Search

Several physical properties of four samples of linear α, ω-dimethoxyfluoropolyethers, varying in length from 6 to 14 chain atoms, i.e. carbon plus oxygen atoms, are reported and discussed. Among them are boiling points, glass and melting transition temperatures, specific volumes and their temperature coefficients, refractive indices and thermodynamic data such as cohesive energy density and melting enthalpies. The experimental data have been analysed, when possible, on the basis of the group contribution method, the results of which are in agreement with literature data. The above results are compared with those of two analogous series of fluoropolyethers whose chain ends are, respectively, OCF 2 H and OCF 3 . As expected, with increasing chain length, the effect of the OCF 2 H and OCH 3 ends decreases progressively so that for chains longer than approximately 15–20 atoms, the differences among the three series should disappear.

The thermal stability of representative hydrofluoropolyether (HFPE) and hydrofluoroether (HFE) compounds has been evaluated. The observed stability order appears to be correlated with the nature of the hydrogenated chain ends; in particular, molecules having fully hydrogenated chain ends (OCH3 and OC2H5) show a significantly lower stability compared with the OCF2H terminated compounds. The main degradation products suggest, however, that the same primary reaction is responsible for the decomposition of all the compounds examined; this reaction involves the fragmentation of the RfOCxHyFz bond with fluorine transfer between the two carbon atoms close to the oxygen, leading to the formation of a hydrofluorocarbon CxHyF(z+1) and an acyl fluoride or a ketone.

Reported here are several physical properties of the following hydrofluoro-compounds: (1) hydrofluoropolyethers (HFPE) of general formulae: HF 2 CO(CF 2 O) n (CF 2 CF 2 O) m CF 2 H; (2) an experimental (HFPE) sample: CH 3 O(CF 2 O) n (CF 2 CF 2 O) m CH 3 ; (3) a linear hydrofluorocarbon (HFC): CF 3 CHFCHFCF 2 CF 3 ; a cyclic HFC: c-CF 2 CF 2 CF 2 CHFCH 2 ; hydrofluoroethers (HFE): C 4 F 9 OCH 3 , C 4 F 9 OC 2 H 5 and C 7 F 15 OC 2 H 5 . The properties of the HFPE series are monotonic functions of molecular weight M . Some of the properties of the HFC and HFE samples differ substantially from the molecular weight dependence shown by the HFPE series, indicating that the overall number and geometry of the H substitution, as well as the molecular geometry of the sample, could perhaps play some role in the control of physical properties.

Samples of virgin and a crosslinked fluorinated rubber, containing different amounts of carbon black and solvated with methylethyl ketone, have been investigated by differential scanning calorimetry. Although the polymer-solvent interaction increases with decreasing temperature, a process of solvent separation was observed for all of the systems. This process can be attributed to a T g regulation effect in which solvent crystallization occurs only when the crystallization temperature is higher than the T g of the system. The interaction between rubber and filler, approximately temperature independent, was found to not influence the glass transition of the rubbermethylethyl ketone systems, even when the carbon black content is on the order of 35 wt %.

The synthesis of a novel perfluoropolyethers (PFPEs) family has been carried out by the low temperature photooxidation of perfluoromethylvinylether. The macromolecular chains are terpolymeric in nature, being constituted by random sequences of the following groups: the regular –CF 2 CF(OCF 3 )O–, the orthoestereal –CF(OCF 3 )O–, and the acetalic –CF 2 O–. The presence of an ethereal oxygen in the short side chains represents the main structural difference with other PFPE families studied up to date [G. Marchionni, G. Ajroldi, G. Pezzin, in: S.L. Aggarwal, S. Russo (Eds.), Comprehensive Polymer Science, 2nd suppl., Pergamon Press, London, 1996.] Several fractions have been obtained, in the present work, by a single sample of the new PFPE, and some of their most important physical and molecular properties have been determined by a variety of different techniques. The properties change regularly with chain length, as expected. The group contribution analysis previously applied by our group to other families of PFPE has been found to be applicable also to chains of infinite length of the present family, which contains oxygen in the side chain.

Random poly(butylene isophthalate/terephthalate) (PBI/PBT) copolyesters of various composition and molecular weight were synthesized and characterized in terms of dilute solution viscosity, molecular weight and thermal properties. From the results of measurements carried out with a capillary rheometer at different temperatures (in the range 200–240 °C) on molten samples, the flow curves η-γ were obtained; these curves permitted to extrapolate the Newtonian viscosity η0, which was correlated with molecular weight. At a fixed temperature, samples with the same weight-average molecular weight and different composition showed a Newtonian viscosity increasing as the content of PBT comonomeric units is increased. Furthermore, the activation energy for melt flow, calculated using an Arrhenius-type equation, was found to be slightly affected by composition only.

The steady-state rheological behavior in the melt of poly(1,4-butylene isophthalate)s containing various amounts of sodium sulfonate groups has been investigated using a capillary rheometer. It has been found that the presence of SO3Na groups strongly enhances the melt viscosity and this result cannot be explained as due to the small increase of the glass transition temperature, but has to be attributed to the relevant interactions between ionic groups. Measurements carried out at different temperatures in the range 190–220°C have shown that the “Newtonian” viscosity of sulfonated poly(1,4-butylene isophthalate) is well correlated to temperature by an Arrhenius-type equation. The activation energy for melt flow has been found to increase linearly with SO3Na group content in the polymer. Fur-thermore, it has been found that for samples having a content of ionic groups of 5% or higher and a sufficiently high molecular weight the time–temperature superposition principle does not hold. © 1993 John Wiley & Sons, Inc.

The concentration dependence of the glass transition temperature T g of binary polymeric systems, for the Couchman entropic and enthalpic treatments, has been described by taking into account the temperature dependence of the specific-heat increments Δc p ( T ) and presence of a characteristic temperature T o . It is shown that at T o the analytical equations undergo a sudden change. Comparison of the equations obtained with a set of experimental data on the poly(vinyl chloride)-dibutyl phthalate system appears to be satisfactory. The treatment allows one to characterize a binary system with suitable thermodynamic parameters.

The cohesive energy density (CED) of linear hydrogenated and perfluorinated paraffins and ethers has been investigated with regards to its dependence on chain length and oxygen content. For high molecular weight compounds (i.e., polymers) the CED has been obtained by extrapolation or by the group contribution method. The oxygen contribution to cohesive energy is found to be very different in the hydrogenated ethers with respect to the perfluorinated ones, and, correspondingly, the dependence of CED on the oxygen to carbon ratio (O/C) is the opposite for the two series. A similar trend, previously described for the glass transition temperature (T g ) is therefore roughly interpretable as related mostly to CED, although intramolecular flexibility parameters also play a role

The present paper refers to the well known fact that the glass transition-composition curves of miscible binary systems (e.g. polymer-diluent and polymer-polymer mixtures) often exhibit a singular point when the Tg values of the two components are far apart. Two different equations apparently apply above and below the temperature of this singularity. Starting from Couchman's entropic treatment of the glass transition of mixtures, and from Angell's suggestion for the calculation of T0 (‘ideal glass transition temperature’ of the component with higher Tg), it is shown that below T0 an equation should be applicable to binary mixtures which differs considerably from the equations normally used in the literature. The new equation gives a satisfactory description of selected experimental data in the proximity of the component with lower Tg.

The theoretical treatment proposed by Couchman to describe the compositional dependence of T g for polymer-diluent systems has been applied, in its several forms, to selected pairs for which heat capacity data are found in the literature. The most accurate equation gives the least satisfactory description of the experimental T g data. This result seems to be accounted for by the assumption, first suggested by Couchman, that in some cases the entropy of mixing is not continuous at T g .

Three copolymeric perfluoroethers with the structure CF3[(OCF2CF2)p(OCF2)q] OCF3, having different p/q ratios, have been fractionated. The fractions obtained have been characterized by Gel Permeation Chromatography and 19F-NMR. The viscosity η the specific volume v and the glass transition temperature, Tg have been measured by standard techniques for all the above samples as well as for some other perfluorinated polyethers. The temperature dependence of viscosity of the unfractionated samples is described by the W.L.F. equation. The values of fg (fractional free-volume at Tg) and of af (free-volume expansion coefficient) are independent of composition, for p/q ratios from 0.53 to 1.15. The critical molecular weight, Mc, is of the order of 8–9,000. From the molecular weight dependence of specific volume, the contribution to the molar volume of the in-chain CF2 group and the excess molar free volume of the chain ends have been determined. The limiting value of Tg for an infinite molecular weight polymer was found to depend linearly on the compositional ratio O/C and the extrapolated values for polytetrafluoroethylene and for the homopolymer (CF2O)n were found to be respectively 200 K and 120 K.

It is shown that for diolic perfluoropolyethers having short chain lengths the “anomalous” (non monotonic) relationship between isothermal melt viscosity and molecular weight, reported in recent papers, is not anomalous at all but fully explainable when one takes into account the fact that the glass transition of the samples Tg is itself a function of the molecular weight. For low molecular weight polymers the viscosities have to be compared not at the same T but at the “normalised” temperature T−Tg.

Infrared studies of the poly(vinyl chloride) (PVC)/di-n-butylphthalate (DBP) system are reported. Spectra of DBP in different solvents seem to be indicative of the ability of DBP to interact with PVC chains in the same way as with polyesters and of competitive interactions between DBP molecules and DBP with PVC. Plots of the width at half-height of the carbonyl stretching vibration, measured at 25 and at −180°C, as a function of composition, show a sudden change of slope at about 60% PVC concentration by weight, indicating that above this polymer composition interaction between DBP and PVC dominates that between DBP molecules. This change corresponds to a “singularity” observed in the dependence of the glass transition temperature on composition.

Six samples of polypropylene produced by Montepolimeri were carefully characterized in dilute solution and studied, with a capillary extrusion rheometer, in large ranges of shear and temperature. The extrudate swelling behaviour is found to be “regular” as far as its shear stress and temperature dependence is concerned, but the influence of molecular weight is abnormal. With increasing molecular weight the amount of elastic recovery (as measured by the extrudate swell) decreases markedly and the shear modulusG (obtained from the data by making use of simplifying assumptions) appears to be an increasing function ofMw at all the shear stress investigated. No similar results appear to be reported in the literature.

The glass transition temperatures and the viscosities (over a very large temperature range) have been measured as a function of molecular weight, for two different series of perfluoropolyethers. The results have been interpreted according to well-known free-volume models. The Williams-Landel and Ferry equation describes the viscosity data with values of fg (the fractional free volume at Tg) close to 0.04 and with values of αf (the free-volume expansion coefficient) close to 7 × 10−4°C−1. The critical molecular weight, Mc, is approximately 5000–6000 for perfluoropolyethers from hexafluoropropene and 7800 for perfluoropolyethers from tetrafluoroethene.

The glass transition temperature has been measured carefully for the fluoroelastomer, Tecnoflon, a copolymer of vinylidene fluoride and hexafluoropropene, diluted with variable amounts of 2-butanone. Of the several different theoretical treatments found in literature, that presented by Gordon and coworkers appears to give the equation that describes well the experimental results, without making use of adjustable parameters. Some features found at very high diluent content can be interpreted with a treatment based on free volume.

Several problems concerning the molecular structure and the morphology of polyvinylchloride (PVC) have been extensively studied in the last 20 years [1], whereas relatively little is known on the kinetic and thermodynamic aspects of the interaction between the polymer and the additives which are commonly added to it. A typical PVC formulation, of the unplasticized type, may contain from 5 to 10 different additives such as lubricants, stabilizers, processing aids, etc. They are added to the resin and mixed, often at a temperature of about 100°C, to ensure a satisfactory distribution on the surface of the 100-150-um polymer grains. The resulting “dry blend” is then fed to the processing machine, usually a twin-screw extruder, where the original particles are converted into a continuous llmelt” in which the additives are assumed to be dispersed homogeneously.

Samples of poly(vinyl chloride) polymerized at temperatures from −60°C to +90°C, and therefore of different degrees of syndiotacticity and crystallinity, have been thermally treated in an attempt to reduce their content of ordered structures. Changes in order have been monitored by measuring Tg and the corresponding specific heat increment Δcp. Partially syndiotactic samples have higher Tg and lower Δcp than samples that are nearly atactic. After thermal treatment the differences decrease but do not vanish, even for samples heat-treated to about 250°C. This seems to indicate that some ordered structures are very stable even at relatively high temperatures. A quantitative estimate of the ordered structure content, based on Δcp data, is proposed.

The interaction of ox ligamentum nuchae elastin (native, purified, and soluble) with different solvents (water, ethylene glycol, methanol, and 2,2,2-trifluoroethanol) was investigated by means of differential scanning calorimetry. The unfreezable solvent content and the corresponding solvent to elastin residue molar ratio were determined from the melting endotherms of the freezable solvent present in these systems. The molar ratio obtained in the case of water and ethylene glycol agrees with hypothesis of a direct solvation of the main chain peptide group and is interpreted according to a model previously proposed for polyamides. Quite different molar ratios are obtained in the case of the two monofunctional solvents and no model can be proposed at present to explain their interaction with the protein.

Linear and branched PBTP samples were synthesized and characterized in terms of the intrinsic viscosity, the melt-flow-index and, for some, the melt viscosity over a range of shear rates at 250 °C.

Linear and highly branched poly(ethyleneterephthalate) samples were synthesized and characterized in terms of intrinsic viscosity, molecular weight and melt viscosity over a wide range of shear rates at several temperatures, in the range from 265° to 295 °C. Linear samples exhibited Newtonian behavior over a wide range of shear rates, while the branched ones became shear thinning at relatively low shear rates. Our experimental data, as well as data previously reported, were found to be described by a proposed correlation between the melt viscosity ratio and a branching index. Moreover, the activation energy for melt flow was found for the highly branched samples to be a little higher than that of the linear samples.

Samples of poly(vinyl chloride) plasticized with variable amounts of either dibutyl phthalate (DPB) or dicyclohexyl phthalate (DCHP) were investigated by dynamic-mechanical measurements in the β relaxation temperature range. In this range of temperature, a superposition of the relaxation due to the cyclohexyl group with the PVC β peak was found for the samples plasticized with DCHP. By studying the dependence of the activation energy and of the peak broadness on the DCHP concentration it was possible to show that the PVC β relaxation is reduced to zero at the critical plasticizer weight fraction W1 = 0.2. For the PVC-DBP series the β peak disappears at the same plasticizer content. These results strengthen the hypothesis that the β peak of PVC is due to a kind of cooperative motion since 1 mole of plasticizer for every 20 repeating units of the polymer is sufficient to suppress the PVC β relaxation.

This work is concerned with the mass spectrometric characterization of polyaryloxy- and polyaminophosphazenes 1a–e. The mass spectral data show that the pyrolytic breakdown of poly(di-β-naphthoxyphosphazene) (1a) leads to the formation of cyclic oligomers. The poly(di-arylaminophosphazenes) 1b–d decompose at low temperatures by a two-stage process involving amine evolution. Finally, in the case of poly(dipiperidinophosphazene) (1e) the thermal decomposition leads to the complete destruction of the polymer backbone, with formation of ammonia and elemental phosphorus. The wide range of thermal decomposition pathways observed are peculiar of the phosphazene structure and represent a remarkable difference between this class of inorganic polymers and the organic ones.

The more distinctive morphological features which appear in the micrographs of fracture surfaces of polymer blends of poly(butyleneterephthalate)(PBTP) with poly(ethylene-covinylacetate) (EVA), obtained by polymerizing in different conditions dimethylterephthalate (DMT) and 1,4-butanediol (4G) in the presence of EVA (with 4 percent w/w of vinylacetate, VA,), are shown and discussed. The micrograph investigation clearly points out that the adhesion between EVA particles and PBTP matrix increases with the extent of reaction between PBTP and EVA and, for the same time of reaction, by changing from an anchor to a paddle agitator. The mean diameter of the particles is also affected by time of reaction and by the geometry of the agitator, decreasing when the adhesion increases.

The low-temperature dynamic mechanical relaxation of ox Ligamentum nuchae elastin, swollen with water and ethylene glycol at various swelling ratios, has been investigated by means of the Rheovibron DDV II viscoelastometer, at frequencies from 3.5 to 110 Hz. The peak location on the temperature scale, its intensity and activation energy have been studied as a function of solvent content and it has been found that the two solvents influence the dynamic mechanical behavior in a quite different way. A similar interaction of the solvent molecules with the backbone peptide groups is assumed for both solvents, and the differences in the mechanical spectra, which are more marked the higher the solvation degree, are explained in terms of the different size of the protein–solvent relaxing units.

Poly[nitrilo-bis(phenylamino)phosphoranetriyl]s (3a–d) were synthesized and characterized in dilute solution and in the solid state. The synthesis was accomplished by thermal polymerization of of 2,2,4,4,6,6-hexachloro--triazatriphosphorane (1) at 240°C, followed by the reaction of the soluble product, poly(nitrilodichlorophosphoranetriyl) (2), with aniline or its derivatives. Elemental analyses show the presence of some residual chlorine, which, according to the IR spectra, does not seem to be ascribed to an incomplete substitution. All the polymers have intrinsic viscosities in tetrahydrofuran higher than 100ml g−1, and high weight average molecular weights (Mw ≥ 7 · 105). The molecular weight distribution is very broad (Mw/Mn ≥ 13,4) which could suggest either a branched structure or a complex mechanism of polymerization. From DSC measurements the glass transition temperatures are found to be between 70 and 98°C and the related values of specific heat increments, ΔC,p, are between 0,13 J g−1 K−1 and 0,27 J g−1 K−1; the densities are in the order of 1,3 g ml−1.

Linear and branched poly(butyleneisophthalate) samples were synthesized and characterized in terms of the intrinsic viscosity, the molecular weight and the melt viscosity over a wide range of shear rates at 200 °C. An exponent of about 4.6 in the equation relatingη 0 to $$\bar M_w$$ was found for linear samples; this high value is probably due to the high content of cyclic oligomers in low molecular weight samples. Both linear and branched samples exhibited Newtonian behaviour over a rather wide range of shear rates, but for any given melt-viscosity, the branched samples became shear thinning at lower shear rates than the linear ones. Our experimental data were found to fit a previously proposed correlation between the melt viscosity ratio (η 0, b /η 0, 1 ) and a branching indexγ quite well.

A computer simulation model for the latent skin growth inside a hollow fiber polymeric membrane Is developed. The evaporation step is studied using a solvent evaporation Law and a concentration-dependent diffusion coefficient, both derived from data. The solvent concentration profiles inside the fiber are obtained as a function of the evaporation time for different sets of the parameters governing the internal diffusion and the surface evaporation of the solvent. The computer model enables estimation of the role of the parameters in the growth of the asymmetric latent skin.

From an overall analysis of the above results it appears that the rheological properties of plasticized PVC are governed, below about 200 °C, not only by the usual parameters (composition, temperature, molecular weight) but also by the crystallinity of the polymer. Recent estimations of the crystallinity degree in commercial PVC samples vary from 3% (45) to about 10% (46), the actual value depending on the thermal history of the sample. Apparently even small values of the crystallinity degree can affect rather strongly the rheological behaviour of plasticized PVC in the rubbery and flow regions. For the fractions here investigated, the fractional precipitation procedure used to carry out the separation (1) leads probably to a fractionation based not only on molecular weight but also on stereoregularity (47), the result being that the fractions precipitated firstly are more polydisperse and more crystallizable. In dilute solution, they give easily aggregates (1), stable up to relatively high temperatures (2). The dynamic-mechanical data discussed above (fig. 1) as well as the anomalies presented by the rheological results (fig. 7 to 12) can be therefore explained on the basis of a model in which supermolecular structures linked together by crystalline bonds are present in the samples. The marked dependence of PVC mechanical properties on the annealing temperature described recently byAndrews andKazama (48) seems to give considerable support to such a model.

Seventeen different samples of styrene polymers and copolymers were examined in a capillary extrusion rheometer over broad ranges of temperatures and shears. They were injection molded in a spiral mold over the same temperature range. If viscosities at constant shear stress are obtained from the rheometric data, they cannot be correlated with spiral flow results. However, if viscosities are calculated at a proper “high” shear rate, capillary data correlate will with spiral flow results at all temperatures and for all materials.

A study was made of the rheological behavior of 35 samples of nylon 6 polymers of molecular weights varying from 4,000 to 40,000. A capillary extrusion rheometer was employed, and flow curves extending over a large range of shear rates were obtained. The effect of temperature, shear rate, and molecular weight on viscosity is reported, and some peculiarities of the flow behavior of nylon 6 polymers are discussed.

The influence of crystallinity, thermal history, plasticization, copolymerization, and chemical modifications on the dynamic-mechanical properties of poly(vinyl chloride) has been studied by means of a free-oscillation torsional pendulum. It was found that the secondary dispersion maximum is not affected by free volume, crystallinity, or head-to-head microstructure, but is modified by post-chlorination, copolymerization, and plasticization. These results show that the molecular motions responsible for the PVC β peak are not influenced by local differences in the distribution of the Cl atoms along the chain, so that they seem to be largely of a cooperative type.

Accurate measurements of stress relaxation after steady-state flow have been carried out, in the Newtonian flow region, for a polystyrene and a poly(methyl methacrylate) melt, with a cone-and-plate rotational rheometer. From the stress relaxation σ(t) versus t curves the relaxation spectra H were calculated by means of the first approximation equation: . The shear stress–shear rate curves, σ versus were also measured, in large ranges of shear rates, for the same melts, and from these data the relaxation spectra H were obtained by means of equations given by Faucher and Ferry. The Faucher equation, , has been found to give results which compare satisfactorily with those obtained from the first approximation equation. It has been found that the Ferry equation has to be modified for comparable agreement.

Several fractions obtained from a large-scale fractionation and several unfractionated PVC polymers and blends have been processed both as rigid and plasticized compounds. The latter have been studied by stress–strain, creep, and recovery tests. The recoverable character of the creep results show that a relatively stable network must be present in the samples. The crosslink density is little influenced by molecular weight, as shown by the modulus and compliance results. On the contrary, the ultimate tensile properties depend strongly on molecular weight, which is interpreted as evidence that the stability of the crosslinks increases with increasing chain length of the polymers.

A large-scale fractionation of poly(vinyl chloride) (3000 g.) has been carried out by a fractional precipitation method. The molecular weight distribution of some fractions has been controlled by refractionation. The fractions have been also characterized by viscometry, osmometry, light-scattering, and glass transition temperature measurements. The data available in literature concerning the molecular weight characterization of poly(vinyl chloride) fractions have been critically reviewed. It has been shown that most of the viscometric and osmometric data are in good agreement when the molecular weights are below 105. Above 105 the literature discrepancies can be attributed to aggregation.

Very different concentration dependences of the viscosity of polymer solutions are predicted by the free volume treatments of Fujita and Kishimoto and of Kelly and Bueche. This latter is conveniently extended, and it is shown that it can describe a given set of experimental data over a concentration range much larger than the Fujita-Kishimoto equations.

The dynamic-mechanical properties as a function of temperature and the low- and high-speed tensile properties at 23°C have been determined on specimens of conventional suspension-polymerized PVC and of low-temperature-polymerized PVC roll milled and then compression molded at different temperatures. It has been found that the main transition α and the shear modulus above Tα depend on the thermal history and are strongly affected by crystallinity, whereas the dynamic-mechanical spectrum below Tα is not influenced by these parameters. Room-temperature tensile modulus and yield properties are very little affected by processing history and crystallinity. The elongation at break and the fracture energy, on the contrary, increase, at any fixed strain rate, for conventional PVC with milling temperature. The same trend has been found for low-temperature PVC, but the elongation at break-versus-temperature curve is shifted, as a whole, toward higher temperatures by approximately 50°C. Such results are discussed in terms of homogeneity of the specimens, which is controlled by the melting process of the crystallites. Stereoscanning electron micrographs of fracture surfaces appear to substantiate these conclusions.

The differential scanning calorimetric behavior of a series of alpha-amino acid homopolymers and copolymers was investigated in the range of temperature from -80 to +130 degrees C. The following polymers have been examined: poly(N'-carbobenzoxy-L-lysine), poly(O-carbobenzoxy-L-tyrosine), random copolymers of O-carbobenzoxy-L-tyrosine and N'-carbobenzoxy-L-lysine, poly(Ngamma-carbobenzoxy-L-diaminobutyric acid), and poly(N sigma-carbobenzoxy-L-ornithine). Each sample exhibited a prominent glass-like transition between +30 and +60 degrees C with a specific heat increment deltacp of the order of 0.03-0.11 cal/(g degrees C). Endothermal peaks, developed by annealing, have been also been revealed. It is concluded that in poly(alpha-amino acids) side chain motions are capable of undergoing glass-like transition.

Poly(butyleneterephthalate) (PBTP) samples of different molecular weights, both linear and branched, were synthetized by mass polymerization and studied in the molten state with a melt-flow-index apparatus at different temperatures in the range 245–270 °C. In our experimental conditions (\(\dot \gamma \) ≦ 20 s−1) the behaviour of PBTP samples was Newtonian, as reported previously. The flow activation energyEa,0 was found to increase with degree of branching: typicallyEa,0 was about 47 and 63–79 kJ/mol for linear and branched polymers respectively.

The low temperature water relaxation of synthetic polymers containing the amide group is usually attributed to the motion of a water-polymer complex in the amorphous region. According to Puffr and Sebenda the complex is formed by water and amide group in the stoichiometric ratio 1.5. Elastin, an insoluble protein of vessels and ligaments, seems the ideal polymer for testing the model, being practically free of polar side groups and having a glass transition at 200°C, so that its “water peak” can be easily investigated. Calorimetric investigations show that the molar ratio for “bound water-amide group” is indeed 1.5, as predicted by the model. The activation energy of the “water peak” is very close, for the “dry” protein, to that of nylons, but it decreases with increasing hydration to a limit of 9 kcal/mole. The results give additional support to the traditional interpretation of the polyamide beta relaxation and to the Puffr and Sabenda model for the interaction of water and amide groups in synthetic polymers.

The interaction of ox ligamentum nuchae elastin (native, purified and soluble) with four different solvents was investigated by Differential Scanning Calorimetry. The freezable solvent content was determined, at different total solvent contents, from the melting endotherms, and the amount of unfreezable solvent was obtained by extrapolation. For two bifunctional solvents, water and ethylene glycol, the molar ratio “bound solvent/elastin residue” was the same for all the elastin samples investigated, and very close to 1.5.

The Methyl-, Ethyl-, n-Butyl-, n-Octyl- and n-Hexadecyl esters of Octadecanoic (Stearic) acid were synthetized from the corresponding alcohols and pure Stearic acid. Their physical properties were determined and their viscous flow parameters were calculated according to both the Eyring theory and the Batschinski-Hildebrand treatment of viscosity. The results are favourably compared with similar published data on hydrocarbons and with values for low molecular weight esters. Different amounts of each stearate were mixed with Polyvinylchloride at 90°C. The solubility of the esters in the polymer, determined by measuring the crystallizable undissolved ester by means of a Differential Scanning Calorimeter, decreases from 6.5% to 0.3% with increasing chain length of the stearates. The Flory-Huggins interaction parameter χ, obtained from the solubility results, increases from 1.6 (Methylstearate) to 4.2 (n-Hexadecylstearate). The lengthening of the aliphatic chain, which results in a dilution of the ester groups, apparently weakens the PVC-ester interactions. The glass transition temperature T g decreases with increasing diluent content down to a constant value closely related to the ester solubility. The decrease of T g brought about by the PVC-ester interactions is described by the free volume theory, using a single equation for all the systems examined.

The glass transition temperature-composition dependence for the systems PVC-dimethylphthalate and PVC-di-n-propylphthalate has been determined in the whole range of concentration. The glass transition temperature of the mixtures decreases with increasing diluent content and in both cases shows a discontinuity at a critical composition (w1≃0.4). At diluent weight fractions lower than 0.4, the theoretical equations by Kelley-Bueche and CouchmanKarasz satisfactorily fit the experimental results. The marked changes in the glass transition parameters (Tg, Δcp, width of the transition) which occur in the intermediate range of concentration (0.4