How to Interpret the Unexpected Transitions in DSC Results

Besides providing a wealth of information, Differential Scanning Calorimetry (DSC) is user-friendly in terms of sample preparation, experimental configuration, and interpretation of the results. DSC is a thermal analysis technique which measures the temperature and heat flow associated with transitions in materials as a function of temperature and time.

Such thermal analysis measurements are useful to obtain qualitative and quantitative data about the physical and chemical changes, including changes in heat capacity or exothermic/endothermic processes.

Glass transition temperatures, melting and boiling points, crystallization time and temperature, percent crystallinity, specific heat, oxidative stability, purity, reaction kinetics, degree of cure, rate of cure are some of the specific information derived from DSC measurements.

DSC is user-friendly n terms of sample preparation, experimental configuration, and interpretation of the results, These capabilities make DSC the most widely used thermal analysis technique. There are, however, some common DSC events/ transitions that can be the cause of less than optimum results and/or misinterpretation.

This paper describes several of these events with causes and solutions. Figure 1 is an artificial DSC curve which was generated to illustrate these events/transitions. The curve is artificial in the sense that all of these events would not occur in the same real world DSC curve.

This article helps DSC users to interpret unexpected transitions in DSC results. The overall quality and interpretation of DSC results can be improved through the implementation of some of the recommended procedures and solutions.

Figure 1. Artificial DSC curve

Large Endothermic Start-up Hook

Causes

Significant baseline change, typically endothermic, may be there at the beginning of a programmed thermal analysis, due to variations in the heat capacity of the sample and reference. Since there is a direct relationship between the heat capacity and weight, an endothermic shift implies that the reference pan is very too light to offset the sample weight. Faster heating rates further intensify this effect.

The thermocouple junctions in the DSC cell base become cold when operating subambient, due to transfer of cold from the cell cooling head. This effect increases as the temperature is lowered and/or the time at lower temperatures is increased.

Effects on Results

Detecting weak transitions becomes difficult because of a large “start-up hook” or a sloping baseline. In addition, it may not be possible to reproduce transition temperatures and measured heat flow (DH) during the first 2-3 minutes of the experiment.

Solutions

A series of reference pans of various weights in 2mg increments is built using aluminum foil or additional pan lids. A reference pan with 0-10% more weight than sample pan is used when running a sample. Results with an epoxy prepreg sample are shown in Figure 2. Best results are obtained with 1.5 lids, whereas overcompensation and an exothermic start-up hook are the results of 2 lids.

Figure 2. Effect of reference pan weight

The effect of correct compensation on the glass transition results are shown in Figures 3 and 4. It is to be noted that these results are acquired when heating at 20°C/minute from a 100°C isothermal hold. Starting the heating at a temperature that is at least 2-3 minutes less than the range of interest at the heating rate selected, i.e., the analysis has to be started at least 50°C below the first thermal event of interest when the rate is at 20°C/minute.

Figure 3. Start-up hook and T_g with no reference pan

Figure 4. Start-up hook and T_g with correct reference pan

When operating at a temperature less than 0°C, dry nitrogen purge gas is recommended to be used through the cell base VACUUM PORT at 50cc/minute as well as the normal purge gas. The typical improvement achievable is illustrated in Figure 5.

Figure 5. Proper gas purging improves subambient baseline performance

Transition(s) at 0°C

Causes

Weak transitions around 0°C are an indication of the presence of water in the sample or the purge gas. Although these transitions are typically endotherms, they may appear different from a melting peak. The transition usually appears as illustrated in Figure 6 due to condensation of water on both the reference and sample pans. In addition, the peaks may appear slightly below 0°C because of the dissolution of impurities by the moisture from the cell and pans.

Figure 6. DSC transition due to moisture in the purge gas

Effects on Results

Reproducible results may not be possible due to the presence of water in the sample as it can lower transition temperatures by acting as a plasticizer. During the run, volatilization of water causes an endothermic peak and a baseline shift. A perturbation in the baseline caused by the presence of water in the purge gas poses challenges for detection of real transitions around 0°C.

Solutions

Hygroscopic samples need to stored in a dessicator and then loaded into pans in a dry box. The complete sample pan with sample is weighed before and after the run to measure the weight change, which could give details about an unexpected transition.

A drying tube is placed in the line to dry the purge gas. An epoxy sample after loading at -100°C is depicted in Figure 7. The absence of any transitions at 0°C means that it is possible to eliminate water condensation with proper precautions even under favorable condensation conditions. The liquid nitrogen cooling accessory (LNCA) is essential for sample loading at temperatures below 0°C. Sample loading above 0°C must be performed with any other cooling accessory.

Figure 7. Quench cooling samples at subambient temperatures

Apparent “Melting” at Glass Transition (T_g)

Causes

Heating of a material through its glass transition leads to the release of internal stresses in the material caused by handling, processing or thermal history. The molecule transforms from a rigid to a flexible structure at T_g and thus can move to release the stress.

Effects on Results

Molecular relaxation generally occurs as a weak endothermic transition close to the end of a glass transition (Figure 8), shifting the measured glass transition temperature several degrees or leading to misreading of the T_g as an endothermic melting peak.

Figure 8. Molecular relaxation can cause T_g to appear as a melt

Solutions

Heating the material to a temperature at least 25°C higher than the T_g, followed by quench cooling to a temperature less than the T_g release the internal stresses in it. The same material illustrated in Figure 8 is shown in Figure 9, but after curing at 200°C and then quench cooling to 25°C. Effect of cooling rate on shape of T_g is depicted in Figure 10.

Figure 9. DSC scan of PET after quench cooling

Figure 10. Effect of cooling rate on shape of T_g

Exothermic Peaks Below Decomposition Temperature While Heating

Causes

Exothermic behavior results when a thermoplastic polymer is crystallized or a thermosetting resin is cured. It is possible to use the amount of heat involved in these transitions to find out the degree of cure and % crystallinity if scans of suitable standards are available.

When an exotherm is obtained in the DSC profile of a polymer at a temperature that is suspected to be too low to be decomposition by the operator, running the material in the TGA helps evaluation. The absence of a TGA weight loss that is coincidence with the DSC exotherm reveals that the exotherm is curing or crystallization.

Effects on Results

Thermal history greatly influences the presence or absence of exothermic crystallization peaks in thermoplastic materials. Hence, the thermal history of the sample needs to be tightly controlled to obtain reproducible DSC results. The different results obtained for PET after quench cooling and programmed cooling at 10°C/minute, respectively, are presented in Figures 11 and 12.

Figure 11. DSC scan of PET after quench cooling

Figure 12. DSC scan of PET after slow cooling

Having a well-defined T_g means that the quenched material has a significant amorphous structure, which transforms upon heating to a crystalline structure prior to melting at about 235°C. Since the DH of melting is slightly higher than the DH of crystallization, the initial structure is largely amorphous. The weak T_g of the slowly cooled material indicates that the initial structure is nearly entirely crystalline. However, the material has a crystalline structure at the beginning of the DSC experiment and therefore additional crystallization does not take place before melting at 235°C.

Solutions

To compare thermoplastic materials, a common known thermal history is provided to the materials by program cooling or quench cooling from above the melting temperature. The recommended procedures to give a known thermal history to polymers are defined in the ASTM D3418-82 standard.

Baseline Shift after Endothermic or Exothermic Peaks

Causes

The changes in sample weight, the specific heat of the sample, or heating rate lead to baseline shifts. The specific heat of sample changes subsequent to transition of the sample such as melting, crystallization or curing. The change in the sample takes place during volatilization or decomposition.

Effects on Results

The calculation of ΔH is based on sample weight (J/g, BTU/lb, etc.) and therefore calculating DH subsequent to a weight change will provide erroneous results. It is difficult to integrate a peak involving a baseline shift. In addition, operator subjectivity in setting integration limits and baseline type typically makes this integration less accurate.

Solutions

The sample weight is measured before and after a run to find out the occurrence of weight loss. If the cause of the transition is crystallization or melting, the DH of the transitions is compared using different limits and types of baselines. An example requiring the use of sigmoidal baseline is illustrated in Figure 13.

Figure 13. DSC scan of PET crystallization and T_g on cooling

Sharp Endothermic Peaks During Exothermic Reactions

Causes

Sharp peaks similar to those in Figure 1 above 300°C, are usually the result of experimental phenomena rather than real material transitions. For instance, sharp peaks can be the result of rapid volatilization of gases confined in the material, similar to rapid volatilization of gases trapped in a partially sealed hermetic pan.

Effects on Results

These sharp endotherms may be erroneously interpreted as melting peaks associated with minor components. Sample mass changes during volatilization can affect obtaining accurate quantitative results. The corrosive nature of some of the volatiles such as halogenated flame retardants can cause damage to the DSC cell during extended operation.

Solutions

The sample weight is measured before and after a run to identify whether weight loss has taken place. The temperature limit of further experiments is reduced if failing to obtain useful information due to volatilization. The use of a Pressure DSC cell is recommended.

This information has been sourced, reviewed and adapted from materials provided by TA Instruments.

For more information on this source, please visit TA Instruments.