Melting of polymers by non-isothermal, temperature-modulated calorimetry: analysis of various irreversible latent heat contributions to the reversing heat capacity

This paper provides an analysis of contributions to the apparent, reversing heat capacity when measured by temperature-modulated differential scanning analysis (TMDSC) with an underlying heating rate in the temperature range where irreversible transitions with latent heats occur. To deconvolute the data of a TMDSC scan into a total and reversing part, it is common practice to use the sliding averages and the first harmonics of the Fourier series of temperature and heat-flow rate. Under certain conditions, this procedure produces erroneous reversing contributions which are detailed by experiment and simulation. Unless the response to the temperature modulation is linear, the total heat-flow rate is stationary, and the transition is truly reversible and occurs only once during the temperature scan, one cannot expect a true deconvolution of total and reversible effects. In the presence of multiple, irreversible transitions within a modulation period, however, each process involving latent heat can increase the modulation amplitude, as demonstrated by computer-simulation of polymer melting. As a result, the multiple transitions may give erroneously high latent heats when integrating the apparent reversing heat capacity with respect to temperature.     

Reversible Melting of Semi-crystalline Polymers Frequency dependence of the reversible melting enthalpy

Modulated DSC (MDSC) has been used to study the heat flow during melting and crystallisation of some semi-crystalline polymers i.e. different grades of polyethylene (LDPE, LLDPE and HDPE), and polypropylene (PP). The heat capacities measured by MDSC are compared with the hypothetical complex heat capacities of Schawe and it is shown that numerically they are equivalent; nevertheless, the concept of the complex heat capacity is problematic on a thermodynamic basis. A reversing heat flow (proportional to the experimental heat capacity of the material) was present at all conditions used for the study. In the melting zone of the polymers it depends on the modulation frequency and on the amplitude. Higher amplitude and frequency of modulation reduce the ratio of the reversing heat flow to the total heat flow, the latter is nearly independent on these parameters. The reversible component of the melting enthalpy of polymers depends on the modulation frequency, the modulation amplitude and the type of the polymer. It increases by increasing the branching in polyethylene. The existence of the reversible heat flow during the crystallisation and melting is contrary to the current hypotheses and theories of polymer crystallisation. This revised version was published online in July 2006 with corrections to the Cover Date.     

机械振动对聚丙烯结晶作用的影响

对热塑性高分子作熔体加工时,在此过程中同时再叠加机械振动,设计制造的加工机器一方面具有卓越的技术经济指标,另一方面制品的物理性能也得到了提高[1,2].此类研究的重点多集中在动态成型过程中聚合物的流变行为或者成型后制品的力学性能方面[3,4],对聚合物本身在此外加交变力     

A comparison of different evaluation methods in modulated temperature DSC

Modulated temperature-DSC is a new method for measuring the thermal behaviour of materials. In this method, the response of the sample to a time-dependent signal (sinusoidal temperature change) is measured. Two different methods are known for the evaluation of the measured data. The first is the separation of the measured data into reversing and non-reversing components of heat flow. The second is based on the linear response theory and yields a complex heat capacity with a real part (storage heat capacity) and an imaginary part (loss heat capacity).

The theoretical basis and the possibilities of interpretation of both evaluation methods are investigated. The results of both methods are compared theoretically for the case of simple time-dependent processes. Experimental results are given for the glass transition process.     

对MDTA中测样品热容量的准恒温法的评述

对用调制差热分析（ＭＤＴＡ）准恒温法测样品热容量的情形进行了讨论。通过结合最基本的热传导定律和ＭＤＴＡ模型，指出了目前国际上测量样品热容量的准恒温法只能得到在所测温度范围内的物质热容量平均值，调制温度的幅度越大或调制频率越高，所得到的热容量数据越平滑。在所测温度范围内样品热容量基本不变时，用ＭＤＴＡ准恒温法较好；但当样品热容量在所测温度范围内有明显变化时，用传统差热分析法（ＤＴＡ）更好一些。     

Obtaining Modulated Temperature DSC Curves Through a Non-conventional DSC Method

The DSC curve obtained in conventional equipment usually only shows the resultant thermal effect due to simultaneous phenomena, which may occur during isothermal or dynamic analyses. This does not allow one to identify the processes properly and may cause an erroneous interpretation of the resulting curves. Modulated DSC equipment enhances the operating conditions and the analysis capacity of conventional DSC by superimposing a sinusoidal temperature modulation on the linear temperature control. Thus reversing and non-reversing heat flow curves are obtained, which are, respectively, the heat capacity and kinetic components of the DSC curve. Therefore, events that are related to these components can be separately analyzed. A method to obtain curves similar to the MDSC reversing and non-reversing components was developed using conventional DSC equipment in a non-conventional way. It was applied to analyze samples of poly(ethylene terephthalate) (PET) taken from bottles of mineral water. The second PET crystallization step that occurs during its melting was quantified and an apparent initial crystallinity was obtained from the resulting data. This revised version was published online in August 2006 with corrections to the Cover Date.     

A systematic solution to the problem of sample background correction in DSC curves

The problem of sample background correction in differential scanning calorimetric curves is addressed in this paper. An equation is derived for the heat capacity of the system, which corresponds to the sample background. Thereby, it is assumed that during the thermal event the system is a two-component mixture of the initial substance and the final product. According to this model, the variation in heat capacity of the system is due both to the increase in the partial concentration of the product at the expense of the initial substance and to the physical change in the specific heats of the two components, resulting from the temperature increase. The final result of the derivation is an integral equation which can be solved by means of a numerical technique. The algorithm used is presented in detail. The model is general, and can be applied to diverse exothermic or endothermic processes. The melting of a semi-crystalline polymer and the cure process of a thermoset are given as demonstrative examples. The method improves the reliability and the reproducibility of the data.     

Einfluß der Probenmasse auf die quantitative Bestimmung von Wärmetönungen mit der Differentialthermoanalyse

From theories about differential thermal analysis (DTA) it can be deduced that the DTA peak is proportional to the mass of the sample. Using the method of dynamic difference calorimetry deviations from this proportionality were found if the mass of the sample exceeded a certain value. This effect is demonstrated by measuring melting curves of tin. By investigating the change of the temperature distribution and heat flow in the system furnace-sample holder-sample during the thermal effect it can be shown that the deviation from the proportionality is a correct phenomenon.     

Application and analysis of a DSC-Raman spectroscopy for indium and poly(lactic acid)

Simultaneous measurement system of DSC-Raman spectroscopy and its analysis method are developed. The developed method was applied to the melting of Indium and the optimum laser irradiation condition was determined. The obtained result of the heat flow is similar to the modulated DSC and the precise melting temperature and the heat of fusion can be obtained from the analyzed DSC. DSC-Raman spectroscopy is also applied to PLLA. Analyzed data indicate the existence of the recrystallization behavior in addition to T _g and T _m. Corresponding to these transitions, Raman peak shifts, intensities, and widths varied. From those results, it is proved that DSC-Raman spectroscopy is useful for the analysis of thermal property of the polymer in connection with the polymer structure.     

A rheological investigation of aqueous solutions of poly (vinyl alcohol) during aging. Part IV: Repeated aging of concentrated solutions

Aging of aqueous solutions of poly (vinyl alcohol) obtained by melting of gels, formed in the aging process from solutions prepared by the dissolution of the polymer in hot water, was investigated using non-Newtonian viscometry and normal stress measurement. It was shown that, if the gels undergo melting at the same temperature at which the original solutions were prepared, aging of such solutions proceeds in the same way as in the case of the original solutions. If melting occurred due to heating at lower temperature, the resulting solutions aged more quickly.     

Treatment of water-induced curvature of the DSC heat flow rate signal

In samples containing a volatile phase, quite often the evaporation of the volatile substance during heating causes appreciable curvature of the DSC heat flow rate signal as function of temperature, making it difficult to quantify thermal transitions and reorganization phenomena occurring in the same temperature range. This is the case for e.g. polyamide–water, polyamide–alcohol, and polypropylene–water systems, thus complicating the study of polymer crystallization, melting, and metastability by DSC. In this study, maleic anhydride-grafted polypropylene particles of sub-micrometer diameters dispersed in water are discussed. These samples show, upon cooling from the melt, different degrees of extra supercooling in crystallization and several phenomena in the subsequent heating, like reorganization of a crystalline phase into another one, perfecting of crystallites, and melting. All these phenomena are difficult to analyze quantitatively due to the mentioned curvature of the DSC trace. In this article two methods, the “Reference” and “Extrapolation from the melt” methods, are described to correct for the influence of evaporation on the DSC heat flow rate signal and for the baseline signal, enabling the discussion of the transitions by way of the excess heat flow rate as function of temperature.     

Study of the crystallization and melting region of PET and PEN and their blends by TMDSC

The cold crystallization and melting of poly(ethylene therephthalate) (PET), poly(ethylene 2,6-naphthalene dicarboxylate) (PEN) and their blends were studied using temperature modulated differential scanning calorimetry (TMDSC) at underlying heating rates of between 1 and 3 K min^-1 and periods ranging from 30 to 90 s. The amplitude of modulation was selected in order to give an instantaneous heating rate β≥0. Heat flow is analyzed by the total heat flow signal o, which is equivalent to the conventional DSC signal, and the reversing heat flow o_REV, which only detects the glass transition and the melting processes. The dependence of the melting region in the reversing heat flow on the frequency of modulation is analyzed. The use of the so-called non-reversing heat flow o_NREV (=o-o_REV)) and the effect of frequency and amplitude on the complex heat capacity are also studied. The results show the complexity of these magnitudes. This revised version was published online in July 2006 with corrections to the Cover Date.     

DSC by the TGA/SDTA851e Considering Mass Changes

DSC measurements in open pans are often disturbed by mass losses such as sublimation during melting or release of water during chemical reactions. By simultaneous DSC and TG measurements the DSC signal can be corrected. For this purpose, a temperature dependent calibration function has to be determined by which the SDTA signal from the TGA/SDTA851^e measuring cell can be converted into a heat flow curve (DSC). By this procedure, accurate heat of melting can be determined despite ongoing sublimation in open pans. This method is illustrated with reference of the melting of anthracene. Additionally, condensation reactions were investigated and analyzed by DSC/TG even under ambient pressure, knowing the heat of evaporation. Using phenol formaldehyde resins the influence of the presence or the release of volatile reaction products on the reaction rate and kinetic parameters were studied. In general, the method can be used to correct DSC curves for thermal effects related to mass change. This revised version was published online in July 2006 with corrections to the Cover Date.     

Thermal effects accompanying Nylon-6 dehydration

The thermophysical behaviour of Nylon-6 with various moisture contents was studied. It was shown that the thermal effects occurring below the Nylon-6 melting temperature are due to the dehydration process. The temperature of the heat flow maximum is a function of the state of the water molecules in the polymer. It was found that the thermophysical study of Nylon-6 in the temperature region below the polymer melting temperature allows a more precise value of its melting heat to be obtained.

---

Zusammenfassung Es wurde das thermophysikalische Verhalten von Nylon-6-Proben mit verschiedenem Feuchtegehalt untersucht. Man fand, daß die thermischen Effekte unterhalb der Schmelztemperatur von Nylon-6 in Verbindung mit dem Dehydratationsprozeß stehen. Die Temperatur für den maximalen Wärmefluß ist eine Funktion des Status der Wassermoleküle im Polymer. Es wurde gezeigt, daß die thermophysikalischen Untersuchungen an Nylon-6 im Temperaturbereich unterhalb der Schmelztemperatur des Polymers die Bestimmung präziserer Werte für die Schmelzwärme erlauben.

     

Reversible Melting During Crystallization of Polymers Studied by Temperature Modulated Techniques (TMDSC, TMDMA)

Quasi-isothermal temperature modulated DSC and DMA measurements (TMDSC and TMDMA, respectively) were performed to determine heat capacity and shear modulus as a function of time during crystallization. Non-reversible and reversible phenomena in the crystallization region of polymers can be observed. The combination of TMDSC and TMDMA yields new information about local processes at the surface of polymer crystals, like reversible melting. Reversible melting can be observed in complex heat capacity and in the amplitude of shear modulus in response to temperature perturbation. The fraction of material involved in reversible melting, which is established during main crystallization, keeps constant during secondary crystallization for PCL PET and PEEK. This shows that also after long crystallization times the surfaces of the individual polymer crystallites are in equilibrium with the surrounding melt. Simply speaking, polymer crystals are ‘living crystals’. This revised version was published online in August 2006 with corrections to the Cover Date.     

Polymerization of para-xylylene derivatives (parylene polymerization). III. Heat effects during deposition of parylene N at different temperatures

Thermal effects accompanying the vacuum deposition of poly-para-xylene (Parylene N) at different temperatures have been studied by following the changes in the temperature of the substrate. Similarly to the case of polychloro-para-xylylene (Parylene C), two distinct exothermic effects were observed; one discrete, resulting in sharp exothermic spikes and the other continuous, resulting in the shift of the baseline. The spike effect, attributed to the solid-state polymerization of para-xylylene, is observed at the low-temperature range, the upper limit of which moves higher for higher deposition rates. The shift of a baseline as a function of deposition temperature exhibits two maxima, one independent of deposition rate and the second moving toward higher temperatures (that is, toward the first maximum) for higher deposition rates. First maximum falls at about ? 72°C and is attributed to the melting point of para-xylylene crystals. X-ray diffraction studies of polymer samples have shown that the existence of the second maximum is always followed by the appearance of an additional crystalline phase in the respective range of deposition temperatures. When the deposition rate is high enough, the second maximum merges with the first one, or virtually disappears. Under such conditions the new crystalline phase is no more detectable. It is postulated that the evolution of the additional amount of heat resulting in the appearance of the second maximum is due to the cyclization reaction and the formation of cyclic oligomers. This reaction very likely requires a particular spatial arrangement of monomer molecules and, therefore, it is suggested to take place in certain domains of the crystalline lattice.     

Reversible Melting of Semi-Crystalline Polymers 2. Annealing Near to the Melting Point

Annealing experiments have been carried out at a few degrees below the melting point of different polyethylenes (LDPE, LLDPE, HDPE), of polypropylene (PP) and of Nylon-6. The heat capacities decrease during the annealing, within a 2-4 min time scale, to a lower value which corresponds to the extrapolated heat capacity values obtained for the cooling cycle when the polymer is cooled from the melt. Heat capacities in the heating cycle following the cooling cycle of PP, Nylon-6 and HDPE have the same value as during the cooling section. This is not the case for LDPE and LLDPE. Exothermic total heat flow in the cooling section following the annealing indicates that the crystallisation takes place during the cooling rather than during the annealing period. The total melting enthalpy measured before and after the annealing cycle is the same. The reversing heat flow shows an excellent fit to the change of the crystallinity measured by small angle scattering of synchrotron radiation during a heating cycle at temperatures below the melting peak. A coupled thermodynamic interaction of the crystalline and the amorphous phases is concluded from this study. This kind of interaction is possible at the lateral end of polymeric chains incorporated into the crystalline phase. This is an indication of the portion of tie molecules in the system, i.e. the portion of fringed micelle type of crystalline morphology with respect to that of folded chain lamellae. This revised version was published online in July 2006 with corrections to the Cover Date.     

Thermal behaviour studies of procaine and benzocaine

The analysed substances, procaine and benzocaine, are two anaesthetic agents currently being administered in tablet form, also in the topical (cream, gel, balm) and injectable dosage forms. The TG/DTG/DTA curves were obtained in air at different heating rates. For determination of the heat effects, the DTA curves (in μV) were changed with the heat flow curves (in mW), so that the peak area corresponds to an energy in J g^?1 or kJ mol^?1. The non-isothermal experiments are preformed to investigate the thermal degradation process of these active substances, both as a solid and are performed in a dynamic atmosphere of air at different heating rates, by heating from room temperature to 500 °C. The kinetic analysis was performed using the TG data in air for the first step of substance’s decomposition at four heating rates: 7, 10, 12 and 15 °C min^?1. The data were processed according to an appropriate strategy to the following kinetic methods: Kissinger–Akahira–Sunose, Flynn–Wall–Ozawa, Friedman and NPK, to obtain realistic kinetic parameters, even if the decomposition process is a complex one. Thermal analysis was supplemented using Fourier Transform infrared spectroscopy coupled with the TG device to identify the anaesthetics with any products which may have formed (EGA—the evolved gas analysis).     

Transitions and relaxations in aromatic polymers

Transition and relaxation phenomena in 26 structurally related polyquinoxalines and other aromatic polymers were studied over a temperature range from 70 to 770°K by means of calorimetric, dilatometric, dynamic mechanical, and dielectric techniques. Differential thermal analysis and x-ray data showed these polymers to be essentially amorphous. The lack of crystallinity is attributed to geometric isomerism, resulting in conformational as well as configurational disorder. Calorimetric measurements gave discontinuities in heat capacities ranging from 12 to 54 cal/°C per mole of repeat-unit structures and provided unambiguous assignments of glass transition temperatures of these polymers. Depending upon structure, T_g varied from 489 to 668°K. Thermal expansion curves of annealed bulk polymer samples between 70 and 770°K exhibited only one discontinuity over the entire temperature range, namely at T_g, thus indicating the absence of any motion leading to transitions in the solid state of these polymers. Viscoelastic properties were obtained by means of torsional braid analysis and a longitudinal vibrational apparatus. In a typical case, the dynamic mechanical relaxation spectrum contained three loss maxima. A peak of low amplitude occurring at 483°K was attributed to impurity effects, resulting from endgroups and species of low molecular weight. The second and only major relaxation process occurred at 579°K, in the glass transition interval. A third, weak loss peak of unknown origin was found in the liquid state at 683°K. On the other hand, the dielectric loss curves of various polymers exhibited only one broad and strong absorption maximum at temperatures 30 to 100°K higher (depending upon a particular polymer) than equivalent major mechanical loss peaks. These differences are interpreted from a mechanistic point of view. Major mechanical relaxations occurring in the glass transition interval of these polymers are proposed to result from translational motions.     

Temperature Modulated DSC of Irreversible Melting of Nylon 6 Crystals

A new method is presented to analyze the irreversible melting kinetics of polymer crystals with a temperature modulated differential scanning calorimetry (TMDSC). The method is based on an expression of the apparent heat capacity, , with the true heat capacity, mc_p, and the response of the kinetics, . The present paper experimentally examines the irreversible melting of nylon 6 crystals on heating. The real and imaginary parts of the apparent heat capacity showed a strong dependence on frequency and heating rate during the melting process. The dependence and the Cole-Cole plot could be fitted by the frequency response function of Debye's type with a characteristic time depending on heating rate. The characteristic time represents the time required for the melting of small crystallites which form the aggregates of polymer crystals. The heating rate dependence of the characteristic time differentiates the superheating dependence of the melting rate. Taking account of the relatively insensitive nature of crystallization to temperature modulation, it is argued that the ‘reversing’ heat flow extrapolated to ω → 0 is related to the endothermic heat flow of melting and the corresponding ‘non-reversing’ heat flow represents the exothermic heat flow of re-crystallization and re-organization. The extrapolated ‘reversing’ and ‘non-reversing’ heat flow indicates the melting and re-crystallization and/or re-organization of nylon 6 crystals at much lower temperature than the melting peak seen in the total heat flow. This revised version was published online in July 2006 with corrections to the Cover Date.