Search

Enthalpy recovery for a single polystyrene ultrathin film of 20 nm thickness is studied using Flash DSC over an extended time and temperature range. Results are compared to a bulk sample of the same polystyrene using a similar experimental protocol and analysis procedure in an effort to determine the effects of nanoconfinement. Examined is the cooling rate dependence of the glass transition temperature (Tg) of unaged films which informs the initial fictive temperature (Tfo) and thus the jump size (Tfo - Ta) for a given aging temperature (Ta). Isothermal enthalpy recovery is investigated as a function of both Ta for various cooling rates and as a function of jump size at constant Ta. The apparent activation energies at Tg and along the glassy line are determined and compared, as is the enthalpy recovery aging rate. Although the apparent activation energy along the glass line is the same within experimental error as the bulk, the aging rate is found to be slightly faster in the ultrathin film. Increasing the cooling rate prior to aging increases the aging rate. The results are discussed in the context of the two competing factors which influence the aging rate, namely, the driving force and the molecular mobility. The driving force for aging is dictated by the jump size or cooling rate, i.e., the value of Tfo - Ta. The mobility, on the other hand, is dictated by the relaxation time at the aging temperature, which increases during aging from the value on the initial glass line to that at equilibrium. The initial mobility in the glassy state is dictated by the jump size, being related to Tfo - Ta and the temperature dependence of the relaxation time along the glass line, whereas the mobility at equilibrium is dictated by Ta, Tg, and the temperature dependence and breadth of the equilibrium relaxation time.

Ionic liquids display outstanding properties and have a wide range of potential uses. Their unique ionic and organic nature also identify them as good glass-forming materials. In this work, a series of imidazolium-based ionic liquids with varying functionalities from aliphatic to aromatic groups and a fixed bis[(trifluoromethane)sulfonyl]amide ([NTf2]−) anion are characterized using a commercial rapid scanning nanocalorimetry instrument, the Mettler Toledo Flash DSC. The limiting fictive temperature Tf′, which is equivalent to the glass transition temperature Tg, is measured on heating as a function of cooling rate. The dynamic fragility is obtained for the series of ionic liquids, and using this data along with a compilation of data from the literature for a total of 50 ionic liquids reveals that aromatic structure increases fragility more than expected simply based on the change in Tg. Our results provide an approach to control fragility of ionic liquids through chemical modifications to the side group.

The use of nanoreactors to confine monomers and synthesize polymers results in changes in the reaction kinetics and polymer properties making nanoconfinement a potential tool for manipulating and engineering polymer properties. In this perspective, we cover conventional nanoconfinement hosts, nanopore-confined free radical, step-growth, and ring-opening polymerizations, and changes in molecular weight, tacticity, glass transition temperature (Tg), thermal stability, and electrical properties. We use examples from research in our laboratory, as well as comparisons of the work in the literature, to illustrate the competing forces that drive these changes, namely molecular layering or orientation at the nanopore surface, decreased molecular and segmental diffusion, and catalytic or inhibitory effects caused by chemical moieties on the native or surface-functionalized nanopore surface. The majority of nanoconfined polymerizations are found to be accelerated, and in the case of free radical polymerizations to generally yield higher molecular weights and higher isotacticity. Tgs for the nanoconfined polymers tend to increase if strong interactions exist between the polymer and the confinement surface, but depressions are observed for confined polycyanurates; the importance of removing unreacted monomer and comparing results to the bulk material of same molecular weight and structure is emphasized. Examples are also provided of enhanced thermal stability and conductivity of polymers synthesized under nanoconfinement.

The structural recovery of a single polystyrene thin film is studied using Flash differential scanning calorimetry (DSC) for aging temperatures ranging from 50.5 to 100.5 °C and for aging times of up to 18,000 s (300 min). A high fictive temperature glass with T f ′ = 117.0 ± 0.5 °C, obtained after cooling at 1000 K/s, is employed in this study. Structural recovery is investigated by monitoring the evolution of the fictive temperature, T f , which initially remains unchanged and then decreases smoothly and approximately linearly with logarithmic aging time. Equilibrium is reached when T f = T a at the highest aging temperatures. The length of the initial plateau is longer for the lower aging temperatures and is related to the increase in the relaxation time with decreasing temperature along the glass line, increasing approximately one decade for every 21 K. This temperature dependence yields a normalized apparent activation energy (E a /R) of approximately 13 kK for the mobility along the glass line. On the other hand, the apparent activation energy for the mobility along the liquid line is 102 kK.

The effect of nanoconfinement on the free radical polymerization of ethyl methacrylate (EMA) and n-butyl-methacrylate (BMA) with di-tert-butyl peroxide (DtBP) initiator is investigated over a wide temperature range from 80 to 190 °C using differential scanning calorimetry. The effective rates are similar for the two bulk monomers although the BMA reacts approximately 11% faster at 95 °C. For nanoconfined cases, the initial reaction rate for monomer confined in the pores of controlled pore glass is enhanced, with larger effects observed in native pores compared to pores in which the native silanol was converted to trimethyl silyl. The onset of autoacceleration also occurs earlier under nanoconfinement, with decreases in both the conversion and the time required to reach autoacceleration, x g e l and t g e l , respectively, and larger changes for the native pores. The induction time follows Arrhenius behavior and increases under nanoconfinement. At polymerization temperatures above 160 °C, depropagation becomes important as the ceiling temperature is approached and seems to be more pronounced under nanoconfinement than in the bulk.

The thermal properties of bulk and nanoconfined HMX were studied using differential scanning calorimetry in dynamic scanning mode at rates ranging from 0.3 to 100 °C/min. At the slowest heating rates, decomposition occurs in the solid phase; at intermediate heating rates, it starts in the solid phase, melts, and finishes with a faster rate of reaction in the liquid state; and at the highest heating rates, the decomposition is entirely in the liquid phase. The activation energy decreases with conversion and is highest for 12 nm-diameter pores. The decomposition reaction is accelerated for nanoconfined HMX compared to the bulk with the onset of decomposition decreased by 1–5 °C in 50 nm-diameter pores and by 4–11 °C in 12 nm-diameter pores. The kinetics of decomposition is well described by a first-order autocatalytic reaction model with the reaction rate constants increasing with nanoconfinement. In addition, the reaction rate constant is one order of magnitude higher in the melt state than in the solid state.

The effect of nanoconfinement on the equilibrium free radical polymerization of methyl methacrylate (MMA) is investigated using differential scanning calorimetry. The ceiling temperature is shifted to lower temperatures in 13 nm diameter pores, with pore surface chemistry showing no significant effect. The results indicate that the change in the entropy of propagation decreases in nanopores due to confinement effects (i.e. ΔSp,conf is a more negative value than ΔSp,bulk). The change in the entropy of propagation is independent of temperature for the bulk equilibrium polymerization, whereas the change in the entropy of propagation in nanopores becomes less negative and more bulk-like with increasing polymerization temperature presumably due to the lower molecular weight chains produced at high temperature. The data suggest that our system is one of weak confinement with chain entropy scaling with molecular weight to the 1.1 power (i.e. ∼ N1.1).

The fictive temperature (Tf) was defined by Tool in the 1940s as a measure of glassy structure. Tf is generally measured on heating and can be calculated from the enthalpy overshoot in calorimetric studies using a method developed by Moynihan. Prior work has demonstrated that the limiting fictive temperature (Tf′) is similar to Tg (measured on cooling) and depends on the cooling rate in a manner consistent with the Williams–Landel–Ferry (WLF) relationship. Theoretically, the limiting fictive temperature should not depend on heating rate, but this has been experimentally verified only for a very limited range of heating rates. Here, rapid-scanning chip calorimetry and conventional differential scanning calorimetry (DSC) are combined to investigate Tf′ for polystyrene over a broad range of heating rates ranging from 0.017 to 3000 K/s after cooling at different rates. The results show that Tf′ depends on cooling rate following the WLF equation. On the other hand, Tf′ is not a function of heating rate, consistent with theoretical predictions, in spite of the change in the magnitude and placement of the enthalpy overshoot.

The effect of nanoconfinement on the free radical polymerization of methyl methacrylate (MMA) is investigated using differential scanning calorimetry, gel permeation chromatography, and 1 H nuclear magnetic resonance. Both hydrophobic and hydrophilic 13 nm-diameter controlled pore glasses (CPG) are used for polymerization under nanoconfinement. The number-average and weight-average molecular weights increase under nanoconfinement because the onset of autoacceleration shifts to shorter times, whereas the polydispersity index at full conversion decreases relative to the bulk value. The tacticity changes from syndiotactic-rich triads for the bulk PMMA to a higher percentage of isotactic-rich triads in hydrophilic pores; the data are described by the first-order Markov model. In addition to the changes in molecular weight and tacticity, the glass transition temperature increases for both pore surfaces compared with the bulk, but the increase in hydrophilic pores is more pronounced.

The reaction kinetics of cyclopentadiene (CPD) dimerization is studied using differential scanning calorimetry (DSC). Both dynamic and isothermal reactions are investigated. The experimental data were fit to a second order autocatalyzed reaction, taking into account the effect of elevated pressure on concentration which results from heating the easily volatilized reactant mixture in a sealed DSC pan. The rate constants and associated activation energies are obtained from the kinetic model, as well as from a model-free method. The results from the modeling with and without pressure effects are compared with one another and literature data.

The kinetics associated with the glass transition is investigated using differential scanning calorimetry (DSC) for germanium selenide glasses with Ge content ranging from 0 to 30 at.% and average coordination numbers ranging from 2.0 to 2.6. As Ge content increases, the glass transition region broadens and the step change in heat capacity at T g decreases. As a result of physical aging, enthalpy overshoots are observed in DSC heating scans and the corresponding change in enthalpy can be calculated as a function of aging time. The change in enthalpy linearly increases with the logarithm of aging time and then levels off at an equilibrium value that increases with decreasing aging temperature. The time required to reach equilibrium increases with decreasing aging temperature and, at a given distance from T g , the time increases with decreasing germanium content. The results indicate that all samples show expected physical aging behavior, and no evidence is found for a Boolchand intermediate phase characterized by high stability and absence of physical aging.

The reaction kinetics of an epoxide-functionalized regenerated cellulose membrane with excess aliphatic amine are examined using differential scanning calorimetry. Both isothermal and dynamic experiments are analyzed. The reaction is found to follow first-order autocatalytic kinetics. The activation energy based on five different methods ranges from 77.0 to 79.7 kJ/mol, indicating that the apparent activation energy for the heterogeneous epoxy/aliphatic amine reaction is considerably higher than the values reported in the literature for the homogeneous reaction.

An expression for the relaxation time, τ, which applies both in the liquid and in the glassy state, is proposed and coupled with the kinetic equation for the volume relaxation of the KAHR (Kovacs, Aklonis, Hutchinson, and Ramos) phenomenological theory. The expression for τ reproduces the WLF (Williams, Landel, and Ferry) behavior above the glass transition temperature, contains the thermodynamic scaling proposed by Casalini and Roland, gives finite values of the equilibrium relaxation time for any value of temperature and pressure, and is assembled in way that the relationship between the logarithmic relaxation time and the internal order parameter is strongly non-linear even when the system is very close to the equilibrium. The resulting model contains ten parameters, and five of these are treated as fitting parameters. With one set of model parameters, the model is able to describe quantitatively the isobaric specific volume response on cooling at various rates and as a function of pressure, the overshoot response on heating after cooling at different rates, the pressure dependence of Tg, the asymmetry of response that characterizes volume relaxation after temperature up and down jumps, and the tau-effective paradox and associated expansion gap. The model also reasonably predicts the overshoot response observed after the two-step memory experiment.

Free radical polymerization of methyl methacrylate in nanopores has been shown to result in a decrease in the time for the onset of autoacceleration. In this work, we simplify our previous kinetic model of nanoconfined methyl methacrylate polymerization, which was based on the work of Verros and coworkers, and incorporate diffusion effects into the model using the Doolittle free volume theory. The simplified model well describes the experimental calorimetric conversion versus time data for isothermal bulk methyl methacrylate polymerization, capturing autoacceleration and the dependence of the limiting conversion on temperature. In order to model the reaction in nanopores, we assume that the diffusion coefficient scales with molecular size to the −3 power and with nanopore diameter to the 1.3 power. Experimental calorimetric conversion versus time data for polymerization in hydrophobic nanopores are well captured by the model, including the decrease in the time to reach autoacceleration with decreasing pore size. The scaling assumed is consistent with that predicted using molecular simulations for good solvent conditions by Avramova and Milchev and by Cui, Ding, and Chen. According to the fit of the experimental data, chain diffusivity is 20–50% of the bulk value in 13 nm-diameter pores.

The effect of nanoconfinement on the rate of isothermal polymerization of methyl methacrylate (MMA) polymerization is investigated using differential scanning calorimetry. Controlled pore glass (CPG) with pore diameters of 13, 50, and 111 nm are used for the confinement of the reaction. Both hydrophilic and hydrophobic pore surfaces are studied. The effective reaction rate and the apparent activation energy at low conversions, prior to autoacceleration, are unchanged in hydrophobic pores. On the other hand, in hydrophilic pores, the reaction rate increases by as much as a factor of 8 in the smallest 13 nm hydrophilic pores, and the effective activation energy decreases. For both pore surfaces, the time required to reach autoacceleration decreases with decreasing pore size, with the effect much more pronounced in the hydrophilic pores. The results are consistent with a model of nanoconfined free radical polymerization which accounts for suppressed termination due to a decrease in the diffusivity of nanoconfined chains.

Nanoconfinement of methyl methacrylate free radical polymerization is known to impact the molecular weight and molecular weight distribution of the polymer produced, with results in the literature generally indicating an increase in molecular weight and a concomitant decrease in polydispersity index. In the present work, the mathematical model described by Verros et al. (2005) for free radical bulk polymerization of methyl methacrylate is extended to account for polymerization in nanopores. The model of Verros et al. (2005) incorporates diffusion effects and is capable of describing the conversion and the number- and weight-average molecular weights of the resulting poly(methyl methacrylate) as a function of polymerization time and process conditions. The model is extended by incorporating the effect of nanoconfinement on diffusivity using the scaling reported in the literature. The calculations indicate that nanoconfinement will lead to higher molecular weights and lower polydispersity, and the gel effect will occur earlier. The results are compared to experimental work and implications discussed.

The viscoelastic properties of poly(α-methyl styrene), its hexamer, and their athermal blends at various concentrations are studied. Master curves for the dynamic shear responses, G′ and G″, are successfully constructed for both the pure materials and the blends, indicating the validity of the time–temperature superposition principle for these systems. The temperature dependence of the shift factor follows the Vogel–Fulcher behavior over the temperature range studied, and the temperature dependence is slightly weaker for the blends. The rubbery plateau modulus scales with the polymer concentration as G N 0 ∝ ϕ 2 . 04 ± 0 . 32 ; the terminal relaxation time scales with the polymer concentration as τ d ∝ ϕ 1 . 35 ± 0 . 33 . The shape of the segmental dispersion appears unchanged by concentration, which differs from our calorimetric studies where mixtures show obviously temperature-broadened glass transitions and depressed enthalpy overshoots. The TNM (Tool-Narayanaswamy-Moynihan) model indicates that the change in the temperature dependence is not sufficient to account for the observed calorimetric broadening. We conclude that the temperature broadening of the glass transition for our blends is not due to a broadening of the dynamic spectrum or to changes in its temperature dependence. The possibility that the broadening is due to changes in the non-linearity parameter x in the TNM model is also considered. While the broadening could be due to a decreasing value of x, we found that this same decrease would lead to increasing enthalpy overshoots on heating, contrary to the experimental observations. The combination of the calorimetric results with the rheological measurements further indicates that the fundamental basis of the TNM-type of model of structural kinetics in glasses is potentially wrong.

Nanoscale constraint is known to have a significant impact on the thermal properties of materials. In this work, differential scanning calorimetry (DSC) is used to investigate the depression in the glass transition temperature ( T g ) and the reactivity of a monofunctional and of a difunctional cyanate ester cured under nanoscale constraint. Both reactants undergo trimerization, but the former forms a small molecular-weight compound, whereas the latter forms a polycyanurate network material. A T g depresssion is observed for both the reactants and their products; the magnitude of the depression seems to be related to the size of the molecule being confined relative to the confinement size. The trimerization reaction is accelerated relative to the bulk when the reactants are confined in nanopores. This is clearly observed by a shift in the reaction exotherms to lower temperatures for dynamic temperature scans. Quantification of the acceleration is accomplished by converting the dynamic temperature scan data to conversion versus time data assuming constant activation energy. The results are consistent with acceleration factors obtained from isothermal cure studies, but the dynamic data is considerably easier to obtain.

The absolute liquid heat capacity of poly(α-methyl styrene) was determined at temperatures far below T g and T K in previous work by use of pentamer/polymer athermal mixtures. Here the data is compared to data compiled by Wunderlich and coworkers from 0 K to above T g in order to obtain the absolute entropy for the polymer in its equilibrium state at temperatures as much as 180 K below the glass temperature or 130 K below the Kauzmann temperature. The results provide no evidence of a second-order transition or of a smeared transition in the entropy. In addition, we find no evidence that the entropy would become negative at a finite temperature.

The activation energy of a kinetic process can be determined as a function of conversion using isoconversion methods. In the past, such isoconversion methods were used to examine the variation in apparent activation energy (Ex) through the glass transition using differential scanning calorimtery (DSC) heating data. In this work we apply a similar isoconversion analysis to data obtained by cooling polystyrene through the glass transition using both capillary dilatometry and differential scanning calorimetry (DSC), and we compare the results to an analysis of DSC heating data for the same material. Although the values of Ex obtained from the isoconversion analyses of cooling data are found to be in good agreement with values of average activation energy (Eave), the values of Ex from analysis of heating data are considerably lower. Hence, we demonstrate that to obtain estimates of the activation energy, the isoconversion analysis should be performed on cooling rather than on heating. We also show that the conversion dependence of Ex can be directly related to the temperature dependence of the equilibrium relaxation time.

The melting behavior of aluminum nanoparticles having an oxide passivation layer is examined using a differential scanning calorimetry (DSC). Both broad and narrow size-distributed particles are studied, and the weight-average particle radius ranges from 8 to 50 nm. With decreasing particle size, the melting response moves towards lower temperatures and the heat of fusion decreases. The effect of the oxide coating on the particles is to apply a compressive force to the aluminum core, thereby increasing the observed melting point and the heat of fusion. The melting point depression, both corrected and uncorrected for the effects of the oxide shell, is linear with the reciprocal of particle radius, as predicted by Gibbs–Thomson equation, although only the corrected data give a value of the solid–liquid interfacial tension comparable to those reported in the literature. The size-dependent heat of fusion is significantly smaller than that predicted by the effects of the surface tension indicating that the solid nanoparticle is at a higher energy than expected, presumably due to the presence of defects or irregularities in the crystal structure at or emanating from the surface. This hypothesis is tested using our data, as well as using data in the literature for tin nanoparticles.

A comparison of the values of the glass transition temperature (T g ) measured on cooling and the limiting fictive temperature ðT 0f Þmeasured on heating as a function of cooling rate is performed for a polystyrene sample using both capillary dilatometry and differentialscanning calorimetry (DSC). The results from both techniques indicate that T 0f is systematically lower than T g presumably due to thebreadth of the relaxation on cooling. The Tool–Narayanaswamy–Moynihan (TNM) model is used to fit the experimental data from dila-tometry and DSC in order to ascertain the origins of the higher value of T g compared to T 0f . 2007 Elsevier B.V. All rights reserved. PACS: 64.70.Pf; 65.60.+a; 81.70.Pg; 82.35.LrKeywords: Glass transition; Measurement techniques; Modeling and simulation; Polymers and organics; Calorimetry 1. IntroductionWhen a glass-forming material is cooled from the equi-librium liquid state, molecular mobility decreases withdecreasing temperature; at some point, the time scale forconformational changes becomes comparable to the timescale of cooling and the material deviates from the liquidline and begins to form a glass. The glass transition is,hence, a kinetic rather than a thermodynamic transitionand depends on cooling rate [1,2]. The glass transition tem-perature (T

Although the relationship between the relaxation timescales of thermodynamic, mechanical, viscoelastic, and dielectric properties in amorphous materials has been studied extensively, no general consensus has been reached. In this work, we examine the relationship between the timescales of volume and enthalpy relaxation for polystyrene using the cooling rate dependence of the glass transition temperature ( T g ) obtained from capillary dilatometry and differential scanning calorimetry (DSC). Our analysis suggests that both volume and enthalpy exhibit similar relaxation timescales at temperatures above and below T g . The divergence of the times required to reach equilibrium noted in the literature at temperatures several degrees below the nominal T g is attributed to the effects of nonlinearity. The relationship between nonlinearity and dynamic heterogeneity is discussed.

Hexamethyldisilazane (HMDS) vapor treatment of plasma-damaged nanoporous organosilicate thin films has been studied as a function of treatment temperature in this work. Although, the HMDS vapor treatment facilitated incorporation of methyl (CH 3 ) groups subsequent to the removal of free hydroxyl (OH) groups in the damaged films at treatment temperature as low as 55 °C, the bonded OH groups were not removed. More significantly, detailed analysis of the results reveals that HMDS vapor modified only the surface of the plasma-damaged samples and not the entire film as expected. This is attributed to the formation of a thin solid layer on the surface, which effectively prevents penetration of HMDS vapors into the bulk. The Fourier transform-infrared (FT-IR) absorption and dielectric constant measurements confirm that the vapor treatment assists only partial curing of the plasma-damaged films. Alternative processes of curing the films with HMDS dissolved in supercritical carbon dioxide (SCCO 2 ) as a medium of reaction in static and pulsed modes were also attempted and the results are presented in this paper.

Glasses are inherently non-equilibrium materials, and consequently, their properties evolve toward equilibrium in a process known as structural recovery or physical aging. Recently, several authors have suggested that the equilibrium liquid line is not reached even when properties have ceased to evolve. In this work, we present measurements of the enthalpy recovery of polystyrene (PS) at temperatures ranging from 90.0 to 103.0 °C, for aging times up to 200 days. The results are analyzed in the context of the TNM model of structural recovery. In addition, we analyze data in the literature to determine whether enthalpy recovery ceases prior to the material reaching the equilibrium liquid line obtained by extrapolation of the liquid line above Tg. The results suggest that, in fact, the liquid enthalpy line is reached at temperatures below Tg when equilibrium is reached, i.e. when properties cease to evolve.

The oxidation reaction of aluminum nanoparticles with oxygen gas and the thermal behavior of a metastable intermolecular composite (MIC) composed of the aluminum nanoparticles and molybdenum trioxide are studied with differential scanning calorimetry (DSC) as a function of the size and size distribution of the aluminum particles. Both broad and narrow size distributions have been investigated with aluminum particle sizes ranging from 30 to 160 nm; comparisons are also made to the behavior of micrometer-size particles. Several parameters have been used to characterize the reactivity of aluminum nanoparticles, including the fraction of aluminum that reacts prior to aluminum melting, heat of reaction, onset and peak temperatures, and maximum reaction rates. The results indicate that the reactivity of aluminum nanoparticles is significantly higher than that of the micrometer-size samples, but depending on the measure of reactivity, it may also depend strongly on the size distribution. The isoconversional method was used to calculate the apparent activation energy, and the values obtained for both the Al/O 2 and Al/MoO 3 reaction are in the range of 200–300 kJ/mol.

The calorimetric glass transition temperature, T g , was measured for both linear and cyclic polystyrenes freeze-dried from dilute solutions of 0.10, 0.05, and 0.02% of polymer by weight in benzene. Upon freeze-drying, T g was found to be depressed by 4–15 K depending on the sample, solvent concentration, and freezing conditions. Annealing under vacuum at moderate temperatures, from 40 to 140 °C and 0.05 Torr, resulted in the shift of T g back towards its bulk value and was accompanied by a decrease in sample weight. The data is consistent with the observed weight loss being due to residual solvent. The amount of residual solvent is a strong function of the annealing temperature and the initial freeze-drying solution concentration; exposure to vacuum at temperatures far below T g is generally insufficient for residual solvent removal.

We report results from a study of isotropic residual stress development for thermosetting resins during cure and subsequent thermal cycling. We use a recently developed thick-walled tube method to impose three-dimensional isotropic constraints on the resin. The strains at the outer surface of the thick-walled tube are monitored by strain gauges. Experiments are performed for a closed tube where the resin is cured in the compressive state and in an open tube where tensile stresses build up at gelation. The cure-induced tensile stresses are much lower than expected from cure shrinkage due to the inability of the curing resin to sustain large tensile stresses in the gel-state.

Post treatments by annealing or supercritical carbon dioxide (SCCO2) exposure of plasma-enhanced chemical vapor deposited hydrogenated amorphous silicon carbide (a-SiC:H) films are reported to reduce the dielectric constant up to 2.1. The a-SiC:H films were prepared using diethylsilane diluted in methane. The deposition precursors and conditions were chosen to fabricate a thermally unstable a-SiC:H film with a high concentration of CHn (n = 1, 2, 3) functional groups. The Fourier transform infra-red spectroscopy measurements of as-deposited films show a-SiC:H networks and presence of a-C:H moieties throughout the film. The films were then annealed in vacuum up to 450 °C to remove thermally unstable CHn as well as other species. After annealing, the dielectric constant of the films decreased from 4.2 up to 2.1 due to the decrease in film density. The reduction suggests the presence of molecular voids/pores. Annealing produced a stable film with thickness decreasing by a maximum of 6% only. SCCO2 treatment was also explored and found to be more effective in the extraction of CHn and other species at 200 °C.

The temperature-modulated differential scanning calorimetry (TMDSC) responses during cure of both epoxy/aromatic amine and dicyanate ester/polycyanurate systems are modeled using chemical reaction kinetics with diffusion control. Physical aging effects are incorporated into the model using the Tool–Narayanaswamy–Moynihan (TNM) equation. We investigate the assumption that the mobility factor, which may be obtained from experimental temperature-modulated differential scanning calorimetry reversing heat flow data, is related to the diffusion factor for the two systems.

New volume recovery measurements have been made on an epoxy glass for temperature down- and up-jumps from 1.0 to 4.0 K to investigate the τ-effective paradox. Due to the sensitivity of the measurements and to the fact that data points are obtained linearly in time, we are able to accurately measure the effective relaxation time at much smaller departures from equilibrium than those examined originally by Kovacs and subsequently by other researchers. The paradox is observed for the largest (2.0 and 4.0 K) up-jumps performed at the highest aging temperature. The magnitude of the expansion gap disappears as the departure from equilibrium approaches zero indicating a resolution of the paradox.

The calorimetric glass temperature was measured for three cyclic polystyrenes with apparent molecular weights ranging from 4.0×10 3 to 195.5×10 3 g/mol for both bulk material and for samples freeze-dried from dilute solution. Freeze-drying from dilute solution was found to reduce the glass temperature by 7–14 K depending on the sample. These T g depressions are 5–12 K greater than those found previously for freeze-dried linear polystyrene. Annealing at 403.2 and 443.2 K (130 and 170 °C) resulted in recovery of the T g back to the bulk value with the time scales depending on both temperature and the magnitude of the T g reduction; the low apparent activation energy dependence of the recovery of T g precludes its being due to viscous flow.

The physical aging behavior of amorphous selenium has been investigated using differential scanning calorimetry and conventional and interrupted creep experiments. As a result of physical aging, enthalpy decreases and the creep and recovery curves shift to longer times. The times required to reach equilibrium for enthalpy recovery and creep appear to have different temperature dependences resulting in enthalpy reaching equilibrium before creep at aging temperatures a few degrees below the nominal glass temperature. In the nominal glass transformation range, however, the times required to reach equilibrium are approximately the same. A general picture of aging behavior has emerged from our data on selenium coupled with past work on polyetherimide and polystyrene.

Viscosity and recoverable shear creep compliance measurements of amorphous selenium and Se x As 1− x binary solutions have been made below and above the glass transition temperature. The measurements suggest that selenium is a relatively low-molecular weight polymer and that the arsenic–selenium solutions are dynamic networks which display viscosities similar to those of uncrosslinked materials. We observe no evidence for the presence of rings.

Creep compliance measurements have been made on low moisture and bubble-free amorphous boron trioxide, B2O3, over the temperature range from 250°C to 320°C in an attempt to determine the existence and form of molecular moieties in the material. No polymeric modes were found to be present in the retardation spectrum, L(τ), which was calculated from the recoverable compliance measurements. Based on the shape of L(τ) and the presence of liquid flow, it is concluded that boron trioxide behaves similarly to a low molecular weight glass former or a tightly cross-linked, but labile, network. Additional evidence is given that appears to rule out the existence of a network.

The status of temperature-modulated differential scanning calorimetry is reviewed, including current methods of data analysis, and the experimental conditions and calibrations necessary for meaningful data interpretation. The theory and application of TMDSC to absolute heat capacity measurements, the glass transition, and crystallization and melting are also reviewed and discussed.

Volume and enthalpy recovery measurements were used to study the physical aging behavior of a polystyrene. Isothermal aging at temperatures near Tg was studied with aging times ranging from several minutes to several days. The data are satisfactorily fit using the Tool–Narayanswamy–Moynihan model of structural recovery. The times required to reach equilibrium are the same for the two properties at temperatures where equilibrium was achieved. Extrapolation to lower temperatures using the model indicates that the times required to reach equilibrium may be different for the two properties at these lower temperatures. The recovery data were plotted as the departure from equilibrium and then were shifted horizontally to superpose at long times. The shift factors for volume and enthalpy recovery data agree with one another above 94°C but they diverge below this temperature. The shift factor data deviates from the WLF or VTFH equation for temperatures below Tg. In addition, we find that the normalized rate of approach is the same for volume and enthalpy recovery at 97°C in the limit of a linear temperature jump, but it differs between the two properties for a nonlinear jump even though the times required to reach equilibrium are the same.

Temperature modulated differential scanning calorimetry (TMDSC) data are simulated through the glass transition region using the Tool–Narayanaswamy–Moynihan (TNM) model of structural recovery. The simulated data are analyzed using a fast Fourier transform similar to conventional Temperature-modulated differential scanning calorimetry techniques. The glass transition temperature (Tg) and enthalpy of aging (ΔHa) were calculated from the simulated data and compared with the values expected based on linear isothermal simulations for the case of the glass temperature and based on the input simulated data for ΔHa. The errors associated with the TMDSC data analysis were thus quantified.

The thermal stability of poly(methyl methacrylate) doped with 0.16 wt% copper naphthenate was examined by monitoring weight loss over time at 220°C. The addition of copper naphthenate decreased the thermal stability by about 2 orders of magnitude. Significant enhancement of the thermal stability was accomplished by extracting the copper ions from the polymer matrix. The extraction was performed in liquid carbon dioxide at room temperature and 2 500 psi with thenoyltrifluoroacetone employed as a carbon dioxide-soluble chelating agent. After the extraction process was completed, the thermal stability of the polymer/metal system was within about 0.6 orders of magnitude of the value for the neat polymer.

It is well known that structural recovery in polymeric glass formers leads to the enthalpy overshoot in nonisothermal experiments. One common model for describing this phenomenon is the Tool-Narayanaswamy-Moynihan (TNM) equation. Here we apply the TNM equation to analyze a typical temperature-modulated DSC (TMDSC) trace and examine the influence of the material nonlinearities on the dynamic heat flow. We show that, in the glass-transition region, the oscillating heat-flow response to the sinusoidal temperature input is a distorted sine wave. Hence, linear analysis of the phase lag between heat flow and temperature is not physically meaningful. The degree of distortion of the sinusoidal heat flow increases as the magnitude of the excess enthalpy annealing peak increases. In addition, we perform a thermal analysis of samples having different geometries and show that an apparent phase lag can result in the measured response due to the presence of thermal gradients in the sample. The significance of the results is discussed.