Melting sample size dependence

The major contributions to dipole polarizations are additive and are similar whether the moiety is within a small or a large (polymeric) molecule. Even so, the secondary contributions to the overall dipole polarization of a sample are dependent on both the chemical and the physical environment of the specific dipole unit and on the size and the mobility of that unit. Thus dipole contributions can be utilized to measure glass transition temperature Tg and melting point Tm. 

Properties can be classified as either intensive or extensive, depending on whether their values change with the size of the sample. Intensive properties, like temperature and melting point, have values that do not depend on the amount of sample. Thus, a small ice cube might have the same temperature as a massive iceberg. Extensive properties, like length and volume, have values that do depend on the sample size. An ice cube is much smaller than an iceberg. 

The change in sample size has been shown to influence product ratios, fractional conversions, and rates of pyrolysis for several polymers [14]. However, the dependence of pyrolysis results on sample size does not have only a negative impact on data reproducibility. This effect may reveal mechanistic features of the pyrolysis process, such as secondary reactions that occur during the residence time of primary products in the melt and diffusion-restriction of the interaction of long-chain radicals, when the thickness of the degrading film is comparable with the diameter of the hydrodynamic volume of the polymer molecules. 

All the DSC methods of purity determination depend on the applicability of the van t Hoff equation. This restricts the method to systems where the impurity forms a simple eutectic phase diagram with the major component that is, the impurity or impurities are soluble in the melt and the components do not form solid solutions (53). Use of the van t Hoff equation assumes that the solution of impurity in major components above the melting point is an ideal solution in the thermodynamics sense. Also, the method assumes that the solid-liquid system is essentially in true thermodynamic equilibrium during the measurements. Failure to meet any of these conditions will lead to erroneous results. Other possible errors are associated with the instrumentation employed. This involves the use of the smallest possible sample size consistent with homogeneity (50), proper encapsulation to minimize temperature gradients within the sample, and the slowest possible heating rate lo approach equilibrium conditions. It is recommended that the melting... 

Discussion should draw attention to the independence from sample size - if melting temperature depended on amount, the experiment would not be worth doing. (Of course, the time it takes for a sample to melt does depend on amount, for example snowflake versus iceberg.)... 

Jahnert et al. (2008) showed that there are the pronounced differences in the magnitude and pore size dependence of the transition temperature T d) of water obtained by the spin-echo method due in part to the chosen delay time x (Figure 1.277). For instance, the X value (2-20 ms) plays an important role as the pore size of the silica samples decreases. The melting point depression of water in silica pores of nominal pore size 4 nm was found to be 14 K when measured with x=l ms, but only 6 K when measured with x=20 ms. This was attributed to a pore size dependence of the relaxation time T2 of the confined liquid (Jahnert et al. 2008). 

The melting temperature-composition relations that were described above were for rapidly crystalhzed samples. This crystallization procedure results in relatively small crystallite sizes. In an alternative procedure the crystallization can be conducted isothermally at elevated temperatures and never cooled prior to fusion. It is then found that the melting temperatures are dependent on the nature of the comonomer.(79) Ethylene butene and hexene copolymers behave similarly to one... 

If a small pinhole is made in the cover of the pan, or if a crimped or open pan is used, then volatiles can escape, but the shape of the DSC curve and the temperatures involved will depend upon the rate at which the volatiles can escape, which in turn depends upon the extent of crimping or the size of a hole made in the pan together with experimental conditions, such as heating rate and sample size. As a result, data obtained may not be very reproducible. In some cases, both melting of the solvate and desolvation in the solid state may happen, which gives rise to two endothermic peaks. 

---