Tips & Tricks

Πάντα ῥεῖ (Panta Rhei) "Everything Flows" – Thermal Analysis on Liquids

In literature, it can often be seen that – in addition to solids (compact samples, powders, etc.) and highly viscous materials (e.g., gels or pastes) – low-viscosity liquids can also be measured.

The purpose of this article is to provide some advice as to which material properties should be taken into consideration for sample preparation and which measurement conditions are recommended for the investigations. Main focus is on Differential Scanning Calorimetry (DSC), Thermogravimetric Analysis (TGA) and the Laser Flash Technique (LFA).

Differential Scanning Calorimetry (DSC):

In DSC, contact between the sample and the crucible bottom is crucial for the signal intensity. Many liquids, however, exhibit capillary ascension upon contact with the crucible wall; i.e., a concave surface is formed, and the liquid is lifted at the wall, due to adhesion forces between the liquid and the solid which are stronger than intermolecular adhesion forces in the liquid itself. As a result, it is often the case that only a small amount of substance remains on the crucible bottom.

In order to circumvent this effect, it is advisable (see fig. 1) to insert just a small amount of liquid by means of an injection or pipette so that only the bottom is covered.

Fig 1. Filling level for DSC crucibles

As an example for liquid sample testing, figure 2 shows the thermal behavior of butyl acetate, a colorless solvent with the elemental formula C6H12O2. After cooling the liquid down to -170°C, the solid which is formed initially remains amorphous, then crystallizes at -109°C (peak temperature) and melts again at -77°C (extrapolated onset temperature).

Fig. 2. DSC measurement on butyl acetate; Al crucibles with pierced lid, heating rate: 10 K/min, nitrogen atmosphere

Another factor that must be taken into consideration is the vapor pressure of the sample components as a function of temperature. A high vapor pressure in open crucibles during heating leads to an early onset of evaporation and to broad endothermal peaks. This may cause other interesting effects to get overlaid – as is the case with some liquid resins.

For pure substances, the molar evaporation enthalpy (and also the heat of evaporation under consideration of the molar mass) can be determined by means of vapor-pressure measurements, e.g., high-pressure DSC (in accordance with ASTM E 1782).

With hermetically sealed aluminum crucibles, internal pressure build-up may ultimately result in deformation or even bursting of the crucible. Depending on the desired temperature range and the objective of the investigation, it is therefore sometimes necessary to use crucibles which are better pressurized. In addition to low-pressure crucibles made of aluminum, medium-pressure stainless steel crucibles or high-pressure stainless steel or titanium crucibles are available.

Thermogravimetric Analysis (TGA):

The early onset of evaporation described above manifests as a mass change at a temperature far below the boiling point (fig. 3). On the other hand, if a lid with an extremely small hole is employed, evaporation will be delayed until close to the boiling point (see also fig. 3). The mass loss itself is considerably faster in this case; the corresponding TGA curve exhibits a sharp downward slope. For these kinds of investigations, aluminum lids with a 50-μm hole can be employed.

Shown in figure 3 are two measurements on water: one in an open crucible (blue), the other in a crucible which has a lid with a microhole (red). The two curve profiles differ from each other significantly.

Fig. 3. TGA measurement on water, Al crucibles, heating rate: 10 K/min, N2 atmosphere

Laser Flash Analysis (LFA):

For determination of the thermal diffusivity by means of LFA, containers are used which guarantee an even sample thickness layer. This is necessary since the thickness of the sample goes into the calculation formula in squared form. Brand new in this regard is the sample holder shown in figure 4, featuring very easy handling, high measuring accuracy and high reproducibility. From bottom to top, the sample container consists of a supporting ring, two sealing discs of stainless steel with an intermediate plastic sample ring featuring two feed openings for liquid samples, and an upper cover plate. The plastic ring and stainless steel discs can be exchanged at low cost.

Fig. 4. LFA sample container for liquids; the upper sealing disc is painted orange-red. The two small holes in the middle symbolize the feed openings for liquid samples.

The sample holder is especially well suited for water-based liquids, oils or resins, but also for highly viscous materials such as fats.

Figure 5 depicts three measurements on water, taken with different sample holders of the type shown in figure 4 in the temperature range from RT to approx. 80/85°C. The evaluation was made on each using a 3-layer model. All test result values are within+/- 5% of literature data.

Fig. 5. Comparison of three measurements on water; the red symbols represent the measured data

Figure 6 shows the corresponding detector signal together with a model fit at 60°C. It is clear that the experimental and adapted data are in good agreement.

Fig. 6. Detector signal with model fit at 60°C