A three-phase microstructural model to explain the mechanical relaxations of branched polyethylene: a DSC,WAXD and DMTA combined study

The thermal transitions of well-characterised single-site catalysed polyethylenes having various degrees of short chain branching have been studied by differential scanning calorimetry, X-ray diffraction and dynamic mechanical thermal analysis. A critical discussion based on the results obtained by means of the different techniques is presented. The results suggest that the γ transition is independent of the branching content and degree of crystallinity, pointing towards a sub-glass local relaxation mechanism related to both amorphous and crystalline fractions. The temperature of the β transition, T _β from dynamic mechanical measurements, is in agreement with the glass transition temperature obtained by calorimetry, T _g. Moreover, T _γ, and also T _β are directly related to a change in the thermal expansion coefficient of the amorphous phase observed by X-ray scattering. It is found that the corresponding scattering distance of the amorphous halo depends on crystallinity. In addition, the calorimetric heat capacity values at T _β do not account for the total amorphous fraction determined for each material. The relaxation motions assigned to the amorphous phase glass transition seems to parallel the subsequent melting of the crystalline structure, suggesting a hierarchical motion of different structures as temperature increases. Dynamic mechanical thermal analysis supports these observations, showing a broad transition in the phase angle involving first the relaxation of amorphous phase, then the (presumable) more rigid intermediate phase, and finally the crystalline phase, as the temperature increases.     

Investigation of the rigid amorphous fraction in Nylon-6

A three-phase model, comprising crystalline, mobile amorphous, and rigid amorphous fractions (χ _c, χ _MA, χ _RA, respectively) has been applied in the study of semicrystalline Nylon-6. The samples studied were Nylon-6 alpha phase prepared by subsequent annealing of a parent sample slowly cooled from the melt. The treated samples were annealed at 110°C, then briefly heated to 136°C, then re-annealed at 110°C. Temperature-modulated differential scanning calorimetry (TMDSC) measurements allow the devitrification of the rigid amorphous fraction to be examined. We observe a lower endotherm, termed the ‘annealing’ peak in the non-reversing heat flow after annealing at 110°C. By brief heating above this lower endotherm and immediately quenching in LN₂-cooled glass beads, the glass transition temperature and χ _RA decrease substantially, χ _MA increases, and the annealing peak disappears. The annealing peak corresponds to the point at which partial de-vitrification of the rigid amorphous fraction (RAF) occurs. Re-annealing at 110°C causes the glass transition and χ _RA to increase, and χ _MA to decrease. None of these treatments affected the measured degree of crystallinity, but it cannot be excluded that crystal reorganization or recrystallization may also occur at the annealing peak, contributing to the de-vitrification of the rigid amorphous fraction. Using a combined approach of thermal analysis with wide and small angle X-ray scattering, we analyze the location of the rigid amorphous and mobile amorphous fractions within the context of the Heterogeneous and Homogeneous Stack Models. Results show the homogeneous stack model is the correct one for Nylon-6. The cooperativity length (ξ_A) increases with a decrease of rigid amorphous fraction, or, increase of the mobile amorphous fraction. Devitrification of some of the RAF leads to the broadening of the glass transition region and shift of T _g.     

The Influence of Drawing in Hot Water on The Morphological Properties of Pet Films as Measured by DSC and Modulated DSC

PET films uniaxially drawn in hot water are studied by means of conventional DSC and modulated DSC (MDSC).Glass transition is studied by MDSC which allows to access the glass transition temperature T _g and the variations of ΔC _p=C _p1−C _pg (difference between thermal capacity in the liquid-like and glassy states at T=T _g). Variations of T _g with the water content (which act as plasticizer) and with the drawing (which rigidifies the amorphous phase) are discussed with regard to the structure engaged in these materials. The increments of ΔC _p at T _g are also interpreted using a three phases model and the 'strong-fragile’ glass former liquid concept. We show that the ‘fragility’ of the medium increases due to the conjugated effects of deformation and water sorption as soon as a strain induced crystalline phase is obtained. Then, ‘fragility’ decreases drastically with the occurring rigid amorphous phase. This revised version was published online in August 2006 with corrections to the Cover Date.     

Mechanical properties and transition temperatures of crosslinked-oriented gelatin

 This second part of a systematic study of the properties of crosslinked-oriented gelatin involves the effects of orientation and water content on the glass transition temperature T _g and on the melting behavior. The samples were the same as those in the preceding study, and their transition temperatures were determined by both differential scanning calorimetry and dynamic mechanical thermal analysis. The crosslinked gelatin which had been room-conditioned showed two transition temperatures: the lower one was attributed to T _g of the water-plasticized gelatin, and the higher one was interpreted as T _g of dried gelatin superimposed by melting. A rather unusual situation arose because of the fact that the T _g and melting temperatures T _m (217 and 230 ^°C, respectively) are so similar. Using water as plasticizer not only decreases T _g but produces imperfect crystallites which melt below the T _g of the system. The presence of the amorphous phase in the glassy state would presumably make it essentially impossible to define a melting point or crystallization temperature in the normal manner, as an equilibrium between crystalline and amorphous phases. Received: 8 October 1996 Accepted: 2 November 1995     

Glass transition temperature and thermal decomposition of cellulose powder

Cellulose powder and cellulose pellets obtained by pressing the microcrystalline powder were studied using differential scanning calorimetry (DSC), differential thermal analysis (DTA), and thermal gravimetry (TG). The TG method enabled the assessment of water content in the investigated samples. The glass phase transition in cellulose was studied using the DSC method, both in heating and cooling runs, in a wide temperature range from −100 to 180 °C. It is shown that the DSC cooling runs are more suitable for the glass phase transition visualisation than the heating runs. The discrepancy between glass phase transition temperature T _g found using DSC and predictions by Kaelbe’s approach are observed for “dry” (7 and 5.3% water content) cellulose. This could be explained by strong interactions between cellulose chains appearing when the water concentration decreases. The T _g measurements vs. moisture content may be used for cellulose crystallinity index determination.     

Dielectric relaxation of poly(phenylene sulfide) containing a fraction of rigid amorphous phase

The dielectric relaxation behavior of poly(phenylene sulfide), PPS, has been investigated from room temperature to 180°C. This study was undertaken to examine the mobility of the amorphous phase through the glass transition region, to determine the contribution that rigid amorphous phase material makes to the relaxation process. Semicrystalline samples contain a fraction of the rigid amorphous phase, which was determined from the heat capacity increment at the glass transition, using degree of crystallinity determined from x-ray scattering. In the dielectric experiment, we measured the temperature and frequency dependence of the real and imaginary parts of the dielectric function. ε″ vs. ε′ was used to determine the dielectric relaxation intensity, δε = ε_s–ε∞, at temperatures above the glass transition. For amorphous PPS, δε decreases as temperature increases, while for all semicrystalline PPS, δε increases with temperature. The ratio of semicrystalline intensity to amorphous intensity determines the total fraction of dipoles which are already relaxed at a given temperature. Results indicate that more and more rigid amorphous phase material relaxes as the temperature is increased. This provides the first evidence that rigid amorphous phase material in PPS contains chains that possess different levels of molecular mobility. Finally, to the temperature of the loss peak maximum, at a given frequency, we assign the value of the dielectric T_g. For both melt and cold crystallization, the dielectric T_g systematically decreases as the crystallization temperature increases, and as the fraction of rigid amorphous phase decreases.     

Effects of thermal treatment of biaxially oriented poly(ethylene terephthalate)

Two distinguishable effects of thermal exposure of biaxially oriented poly(ethylene terephthalate) (PET) have been observed in the temperature range from room temperature to 140°C. Upon heating above the glass transition temperature T_g of the film an irreversible shrinkage of a few percent occurred with a concomitant decrease in the rate of creep. Some loss of orientation in the noncrystalline phase with an attendant slight increase in density is believed to be responsible. Since the film was anisotropic in its plane, different amounts and rates of shrinkage were observed along with differing thermal expansion coefficients in various directions relative to the primary optic axis. Upon cooling the 50% crystalline PET from above T_g to lower temperatures, reversible “physical aging” was observed. Creep rates were found to decrease with the residence time below T_g. As with purely amorphous polymers, the effects of the aging are removed by heating the specimen above T_g where the density of the amorphous phase achieves equilibrium values.     

The temperature dependence of the local free volume in polyethylene and polytetrafluoroethylene: A positron lifetime study

Positron lifetime measurements, performed in the temperature range 80–300 K, are reported for polyethylene (PE) and polytetrafluoroethylene (PTFE). The lifetime spectra have been analyzed using the data processing routines LIFSPECFIT and MELT. Two long-lived components appear, which are attributed to pick-off annihilation of ortho-positronium in crystalline regions and at holes in the amorphous phase. The ortho-positronium lifetimes, τ₃ and τ₄, are used to estimate the crystalline packing density and the size of local free volumes in the crystalline and amorphous phases. The interstitial free volume in the crystals exhibits a weak linear increase with the temperature which is attributed to thermal expansion of the crystal unit cell. In the amorphous phase, the hole volume varies between 0.053 and 0.188 nm³ (PE) and between 0.152 and 0.372 nm³ (PTFE). Its temperature variation may be fitted by two straight lines, the intersection of which is used to estimate a glass transition temperature of T_g = 195 K for both PE and PTFE. The slopes of the free volume in the glassy and crystalline phases with the temperature correlate well with each other. The coefficients of thermal expansion of the hole volume are compared with the macroscopic volume change below and above the glass transition. From this comparison a fractional hole volume at T_g of 4.5 (PE) and 5.7% (PTFE) and a number of 0.73 (PE) and 0.36 (PTFE) × 10²⁷ holes/m³ is estimated. Finally, it is found that the intensity of o-Ps annihilation in crystals shows a different temperature dependence to that in the amorphous phase. © 1998 John Wiley & Sons, Inc. J Polym Sci B: Polym Phys 36: 1513–1528, 1998     

Effect of diphenylsiloxane unit content on relaxation behavior of poly(dimethylsiloxane‐co‐diphenylsiloxane)

The relaxation behaviors of poly(dimethylsiloxane‐co‐diphenylsiloxane)s with different compositions were investigated using dynamic mechanical analysis (DMA) and differential scanning calorimetry (DSC). It is indicated that the content of Ph₂SiO unit, which is closely associated with crystallinity of polysiloxane, has a remarkable influence on its relaxation behavior. Two‐phase (crystalline and amorphous phase) structure in the semicrystalline polysiloxane of the present system can be determined for discussing relaxation behavior. An increase in relaxation strength can be reasoned to a cooperative effect of decrease in fraction of crystalline phase and increase in friction between molecular chains. And enhancements in glass transition temperature (T_g) and effective activation energy for glass transition (E_a(eff)) were ascribed more to the stiffness imposed by Ph₂SiO unit than decrease in fraction of crystalline phase. © 2008 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 46: 1652–1659, 2008     

Succinic or glutaric anhydride modified linear unsaturated (epoxy) polyesters

A series of N-alkyl-N-alkyl′-pyrrolidinium-bis(trifluoromethanesulfonyl) imide (TFSI⁻) room temperature ionic liquids (RTILs) has been investigated by means of thermogravimetric analysis (TG), differential scanning calorimetry, FT-IR spectroscopy, and X-ray diffraction analysis. These compounds exhibit a thermal stability up to 548–573 K. The mass loss starting temperature, T _ml, falls in a narrow range of temperatures: 578–594 K. FT-IR spectra, performed before and after 24 h isothermal experiments at 553 and 573 K, have confirmed their great thermal stability. Below the ambient temperature, these compounds exhibit a complex behavior. N-methyl-N-propyl-pyrrolidinium-TFSI is the sole liquid which crystallizes without forming any amorphous phase even after quenching in liquid nitrogen. Its crystalline phase has a melting point, T _m, of 283 ± 1 K. When the amorphous solid is heated, the N-butyl-N-ethyl-pyrrolidinium-TFSI presents a glass transition temperature, T _g, at 186 K followed by a cold crystallization, T _cc, at 225 K, and a final T _m at 262 K. The N-butyl-N-methyl-pyrrolidinium-TFSI exhibits a T _g between 186 and 181 K, its cold crystallization leading to two different solid phases. Solid phase I has a melting point T _I,m = 252 K and phase II, T _II,m = 262 K. When the amorphous phase is obtained at a cooling rate of 10 K/min, its T _cc is 204 K, and a metastable solid phase (III) is obtained which transforms into the phase II at 226 K. However, when the sample is quenched, the amorphous phase transforms into phase II at T _cc = 217 K and phase I at 239 K. P₁₅-TFSI exhibits the most complicated pattern as, on cooling, it leads to both a crystallized phase at 237 K and an amorphous phase at 191 K. On heating, after a T _g at 186 K and a T _cc at 217 K, two solid–solid phase transitions are observed at 239 K and 270 K, the final T _m being 279 K.     

Plastic deformation of poly(tetramethylene oxide). II. Molecular orientation as measured by x-ray diffraction and NMR measurements

The orientation of the crystalline and amorphous phases in uniaxially drawn samples of polytetramethylene oxide has been studied by x-ray and NMR methods. Pole figure measurements give the orientation of the crystalline phase. Its dependence on the draw ratio does not obey the pseudoaffine model. The crystalline fraction is derived from NMR measurements at temperatures between T_g and T_m, where the free induction decay (FID) of the unoriented sample can be analyzed in terms of a rigid (crystalline), a constrained, and an amorphous phase. Low temperature (T < T_g) measurements of the NMR second-moment anisotropy in protonated and deuterated samples give a mean orientation of the chains higher than that corresponding to the crystalline phase. The lack of anisotropy in the tail of the FID at temperatures between T_g and T_m indicates no appreciable anisotropy in the truly amorphous interlamellar phase. From this and from the ratio of the P₄/P₂ orientations, factors which obey the “most probable distribution,” it is concluded that the amorphous orientation is due to the layer of constrained chains at the surface of the lamellae.     

Phase transitions in chlorobenzene-cis-decalin mixtures

The dielectric constant ?' and loss tangent tan δ of chlorobenzene-cis-decalin mixtures have been measured in the temperature range 77 K to 330 K and frequency range 0.1 to 100 kHz. On cooling, ?' increases with decreasing temperature upto about 135 K, after which it drops rapidly with decreasing T followed by a slow decrease. This indicates that the liquid mixture goes to an amorphous phase which transforms to a glass phase of restricted dipole rotation below T_g; however, the peak in ?' is due to relaxation in the amorphous phase (α relaxation) and does not give an exact T_g. On heating, the behaviour of the cooling curve is retraced upto 160 K, after which ?' drops suddenly to a value lower than that at 77 K in the glass phase. This indicates the transition to a crystalline phase in which dipole rotational freedom is completely lost. The crystalline phase changes to a eutectic liquid phase of high ?' at a temperature (200 K) lower than the melting point of chlorobenzene and cis-decalin. Dielectric dispersion is observed only in the glass and amorphous phases. The dielectric relaxation time is independent of the concentration of chlorobenzene.     

Determination of dual glass transition temperatures of a PC/ABS blend using two TMA modes

An inherent challenge with polymer blends is the difficulty in resolving the glass transition, T _g, for the smaller mass fraction component. The objective of this work was to determine the practical scanning conditions for identifying the dual T _g’s for a 75:25 polycarbonate/acrylonitrile-butadiene-styrene (PC/ABS) blend using a thermomechanical analyzer (TMA). Scanning rates up to 20°C min⁻¹ using dilatometer and expansion modes were studied. Heating and cooling rates were found to affect both T _g values but the effects were not simple relationships. T _g values could either increase or decrease depending on the scanning rate applied. Higher rates resulted in large thermal lags which opened the accuracy of measurements to question. Generally, higher rates tended to display only one T _g but the duality of T _g’s can be detected with scanning rates between 0.5 and 5°C min⁻¹ for both modes.     

Water sorption and glass transition of freeze-dried camu-camu (myrciaria dubia (H.B.K.) Mc Vaugh) pulp

Differential scanning calorimetry (DSC) was used to determine phase transitions of freeze-dried camu-camu pulp in a wide range of moisture content. Samples were equilibrated at 25°C over saturated salt solutions in order to obtain water activities (a_w) between 0.11–0.90. Samples with a_w>0.90 were obtained by direct water addition. At the low and intermediate moisture content range, Gordon–Taylor model was able to predict the plasticizing effect of water. In samples, with a_w>0.90, the glass transition curve exhibited a discontinuity and T^’_g was practically constant (–58.8°C), representing the glass transition temperature of the maximally concentrated phase(T_g ).     

The rigid amorphous fraction in semicrystalline macromolecules

The first experimental evidence of the existence of the rigid amorphous fraction (RAF) was reported by Menczel and Wunderlich for several semicrystalline polymers. It was observed that the hysteresis peak at the glass transition was absent when these polymers were heated much faster than they had previously been cooled. In the glass transition behavior of poly(ethylene terephthalate) (PET), the hysteresis peak gradually disappeared as the crystallinity increased. At the same time, it was noted that the ΔC _p of higher crystallinity PET samples was much smaller than could be expected on the basis of the crystallinity calculated from the heat of fusion. It was also observed that this behavior was not unique to PET only, but is characteristic of most semicrystalline polymers: the sum of the crystallinity calculated from the heat of fusion and the amorphous content calculated from the ΔC _p at the glass transition is much less than 100% (a typical difference is ~20–30%). This 20–30% difference was attributed to the existence of the “RAF”. The presence of the RAF also affected the unfreezing behavior of the “mobile (or traditional) amorphous fraction.” As a consequence, the phenomenon of the enthalpy relaxation diminished with increasing rigid amorphous content. It was suggested that the disappearance of the enthalpy relaxation was caused by the disappearance or drastic decrease of the time dependence of the glass transition. To check the validity of this suggestion, the glass transition had to be also measured on cooling in order to overlay it on the DSC curves measured on heating. However, before this overlaying work could be accomplished, the exact temperatures on cooling had to be determined since the temperature of the DSC instruments that time could not be calibrated on cooling using the usual low molecular weight standards due to the common phenomenon of supercooling. Therefore, a temperature calibration method needed to be developed for cooling DSC experiments utilizing high purity liquid crystals using the isotropic → nematic, the isotropic → cholesteric, and other liquid crystal → liquid crystal transitions. After the cooling calibration was accomplished, the cooling glass transition experiments indicated that the glass transition in semicrystalline polymers is not completely time independent, because its width depends on the ramp rate. However, it was shown that the time dependence is drastically reduced, and the midpoint of the glass transition seems to be constant which can explain the absence of the enthalpy relaxation. The work presented here has led to a number of studies showing the universality of the rigid amorphous phase for semicrystalline polymers as well as an ASTM standard for DSC cooling calibration.     

Difference in Glass Transition Behavior Between Semi Crystalline and Amorphous poly(lactic acid) Thin Films

Summary: Semi crystalline and amorphous poly(lactic acid) (PLA) thin films exhibit different glass transition temperature and behaviour, as revealed by ellipsometry. For semi-crystalline poly(L-lactic acid) (PLLA) thin film (with crystalline content between 40 and 60%), the glass transition temperature (T_g) is found to decrease below a film thickness of 50 nm. This depression was interpreted in term of disentenglement effect which is likely to occur upon confining the amorphous PLA phase near a non interacting surface. New results performed on non completed films, i.e. isolated objects, also reveal the lower transition temperature, thus underlying the importance of the entanglement state of the polymer chains on their mobility. For amorphous PLA thin film, obtained from the L and D copolymer, two distinct glass transitions were observed, with the highest Tg attributed to the presence of some nano-phase domains, formed by a possible cooperation of the D and L blocks to form stereocomplexes sequences, within the film. Furthermore, if these T_g remained constant as film thicknesses decrease down to 50 nm, they were also found to slightly decrease for isolated objects, thus supporting the importance of the entanglement hypothesis on the glass transition.     

On the scaling law of the evolution of lap-shear strength with healing temperature at amorphous polymer−polymer interfaces and surface glass transition

The evolution of lap-shear strength (σ) with healing temperature T _h at symmetric and asymmetric amorphous polymer−polymer interfaces formed of the samples with vitrified bulk has been investigated. It has been found that the square root of the lap-shear strength behaves with respect to healing temperature as σ ^1/2 ~ T _h both at symmetric and asymmetric interfaces. Basing on this scaling law between σ and T _h, the values of the surface glass transition temperature ( T_g^surface ) \left( {T_{\rm{g}}^{\rm{surface}}} \right) have been estimated for a number of amorphous polymers by the extrapolation of the experimental curves σ ^1/2 ~ T _h for symmetric polymer−polymer interfaces and, in some cases, for asymmetric, both compatible and incompatible, polymer−polymer interfaces, to zero strength. A significant reduction in surface glass transition temperature T_g^surface T_{\rm{g}}^{\rm{surface}} with respect to the glass transition temperature of the polymer bulk ( T_g^bulk ) \left( {T_{\rm{g}}^{\rm{bulk}}} \right) , reported earlier, has been confirmed by the use of the new proposed approach. The quasi-equilibrium surface glass transition temperature T_g^surface T_{\rm{g}}^{\rm{surface}} of amorphous polystyrene (PS) has been predicted in the framework of an Arrhenius approach using the plot “logarithm of healing time − reciprocal surface glass transition temperature T_g^surface￠￠ T_{\rm{g}}^{\rm{surface}}\prime \prime and the activation energy of the surface alpha-relaxation of PS has been calculated.     

The extent of the glass transition from molecular simulation revealing an overcrank effect

A deep understanding of the transition between rubber and amorphous state characterized by a glass transition temperature, T_g, is still a source of discussions. In this work, we highlight the role of molecular simulation in revealing explicitly this temperature dependent behavior. By reporting the specific volume, the thermal expansion coefficient and the heat capacity versus the temperature, we actually show that the glass transition domain extends to a greater range of temperature, compared with experiments. This significant enlargement width is due to the fast cooling rate, and actually explains the difficulty to locate T_g. This result is the manifestation of an overcranking effect used by high‐speed cameras to reveal slow‐motion. Accordingly, atomistic simulation offers the significant opportunity to show that the transition from the rubber state to the glass phase should be detailed in terms of the degrees of freedom freeze. © 2017 Wiley Periodicals, Inc.     

How did we find the rigid amorphous phase?

The first experimental evidence of the existence of the rigid amorphous phase was reported by Menczel and Wunderlich [1]: when trying to clarify the glass transition characteristics of the first main chain liquid crystalline polymers [poly(ethylene terephthalate-co-p-oxybenzoate) with 60 and 80 mol% ethylene terephthalate units] [2], the absence of the hysteresis peak at the lower temperature glass transition became evident when the sample of this copolymer was heated much faster than it had previously been cooled. Since this glass transition involved the ethylene terephthalate-rich segments of the copolymer, we searched for the source of the absence of the hysteresis peak in PET. There, the gradual disappearance of the hysteresis peak with increasing crystallinity was confirmed [1]. At the same time it was noted that the higher crystallinity samples showed a much smaller ΔC _p than could be expected on the basis of the crystallinity calculated from the heat of fusion (provided that the crystallinity concept works). Later it was confirmed that the hysteresis peak is also missing at the glass transition of nematic glasses of polymers. When checking other semicrystalline polymers, the sum of the amorphous content calculated from the ΔC _p at the glass transition, and the crystallinity calculated from the heat of fusion was far from 100% for a number of semicrystalline polymers. For most of these polymers, the sum of the amorphous content and the crystalline fraction was 0.7, meaning that ca. 30% rigid amorphous fraction was present in these samples after a cooling at 0.5 K min⁻¹ rate. Thus, the presence of the rigid amorphous phase was confirmed in five semicrystalline polymers: PET, Nylon 6, PVF, Nylon 66 and polycaprolactone [1]. Somewhat later poly(butylene terephthalate) and bisphenol-A polycarbonate [3] were added to this list.     

10.1007/s10973-006-8228-4

The high-resolution transmission electron microscopy HRTEM study of the atomic scale mechanism of crystal structure organization within the amorphous polymeric structure of the model multicomponent glass TiO₂–MgO–Al₂O₃–SiO₂– in the glass transformation temperature range has been undertaken. In the glass transition (T _g) temperature range, glass transforms from the solid of rigid amorphous structure into viscoelastic state of weakened chemical bonds. This is an example of nuclei formation and crystal growth in the polymeric amorphous structure of low atomic scale homogeneity due to middle range ordering. It has been demonstrated that in this case crystal structure formation proceeds by successive displacement and local ordering of atoms in the amorphous structure, like disorder-order transformation in crystalline solid bodies. As the consequence in the crystallization by parent structure reorganization mechanism, traditional model of glass crystallization as well as kinetic models of reactions in solid bodies according Avrami or others, are worthy to be revised.