Thermal analysis by structural characterisation

We hear from thermal analysis pioneer Professor Mike Reading on a new general purpose thermal analysis technique

Thermal methods are amongst the most powerful techniques for characterising materials. The most widely used is differential scanning calorimetry or DSC (often with the variant of using modulated temperature, MTDSC) which, amongst other quantities, measures transition temperatures. From this information and the details of the form of the transition a great deal of useful information can be obtained on composition and structure.

However, DSC only provides global information; it can determine how much of a particular material is present but, when there is more than one component in a sample, it cannot provide information on how the components are distributed in space. A similar comment applies to techniques that are often used in conjunction with DSC such as dynamic mechanical analysis and thermomechanical analysis. Hot stage microscopy, abbreviated to HSM (also called thermomicroscopy or thermooptical analysis), is a well-established technique with a number of well-defined applications^1,2, notably characterising how crystal structure changes with temperature. However, it is very limited in what it can achieve when analysing samples that are filled and it cannot be routinely used for detecting glass transitions.

Here, I describe a recent technique called TASC that overcomes these limitations; it relies on using image analysis to process videos made using HSM. Thermal analysis by structural characterisation (TASC)³ is a method based on characterising how a selected structure changes with temperature. The capabilities of this approach include: Determining melting temperatures.

• Determining glass transition temperatures.
• Transition temperatures at different locations on the sample can be measured (local thermal analysis or LTA).
• Surface analysis can be performed so the thermal behaviour of topmost layer can be characterised.
• Transition temperatures of individual particles in a particle mixture can be determined.
• Changes in any structure that can be seen by an optical microscope can be followed as a function of time and temperature.
• Ultra-small samples can be analysed.
• Analysis of dissolution behaviour is possible.

No single instrument offers this range of capabilities and some of them are unique.

[caption id="attachment_49699" align="aligncenter" width="583"] Figure 1: An example of a TASC experiment; left is an image of four indentations in a sample of filled polystyrene. The middle image is at 150oC, i.e. above the glass transition; it can be seen that the indentations disappeared under the action of temperature. Right is the TASC output co-plotted with DSC data; the offset of the DSC transition coincides with the onset of the TASC onset.[/caption]

The TASC algorithm consists of first parameterising a designated structure, locating this structure within a selected area and quantifying the extent to which it conforms to its original form; in this way it tracks how the structure changes over time. If the sample moves, as often happens when a sample is heated, the algorithm compensates for this so that only structural changes are quantified. One type of experiment that employs this approach is to impose an indentation on the surface of a sample and measure how it changes with temperature. This is illustrated in Figure 1, which shows images taken with a hot stage microscope of four indentations that were imposed on the surface of a sample of filled polystyrene. As the temperature of the sample was increased to above its glass transition temperature the indentations disappeared due to the action of surface tension when the material became fluid. Also in Figure 1 is an example of the TASC output for one of the indentations; it is co-plotted with a DSC measurement for the same sample. The onset of the TASC measurement corresponds to the offset of the change in heat capacity measured by the DSC. As might be expected, the two measurements probe different molecular motions. In the case of TASC cooperative motions leading to polymer chains moving over one another have to take place before the indentation starts to relax to form a smooth surface. Calorimetry is sensitive to in-chain moieties acquiring additional degrees of freedom. These two methods are, therefore, highly complementary.

[caption id="attachment_49700" align="aligncenter" width="600"] Figure 2: An example of local thermal analysis using TASC. The optical micrograph of the sample suggests two regions of different appearance and possibly, therefore, different compositions. The points A and B at which the indentations were made are indicted in the image. The TASC results clearly demonstrate two different transition temperatures. It follows that there are compositional or structural differences between the two domains.[/caption]

TASC can be used for local thermal analysis because an indentation can be made anywhere on the surface of a sample; this is illustrated in Figure 2. The micrograph indicates two phases and thus some form of analysis is needed to identify them. By making an indentation in each phase then heating the sample TASC determined that the materials were different and the transition temperatures confirmed that A) was PCL and B) was polystyrene. This capability to locate and identify phases within a sample is not available with any mainstream thermal analysis technique.

[caption id="attachment_49701" align="aligncenter" width="400"] Figure 3: The application of TASC to particles of polystyrene (PS), high density polyethylene (HDPE) and polypropylene (PP). The change in appearance in each case is shown above the graph. In all cases a transition temperature is clearly delineated by TASC both for the glass transition (PS) and melting (HDPE and PP).[/caption]

In the examples given above TASC was used to follow how an imposed structure (an indentation) changed with temperature; Figure 3 gives examples of following changes in the pre-existing structure of polymer particles in order to determine the relevant transition temperatures. Even in this case it is possible to measure a glass transition as well as melting points. Again, DSC could determine that these materials were present but would not be able to identify the location and characteristics, such as size distribution, of each material.

[caption id="attachment_49703" align="aligncenter" width="600"] Figure 4: The application of TASC to the surface of a foodstuff is shown. Left; the white box shows the outer sugar-based layer. The TASC plot shows a transition on the red surface consistent with that of Carnauba wax (melting range 82-86oC).[/caption]

The plane of focus in a transparent sample can be adjusted to lie within the sample so TASC does not always have to be a surface technique, however, typically it is a surface analysis method and this can be exploited to look specifically at coatings in complex samples. Figure 4 illustrates the analysis of the surface of a foodstuff. The TASC plot shows a low temperature transition that corresponds to the melting of the Carnauba wax that covers the surface of the outer sugar-based coating. Using a conventional thermomechanical analyser might well show a transition but whether it came specifically from the surface or from somewhere within the sample would not, in general, be clear.

[caption id="attachment_49704" align="aligncenter" width="600"] Figure 5: An example of following the dissolution of crystals in water. At the start (the left image) there are a number of crystals and two are marked with coloured squares and the dissolution of these crystals is tracked and displayed in the graph below using this colour coding. The red crystal, though large, disappears quickly and is dissolved by 200 sec. (the middle image). By 400 sec. (the right image) there has been a substantial reduction in the size of the blue crystal (and the other crystals) though it is not completely dissolved.[/caption]

Figure 5 shows micrographs of a set of crystals in a DSC pan surrounded by water as the solvent, a heating rate of 10ºC/min. was used to ensure dissolution did not take too long. The dissolution process was followed for two crystals using the TASC algorithm as shown in the graphs below the images. It is the case that the rate of dissolution expressed as a fraction will be lower for larger crystals than smaller ones of the same material i.e. larger crystals take longer to completely dissolve than smaller ones. It is clear from just these two results that there are at least two materials present because the larger crystal dissolved faster than the smaller one.

[caption id="attachment_49705" align="aligncenter" width="600"] Figure 6: A line was fitted to the TASC data for each crystal in the image shown in Figure 4 and c) above. The slope of the line is a measure of the rate of dissolution. These were then plotted against the surface area of each crystal to obtain b). It is expected that the dissolution rate is a monotonic function of crystal size though scatter is expected because of the irregularities of their shape. Two classes of behaviour are clearly seen; the points are plotted in red and green to indicate the different groups. Each crystal can then be assigned to its group and this is shown in c).[/caption]

In Figure 6 an analysis of all of the crystals entirely within the field of view of the microscope is shown. To characterise the dissolution kinetics a linear equation was fitted to each plot up to 0.8 of the normalised value, see Figure 5a; this is a measure of the rate of dissolution. The slopes were plotted against the surface area of each crystal to obtain Figure 5b. As stated above, this measure of the rate of dissolution is related to the size of each crystal and it is expected that the rate should be a monotonic function of an appropriate measure of their sizes. We see by inspection two markedly different behaviours. It is clear that there are two categories of materials as distinguished by their dissolution kinetics, one of these is shown in red the other in green. The Crystals in each category can then be assigned within the original image; this is shown in in 5c. Chemical analysis of the solvent and residue identified the ‘green’ crystals as sucrose and the ‘red’ crystals as salicylic acid.

In conclusion; TASC is showing itself to be a versatile tool that has a very wide range of applications. It can measure changes in any structure that can be seen by the particular imaging technique that is being used. Using it with optical microscopy creates a simple but powerful way of measuring transition temperatures so that hot stage microscopy becomes a general thermal analysis tool with unique capabilities not available with conventional mainstream thermal analysis methods. These include the ability to perform local thermal analysis, analyse individual particles in a mixture and provide surface analysis. In addition it is a convenient tool for measuring transition in pure materials. TASC also provides a unique way of characterising the dissolution behaviour of materials at the level of individual particles. Where there are two or more different types of kinetics within a mixture this can be detected. For these types of experiment temperature control is essential and temperature programming can be used to shorten the time of the experiment and to investigate the effects of temperature on dissolution kinetics. By combining dissolution analysis with chemical analysis of the solutes and residues, materials with different compositions and physical states can be mapped.

References:

1. Weidemann H.G, Felder-Casagranda S., Thermomicroscopy, in Brown M.E. editor, Handbook of Thermal Analysis and Calorimetry, Elsevier, 1998. Chapter 10.
2. Vitez I.M., Newman A.W., Thermal Microscopy: in Reading M., Craig Q.M., editors, Thermal Analysis of Pharmaceuticals, CRC Press, 2007. Chapter 7
3. New Methods of Thermal Analysis and Chemical Mapping on a Micro and a Nano Scale by Combining Microscopy with Image Analysis, M.Reading, M. Morton, M. Antonijevic, D. Grandy, D. Hourston and A. Lacey, Microscopy Advances in Scientific Research and Education, Ed. A. Mendez-Vilas, Formatex Research Centre, September 2014, Vol. 2, pp. 1083-1089

The author:

Professor Mike Reading invented the algorithm named TASC (Thermal Analysis for Structural Characterisation). Linkam, which manufactures hot stages for optical microscopes, has now incorporated TASC into one of its products.