Search

Considerations based on the Maron theory show that for solutions of an amorphous polymer in a solvent the values of A obtained from heats of solution, heats of dilution, osmotic pressure, and vapor pressure should all be identical. Here λ = (aWaT)2, where μ is the Maron theory interaction parameter, T the temperature, and v2 the volume fraction. On the other hand, for solutions of a crystalline polymer in a solvent the A's obtained from heats of solution should always differ from those given by the other measurements. The difference is due to the heat of fusion involved in melting the crystalline portion of the polymer. A method is presented by means of which this heat of fusion can be found from the observed heats of solution, and this heat of fusion can either be used as a relative measure of crystallinity, or it can be converted to percent crystallinity when the heat of fusion of the completely crystalline polymer is known. The method described is very precise, and highly sensitive to low degr...

Maron's theory of the thermodynamics of nonelectrolyte solutions has been used to obtain from published osmotic pressure, light-scattering, and partial miscibility data for the system polystyrene-cyclohexane (μ° – σ°) and σ° values as functions of molecular weight and temperature. These functions were combined then with intrinsic viscosities to calculate the partial miscibility curves for five samples of polystyrene of varying molecular weight. Except for minor discrepancies, which are shown to be due to molecular weight heterogeneity, the calculated plots are in good agreement with the experimental results. The data have also been used to evaluate the interaction parameters at the points of critical miscibility. These calculations confirm the predictions of Maron and Nakajima that for infinite molecular weight these parameters go to the limits μc° = 2/3, σc° = 1/6, (μc° – σc°) = 1/2, and μc = 2/3.

The flow properties of low temperature GR-S X-667 latex have been investigated in a new low shear stress viscometer at 30.0°C. The range of shear stresses covered was 0.08 to 20.0 dynes/cm.2. The concentration was varied from 5% to 59% solids in approximately 5% increments. The latex was found to be completely Newtonian below a volume fraction of v = 0.246. Above v = 0.246 the latex flow was Newtonian at low shear stresses, changing to non-Newtonian as Fm was increased. Furthermore, the range of Fm over which the flow was Newtonian proved to be a linear function of the concentration between v = 0.320 and v = 0.544. Above 0.544 the flow behavior was non-Newtonian over the entire range of Fm covered in this study. It was found possible to express the flow behavior of this latex as a function of Fm and v in terms of the modified Eilers equation involving two parameters α and β; α was found to be a constant, while β was a function of Fm. This equation predicts that at infinite dilution 2α should equal the Einstein 2.50 for suspensions of spheres. When the observed values of 2α were corrected for the volume occupied by the layer of adsorbed soap on the latex particles, the value obtained for this factor was 2.52 ± 0.01. Again β, which is the volume fraction at which the viscosity becomes infinite, was found at the lower shear stresses to be of an order of magnitude expected for spheres with a mixture of orthorhombic and tetragonal packing. With increasing Fm the packing became tighter, and appeared to approach the densest, or rhombohedral, packing. The equation was used also to obtain an expression for the deviation of the latex from Newtonian behavior as a function of Fm and v. At any given shear stress the deviation was the greater the higher the volume fraction. Again, at volume fractions above v = 0.246, and below v = 0.544 the degree of non-Newtonian behavior decreased regularly with decrease in Fm, reaching a Newtonian limit at some low shear stress. However, no Newtonian limit was found above v = 0.544.

Maron's theory of the thermodynamics of non-electrolyte solutions has been used to examine the light scattering behavior of both dilute and concentrated solutions of polystyrene in several solvents. It is shown that for each polymer-solvent system at a given temperature the behavior can be characterized, as in the case of osmotic pressure, by a single interaction parameter μ which varies linearly with volume fraction v2, i.e. μ=μ°+σ°v2, where μ° and σ° are constants. Introduction of this equation into Maron's theory leads to functional plots in dilute solutions which are linear even when the conventional plots of HC/τ versus C plots are not. These plots yield the molecular weight from the intercepts and (μ° − σ°) from the slopes. However, to obtain the individual values of μ° and σ° data are required for concentrations higher than those normally used for molecular weight determinations. It is further shown that once μ° and σ° are evaluated, they may be used to calculate satisfactorily the turbidities up to a concentration of v2 = 0.62. Under the same circumstances the Flory-Huggins interaction parameter χ does not reproduce the observed data as well even when χ is taken to be a linear function of v2.

Summary A study has been made of the effect of polydispersity on the flow of synthetic latices by investigating the behavior of mixtures of type III and type V GR-S rubber latex in all proportions and up to total rubber contents of 55%. The two latices differed in volume to surface average particle diameter by a factor of two. Mixtures of the two latices showed Newtonian flow below total volume fractions of φ = 0.25, and non-Newtonian flow at higher concentrations. Up to φ = 0.47, a single exponential flow equation covered the entire range of shearing stresses used. Above this concentration, separate exponential flow equations had to be used for the low and high stress ranges. The flow constants N and η were found to depend both upon the total concentration of polymer and the proportion of the two latices in the mixtures. The equation log ( η ′ / η 0 ) = α b φ 1 − α φ again represented well the dependence of η′ on the total concentration of polymer, and yielded α's dependent only on the proportion of the two latices in the mixtures. At low concentrations this equation gave for the proportionality constant between η′ and φ values ranging from 2.20 to 2.40 as against the Einstein value of 2.5. Again, extrapolation of the equation to infinite viscosity gave volume fractions of rubber plus adsorbed soap which varied with the proportion of the two latices, and showed that an increase in heterogeneity of particle-size distribution leads to a considerably tighter packing of the spherical particles. The maximum packing fraction of φ t = 0.81 was observed at a polymer volume fraction of 0.76 type V latex in the mixture. This is to be compared with φ t = 0.74 for the type III latex and φ t = 0.77 for the type V latex. These results show that at the point of tight packing the larger spheres are arranged rhombohedrically, with the smaller spheres filling the voids between the larger particles. They also indicate that polydisperse systems should exhibit flow to concentrations higher than corresponding monodisperse systems.

The Maron theory of nonelectrolyte solutions has been used to examine the vapor pressure behavior of solutions of polyisobutylene in benzene and in cyclohexane between 25 and 65°C and in n-pentane between 25 and 55°C. The results show that the theory can predict the vapor pressures of these solutions over the full concentration range, provided the polymers involved are amorphous. When the polymers are not completely amorphous, then discrepancies between theory and experiment are observed at high solution concentrations which are due to persistence of polymer order in the highly concentrated solutions. A method is described by means of which these discrepancies can be used to obtain on an absolute basis both the degree of polymer order in solution and in the original polymer samples used.

Type V GR-S latex is non-Newtonian at all concentrations above about 0.25 rubber by volume, Newtonian below. The exponential flow equation may be used to convert experimental observations in capillary and concentric cylinder viscometers to rate of shear and shearing stress values. For volume fractions above about 0.40, however, separate exponential equations must be used for the high and low shearing stress ranges, due to the inadequacy of a single two-parameter equation to express the wide variation observed. The constants N and η′ of the exponential equation increase in a regular manner with concentration for both shearing stress ranges up to a volume fraction of about 0.60, and either decrease or “level off” from 0.60 to 0.62, the highest concentration studied. This unusual behavior, equivalent to a decrease in viscosity with increasing volume fraction, is tentatively attributed to the setting in of a degree of order, which would restrict Brownian motion of the particles, and permit flow to occur along more or less definite planes. An empirical extension of the Einstein equation of the form log η′ η 0 = bZ, where Z = αφ 1 − αφ , holds well all the way from 0 to 0.60 volume fraction of rubber. The limiting form of this equation at low volume fraction is identical with that of the Einstein equation, with a factor not far from Einstein value of 2.5. Extrapolation of the formula predicts infinite viscosity at a total volume fraction of rubber plus an adsorbed soap monolayer of about 0.75-0.77, a value which corresponds closely to the maximum obtainable by tight packing of spheres.

A study has been made to show that Maron's theory of the thermodynamic behavior of nonelectrolyte solutions is applicable to systems with low molecular weight solutes as well as polymers. In this connection were investigated three systems of varying degrees of complexity, namely: (1)carbon tetrachloride–cyclohexane, (2)acetone–chloroform, and (3)water–ethyl alcohol. The first of these systems yields a small positive interaction parameter μ which is essentially independent of volume fraction v2 but which decreases with temperature T. On the other hand, the system acetone–chloroform gives μ's which are large, negative, and strongly dependent on both T and v2. Similarly the system water–ethyl alcohol yields μ's which vary with both v2 and T, but in this case they are large and positive. The relations deduced for the μ's as a function of T and v2 reproduce very satisfactorily the available thermodynamic data for these systems. A discussion is also given of the significance of the observed interaction parameters.

Various technical applications of latex call for the formation of a latex film and its drying by evaporation of the water in which the rubber is dispersed. The same problem is encountered also in the preparation of dried latex films, both uncompounded and compounded, for film testing for tensile, elongation, and other properties. Very frequently the preparation of such films is complicated by the development of surface cracks and ridges in the drying films, which mar the appearance of the films, detract from their strength, and make testing difficult and uncertain. These imperfections in the films generally arise from the formation of impervious surface skins on the drying latex. Such impervious skins lead to shrinkage and subsequent cracking of the films on the one hand, and to bubble formation on the other. The result is that the preparation of dried latex films by this method becomes a very slow process, which yields as a rule films of inferior appearance and quality. It was found in this laboratory that many of the difficulties encountered in the preparation of latex films by evaporation can be overcome by appropriate control of humidity in the initial drying stages, and the use of an agent which prevents formation of the impervious surface skins on the drying latex. The agent found to be most suitable with alkaline latexes for the latter purpose is carbon dioxide gas. By utilizing moisture and carbon dioxide in a definite technique, films can be prepared from various latexes which are smooth and uniform in thickness and possess improved appearance and properties. The purpose of this paper is to describe the technique developed for preparation of dried latex films by casting, and to give some results for tensile and elongation obtained on GR-S, Hycar, Neoprene, and natural latexes.

Unpolarized incident light was used to determine the polarization ratio, ρ u , for a series of polystyrene and butadienestyrene latices. The experimentally observed ρ u 'S were found to differ appreciably from those predicted by Mie theory. The discrepancy is shown to be due to inherent anisotropy of the latex particles, and this conclusion is confirmed by a study of the polarization ratios with vertically and horizontally polarized incident light. This presence of anisotropy makes uncertain the determination of the particle size of latex from ρ u measurements at 90° by use of Mie theory. Since latex particles are spherical, the anisotropy is believed to arise from nonrandom orientation of polymer chains in the colloidal latex particles.

Turbidities and Rayleigh ratios have been measured at 25°C. and at wave lengths of 4358 and 5461 A. for benzene, carbon tetrachloride, toluene, and methyl ethyl ketone. The results obtained are in agreement with the theoretical Einstein equation for the scattering of light by pure liquids, the data yielding for Avogadro's number a value of 5.98 × 1023 ± 4.7%, as against the accepted value of 6.02 × 1023. Further, the Rayleigh ratios are in accord with the published measurements of Doty, Zimm, Brice, and their co-workers, but not with the older literature or the values of Stamm and Button.

Six samples of polyvinyl chloride (PVC) ranging in molecular weight from 23,200 to 155,400, were used to determine in cyclopentanone the intrinsic viscosities, molecular weights, and the polymer-solvent interaction parameters over a wide range of temperatures. The solutions were found to behave normally and to exhibit no evidence of “aggregate” formation. The molecular weights obtained were independent of temperature and agreed well with those measured in cyclohexanone. The interaction parameters observed were independent of concentration but were dependent on temperature and molecular weight. Again, the intrinsic viscosities as functions of molecular weight were found to follow the Mark-Houwink relation between 20° and 60°C with a constant a, but with a K linearly dependent on the temperature. From the temperature coefficient of the interaction parameter, it was deduced that, above a molecular weight of ca. 13,000, the heats of mixing of completely amorphous PVC with cyclopentanone as well as th...

Although dissymmetry of light scattering is a simple and attractive method for determination of latex particle size, it is limited in application to values of D/γm below 0.60 for monodisperse systems, and practically inapplicable to polydisperse ones. Further, its use requires Fresnel corrections, and a knowledge of the refractive index ratios of the systems investigated. To obviate some of these difficulties, a method is described which utilizes the intensity ratios for two forward angles differing by 10°, such as 45°/35°. With such forward angle ratios Mie calculations show, and experiment confirms, that particle diameters up to D/γm = 2.00 or higher can be measured for monodisperse systems. Again, no Fresnel corrections are required, and the results are practically independent of refractive index ratio between 1.15 and 1.20. However, like the dissymmetry method, the forward angle ratio method is very sensitive to polydispersity, and results are obtained here which vary with the distribution of the systems, the angle ratio, and the wavelength of light used.

The Maron theory was used to deduce from mostly osmotic pressure and light-scattering data the interaction parameters as a function of concentration, temperature, and polymer molecular weight for solutions of polystyrene in seven different solvents. These interaction parameters were then utilized to calculate the activity coefficients of the solvent in more concentrated solutions for comparison with experimental values deduced from vapor pressure measurements. When order was absent in the polymers (all cases but two), the agreement between calculated and observed activity coefficients was found to be very good. In the two cases where order was present, the activity coefficients observed in the concentrated solutions were higher than those calculated. These deviations were employed to find the degree of polymer order both in solution and in the original polymers.

A scheme is proposed for systematizing the handling of viscometric data through use of the variables apparent fluidity φa and shearing stress F. The rate of shear g(F) is obtained from the relation: g(F)/F =φa[1+Δ(F)]. Methods are presented for obtaining the correction term Δ(F) from capillary and concentric‐cylinder viscometric data. Use of the proposed scheme is illustrated.

From Mie theory calculations it is shown that the angle, θ 1 , at which the first minimum in scattering intensity occurs is given by the relation D λ m sin θ 1 2 = 1.062 − 0.347 m where D is the diameter of the scattering spheres, λ m is the wave length of light in the medium, and m is the refractive index ratio of scattering particles to medium. This relation holds for m = 1.00 to 1.55, and applies to incident unpolarized light and the vertical component of the scattered radiation. Using this equation, particle diameters have been determined at several wave lengths for 23 synthetic latices ranging in D from 2080 to 10,100 A. The results show that D is independent of λ m when the systems are monodisperse. For polydisperse systems D increases with λ m , and only low λ m 's yield the weight average for D . The variation of D with λ m can be used as a qualitative index of size distribution. This method can be made very rapid, and under suitable conditions it can be used to measure diameters ranging from about 0.18 to 4 microns.

The soap titration method for determination of particle size and specific surface area of synthetic latices can be applied only to latices the initial soap content of which is less than that required to yield the critical micelle concentration of soap in solution. To handle latices the soap contents of which exceed the cmc, a procedure has been developed involving the titration of the latex with another the surface of which is still unsaturated by soap. The latter latex is used to take up the excess soap and to yield both latices in equilibrium with the cmc of the soap. The end point of the titration can be determined, as before, by conductance. The end points are sharp, and the requisite data obtained permit a ready calculation of the latex particle size for both fatty and rosin soap latices. Further, the results obtained are in excellent agreement with those obtained by soap titration.

A capillary viscometer is described for the study of the flow behavior of non‐Newtonian fluids at shear stresses of 0.06–20 dynes/cm2. The all‐glass instrument consists of two units connected by a ball joint, one unit of which contains the capillary while the second is essentially a manometer. As the test fluid moves through the capillary, the rate of rise of the fluid in the manometer is observed. This measurement of the column height as a function of time allows calculation of both pressure drop and rate of flow. Different combinations of capillary and manometer units can be employed interchangeably to study liquids of widely varying viscosities. Data and methods are presented which show that the instrument can be used to determine the absolute viscosities of Newtonian liquids and the flow curves of non‐Newtonian fluids over the indicated shear stress region.

A method is presented for the determination of the surface area of carbon blacks by adsorption of laurate or myristate soaps from aqueous solution. It involves conductometric titration of the blacks in aqueous dispersion with the respective soaps. Comparison of results obtained by this method on eleven samples of various blacks with nitrogen adsorption data indicates that up to a nitrogen area of 50 m.2/g. the two methods yield identical results if areas of 23.2 and 35.2 A2 are employed for the laurate and myristate molecules, respectively. However, for the finer blacks the two methods yield divergent surface areas, with those by nitrogen being the larger. These differences are ascribed to surface irregularities present in the higher area blacks, which can be expressed as a surface roughness factor, f, defined as the ratio of the area of a black as determined by nitrogen adsorption to that given by the soap titration. These roughness factors have a value of unity for the smaller area blacks, and increase for the finer blacks to values above two. Since a nitrogen molecule is small and can be accommodated in spaces inaccessible to larger molecules, nitrogen areas appear to be too large for correlation with rubber reinforcement. On the other hand, the areas given by larger molecules, such as soap, should be much nearer to those involved in the interaction of carbon black and rubber.

Data on the osmotic pressures of both dilute and concentrated polystyrene and polyisobutylene solutions in various solvents were examined in terms of Maron's theory. It is shown that for each polymer-solvent system at a given temperature the behavior over the entire concentration range studied can be characterized by a single interaction parameter, mu, whose dependence on volume fraction, v2, of polymer is given by mu = mu-o + sigma-o(v2), where mu-o and sigma-o are constants. For two systems, where data are available, this linear relationship was found to hold from v2 = 0 up to v2 =0.90. With this relationship for mu, Maron's theory leads to functional plots for dilute solutions which are linear even in systems where the usual II/C vs. C plots are not straight lines. These plots yield the molecular weights from the intercepts and the quantities (mu-o -- sigma-o) from the slopes. However, the individual values of mu-o and sigma-o cannot be obtained from the dilute solution data. To arrive at these, data are required at concentrations higher than those customarily used for molecular weight determinations. (Author)

Dissymmetry of light scattering at 45 to 135[ddot] has been used to determine the particle size of both mono- and polydisperse latices. For the monodisperse latices the particle diameters found are independent of the wavelength used, and in good agreement with results obtained by electron microscopy and transmission measurements. However, because of the nature of the Mie plots and experimental difficulties, the method is limited primarily to diameters less than 0.60λm. On the other hand, the diameters obtained for polydisperse latices increase with increase in Am to values considerably above the weight-average diameters of the systems. It is shown that this variation of D is a function of the particle-size distribution of the latices as well as the wavelength, and that it is not possible to define for such latices conditions of measurement which will correspond to a definable type of average for the diameters obtained.

From Mie theory calculations it is shown that the angular positions of the first two minima and maxima in the intensity of the vertically polarized component of the scattered light are given by the relations Dλm sin θi2 = ai − bim and Dλm sin φi2 = Ai − Bim where D is the diameter of the scattering spheres, λm the wavelength of light in the medium, m the refractive index ratio of scattering particles to medium, θi the angles at which minima occur, and φi the angles for the maxima. The subscript i = 1 or 2, depending on the order of the minimum or maximum. Finally, ai, bi, Ai, and Bi are constants dependent only on the value of i. These equations hold for m = 1.00 to 1.55, and apply to unpolarized incident light. Using these equations, particle diameters of four latices ranging in size from 3470 to 10,500 A. have been determined at several wavelengths. The results show that for monodisperse latices the D obtained is essentially the same for all orders of the minima and maxima, as well as the two wavelengths used. For polydisperse latices D is essentially the same for the various minima and maxima at a given wavelength, but it increases as λm is increased. Only at the lowest wavelength used, λm = 2711 A., does D correspond to the weight-average diameter obtained from electron microscopy. The particle sizes which are theoretically determinable under suitable conditions from the first two maxima and minima range from 0.18 to 7.7 microns in diameter. It is also demonstrated that equations of the above form apply to the first four minima and maxima in the total intensity of the scattered light at m = 1.20.

Measurements made with both concentric cylinder and capillary viscometers show that type III synthetic latex is Newtonian at concentrations below ca. 25% solids, and non-Newtonian above. Analysis of the data demonstrates the absence of either a yield point at low shearing stresses or a limiting viscosity at high, ruling out Bingham body behavior. The linearity of logarithmic plots of the data indicates the applicability of the exponential flow equation of Farrow, Lowe, and Neale for the conversion of experimental observables to rate of shear—shearing stress values. However, above ca. 45% solids, separate exponential flow equations must be used at high and low shearing stresses. A composite rate of shear—shearing stress plot, including values obtained from three capillaries and the concentric cylinder viscometer, yields a single smooth curve, confirming the validity of the conversion procedure. Comparison of values of the two constants of the exponential equation, as determined independently using the capillary and concentric cylinder instruments, shows agreement between the two types of viscometer. Similar results have been observed for other latices, both synthetic and natural.

A study has been made of the flow behavior at 30°C. of solutions of three samples of polymethyl methacrylate in three solvents in both the Newtonian and non-Newtonian regions. The concentration range covered was approximately 0–16% by weight. By application of the Ree-Eyring theory it was found possible to represent the observed data in terms of two flow units, and to obtain from the parameters of the theory the effective hydrodynamic volume occupied by the polymer in solution, eυ2. Here υ2 is the volume fraction of the polymer and e an effective volume factor. The volume factor, e, was found to be dependent on both solvent and molecular weight only up to ca. υ2 = 0.04. Beyond υ2 = 0.04, e is independent of solvent and is a function of molecular weight only. These observations indicate that chain entanglement in solution is not probable. Rather, the polymer chain curls on itself, and compacts to a small mass when the effective concentration eυ2 = 1 and the viscosity becomes infinite. The concentration at which this occurs, υ∞, is given by ∞ = 1/e∞. The quantity e∞ has been found to be a linear function of molecular weight, M. Further, the value of e at υ2 = 0, e0, has been found to be an exponential function of M, and to yield expressions relating intrinsic viscosity to molecular weight in good accord with published results.

Maron's theory of the thermodynamics of nonelectrolyte solutions has been applied to the condition of critical miscibility in polymer solutions on the basis of an interaction parameter μ linearly dependent on the volume fraction v2, i.e., μ=μ°+σ°v2. The results yield expressions for μc, μc°, σc°and (μc° − σc°) at the point of critical miscibility which are related to the molar volume ratio, the critical volume v, and effective volume factors. For the limit of infinite molecular weight these equations yield σc°= 1/6, (μc° − σc°) = 1/2 and μc= μc°=2/3. These results differ from the χc=1/2 predicted by the Flory-Huggins theory on the basis of an interaction parameter χ independent of concentration.

The generalized flow theory of Ree and Eyring has been applied successfully to the extensive data of Maron and Fok (3) on the flow behavior of X-667 latex in both the Newtonian and non-Newtonian regions. By postulation of the presence of two flow units, a Newtonian water unit and a latex particle flow unit which may be non-Newtonian, an equation involving three parameters was obtained which reproduced very well dependence of the viscosity on rate of shear at all temperatures and concentrations. Further, the theory predicted the correct dependence of the parameters on temperature, from which could be obtained the heats of activation for flow of both the water and latex particles. Both heats of activation were found to be independent of concentration, with that of the water being the same as for the pure liquid. The theory does not predict the concentration dependence of the parameters. However, empirical equations for such dependence were deduced, from which it was possible to ascertain the values of the parameters in the Newtonian range below a volume fraction ν = 0.25, and to calculate therefrom Newtonian relative viscosities in good agreement with the observed ones at all temperatures and concentrations. Further, on extrapolation to ν = 0, ν sp ν t was found to be 2.60 as against the expected Einstein value of 2.50 for suspensions of spheres, while extrapolation to infinite viscosity gave a volume fraction in very good agreement with that expected for the rhombohedral close packing of nonuniform spheres. From the combination of the theoretical and empirical equations were obtained also expressions for the values of the “molecular” parameters of the theory as functions of temperature and concentration, and the entropy of activation for the flow of the latex particles. The distance between equilibrium positions of the water molecules was found to vary from 13 A. to zero between ν = 0 and the volume fraction at which the viscosity becomes infinite. On the other hand, the distances between equilibrium positions and planes of latex particles varied from infinity to zero between the same concentration limits.

Six samples of polyvinyl chloride (PVC), ranging in molecular weight from 23,200 to 155,400, were used to determine in tetrahydrofuran the intrinsic viscosities, molecular weights, and the polymer-solvent interaction parameters over a range of temperatures. The solutions were found to behave normally and to exhibit no evidence of “aggregate” formation. The molecular weights obtained were independent of temperature and agreed well with those measured in cyclohexanone and cyclopentanone. The interaction parameters observed were independent of concentration and molecular weight, and functions only of temperature. Again, the intrinsic viscosities are related to molecular weight by the Mark-Houwink equation between 20° and 50°C with a constant a and with a K linearly dependent on the temperature. The value of the temperature coefficient of the interaction parameter obtained by osmometry agrees well with that found by microcalorimetry and predicts that heats of mixing of completely amorphous polymer wi...

A study has been made of the flow behavior of solutions of a single sample of polystyrene in five different solvents in both the Newtonian and non-Newtonian regions. The concentration range covered was approximately 0-20% by weight, and the temperatures were 20, 30, 40, and 50°C. By application of the Ree-Eyring theory it was found possible to represent the observed data in terms of two flow units, and to obtain from the parameters of the Ree-Eyring equation the effective hydrodynamic volume occupied by the polymer in solution, ev, where v is the volume fraction of polymer and e an effective volume factor. For any given solvent e varies from [η]/2 at v = 0 to e = ca. 4 at v∞, the volume fraction at which ev = 1 and the viscosity becomes infinite; and, except in very dilute solutions, e is essentially independent of the temperature. Again, for any given value of v, e is a function of the nature of the solvent only between v = 0 and v = 0.04. Above the latter concentration e is completely independent of the solvent in which the polystyrene is dissolved, and becomes merely a function of the solution concentration. These results indicate that chain entanglement in solution, if it does occur, can take place only below a concentration of v = 0.04. Above v = 0.04 no entanglement can be present, and the configuration of the polymer chain, as reflected in the effective volume they occupy, is governed only by the nonspecific factor of the space available for occupancy by each molecule. As this space decreases with increase in concentration, the polymer molecule is forced to reduce its effective volume by coiling into a tighter mass. This process continues until ev becomes equal to unity, at which point the state of tightest coiling is reached without change in the nature of the flow. In the latter state the polymer molecule is shown to occupy a volume 2 to 3 times, and to have a cross section 1.26 to 1.44 times, that of the polymer in bulk.

The oxidation of emulsion polybutadiene has been followed by observation of changes in gel content, swelling index, intrinsic viscosity, and infrared spectra. The results show that, whereas simple heating of the polymer leads primarily to crosslinking, the presence of oxygen causes both scission and crosslinking, with the predominance of either of these processes being dependent on the amount of oxygen absorbed. Increase of temperature or degree of oxidation favors crosslinking. During oxidation acid, ester, aldehyde, ketone, ether, and alcohol groups appear, but their relative proportions remain essentially constant. Also, a large extent of the oxidation occurs without resultant decrease in unsaturation, indicating, in all probability, that the double bond is not the direct point of oxygen attack.

Using a Beckman spectrophotometer, light transmission measurements were made on 12 polydisperse butadiene-styrene latices and one essentially monodisperse polystyrene latex. Cell lengths of 1, 5, and 10 cm. were used, and the wave length range covered was 3500–10,000 A. in 500-A. intervals. By employing total polymer content, c, rather than total solids, ratios of turbidity, τ, to c at c = 0, i.e., (τ/c)0, were obtained which were independent of the cell length at each wave length. Application of Mie theory to the measurements yielded particle diameters showing no regular variation with wave length and in good accord with the weight-average diameters of the latices obtained by electron microscopy. The validity of the average was checked by differential creaming of the latices and also by calculation of the averages to be expected from the latex particle size distributions on summation of the scattering coefficients. The range of diameters covered by the data was α = πDw/λm = 0.30 to 4.74, where DW is the weight-average diameter and λm the wave length of light in the medium. For light of 10,000 A. in vacuo, the upper limit of α corresponds to Dw = 11,400 A.

The generalized flow theory of Ree and Eyring has been applied to the data of Maron and Belner (5) on the flow behavior of X-667 latex at low rates of shear in both the Newtonian and non-Newtonian regions. Whereas in the intermediate shear regions, Maron and Pierce (1) were able to represent the results in terms of two flow units, three such units were required at the low rates of shear. These were a Newtonian water unit, and two latex flow units of different relaxation times. As a result, the dependence of viscosity on rate of shear, at any given concentration in the non-Newtonian range, required five parameters for satisfactory representation of the experimental data. Equations have also been deduced for the dependence of the parameters on concentration, and these were used to predict the Newtonian viscosities measured by Maron and Belner below v = 0.25. Good agreement was found between the observed and calculated viscosities. Further, on extrapolation to v = 0, η nsp v t was found to be 2.67 as against the expected Einstein 2.50 for suspensions of spheres, while extrapolation to infinite viscosity gave a volume fraction in very good accord with that expected for rhombohedral close packing of spheres.

The Maron theory of polymer solutions was used to examine the available osmotic pressure, light scattering, and heat of dilution data for polymethyl methacrylate (PMMA) in fourteen solvents. No vapor pressure or heat of solution data could be found for these systems. Wherever possible interaction parameters, μ, were evaluated as functions of volume fraction, temperature, and molecular weight, as well as X, the temperature coefficient of μ. The equations deduced allow in some instances the prediction of the vapor pressure behavior of solutions of noncrystalline PMMA and also the heat of mixing of random amorphous polymer with solvent. The data were not sufficient to establish the complete dependence of μ on concentration, molecular weight, and temperature for any PMMA-solvent system.

The Maron theory of nonelectrolyte solutions was used to analyze available osmotic pressure data for solutions of polyisobutylene in benzene, chlorobenzene, toluene, and cyclohexane and to evaluate the interaction parameters, μ, for these systems. The results obtained show that for the benzene and chlorobenzene solutions the μ's are linear functions of volume fraction and temperature, but they are not dependent on molecular weight. On the other hand, the solutions of polyisobutylene in toluene and cyclohexane show a dependence of μ on molecular weight and, in all probability, also on volume fraction and temperature. However, the data available are insufficient to establish the exact dependence of μ on v2 and T for the latter two systems.

Heats of solution and dilution for polyvinyl chloride (PVC) in cyclohexanone and tetrahydrofuran have been measured at 30°C for samples ranging in molecular weight from 23,200 to 155,400, as well as for several polymers with special histories. The results show that normal PVC is a mixture of crystalline and glassy phases. The degree of crystallinity is relatively low, ca. 20%, and decreases with increase in molecular weight. Annealing does not modify the crystallinity significantly, but quenching from the melt can reduce it to low levels and possibly even to zero. A tentative method is presented for estimating the degree of crystallinity from residual heats obtained from heat of solution measurements. These residual heats are independent of concentration or solvent used and are properties only of the polymer. Also, Maron theory λ's have been calculated from the heat of dilution data, and these agree with the λ's evaluated from osmometry. This agreement confirms the Maron theory argument that heat...

The flow behavior of a GR-S latex prepared at low temperature was studied at volume fractions up to 0.60 over a shear stress region of 50–800 dynes cm. 2 , and over a temperature range of 20° to 50°C. The latex was found to be Newtonian below a volume fraction of 0.25, and non-Newtonian above this concentration. Unlike some preceding latices investigated, the dependence of flow of this latex on shear stress could not be represented by the exponential equation, nor could the concentration dependence be expressed by the logarithmic equation employed before. However, the flow behavior of this latex could be represented by a two-parameter modified Eilers equation in which one of the parameters is a function of the temperature, the other, of shear stress. The limiting form of the expression as the volume fraction v approaches zero yielded η sp v = 2.50 ± 0.05 , as against the expected Einstein value of 2.50 for dilute suspensions of spheres. Again, the expression predicted for the volume fractions at which the viscosity becomes infinite values ranging from 0.69 at Fm = 50 to 0.75 at F m = 800 dynes cm. 2 . These values indicate that the close packing of the nonuniform spheres in the latex is primarily tetragonal, with a transition to rhombohedral packing as Fm increases. The results also show that the variation of the viscosity of the latex with temperature is due primarily to the change in the viscosity of the water, and only secondarily to the decrease in the energy of activation for flow of the latex with increase in concentration. Finally, it was found that the deviation of the latex from Newtonian behavior is essentially independent of temperature, and dependent only on shear stress and concentration. The dependence on shear stress is more complicated, with maxima occurring in the curves at shear stresses which are the lower the higher the volume fraction of polymer in the latex.

The Tian-Calvet microcalorimeter, as modified by Maron and Filisko, has been used to measure the heats of solution and dilution of a sample of polyethylene oxide (PEO) of M2 = ca. 6000 in chloroform, methylene chloride, and water. By applying the Maron theory to these measurements, it was found from the organic solvent data that the heat of fusion of the PEO sample is 57.1 cal/g polymer, independent of the solution concentration or the solvent used. However, the solution behavior in water was different because PEO in organic solvents exists as a random coil, whereas in water its conformation is helical. By combining the water and organic solvent data, it can be shown that conversion of the random coils to the helix involves the evolution of 39.3 cal of heat per gram of polymer. This heat evolution accounts, in all probability, for the solubility of PEO in water.