a nonoriented (isotropic) porous structure are obtained. When g is assumed to be a function of the local porosity gradient as well, we derive a new expression for g applicable to oriented (anisotropic) porous structures.For a Newtonian fluid with a constant viscosity, the divergence of the local volume-averaged extra stress tensor is proportional to the Laplacian of the averaged velocity vector. Boundary conditions for the averaged velocity vector are discussed. Three problems are solved for the flow of an incompressible Newtonian fluid in a nonoriented permeable medium. These solutions, as well as an order-of-magnitude analysis, suggest that we may often neglect both the Laplacian of average velocity and the boundary conditions for the tangential components of averaged velocity at an impermeable wall.Two specific constitutive equations for g are proposed for the flow of incompressible Noll simple fluids in nonoriented porous structures. Flow through a porous medium bounded by an impermeable cylindrical surface is solved for these two constitutive equations, and the results are compared with previously available experimental data.

Notes: The mechanism by which oxygen is transported from capillaries into tissue for cell respiration has been a subject of interest since the work of August Krogh in 1919. Mathematical analyses of the diffusion process have been obtained by several investigators (11, 12, 18), but are limited in scope by the complicated nature of the problem. However, the solutions that exist give enough insight to inspire a more rigorous treatment. With the advant of electronic computers more complete analyses are now within reach (16).The human brain accounts for less than 4% of the total body weight, yet it uses 15 to 20% of the total body oxygen consumption. This high metabolic rate makes the brain particularly vulnerable to any change in the normal oxygen supply. The fact that following a cessation of cerebral blood flow, unconsciousness develops within 10 sec., and irreversible damage within 10 min., demonstrates the importance of understanding the dynamics of oxygen supply to the brain.