As for any solids, microscopy can give valuable information about the state of the hydrate samples. Microscopy is also a very quick, non-invasive analytical method, which should be standard equipment in laboratories investigating the characteristics of solid samples.

Simply looking at the powder under a normal microscope will give information about the particle size, and if the particles are big enough, additional information about the shape and crystal habit can be gathered. Using the characteristic of crystals to rotate the polarisation axis of light, birefringence, it is quite easy to find out whether the sample is crystalline or not. For this the sample is placed between two polarisers, which are rotated 90˚ from each other. This is also called dark-field microscopy. Crystalline samples normally show up as light particles in the dark-field revealing interference colours depending on their thickness. If the sample appears absolutely dark when rotated in the dark-field, it does not exhibit birefringence and is thus most likely amorphous. The birefringence of a crystal, however, depends on its anisotropy and thus certain crystal symmetries can make perfectly crystalline samples show no birefringence. The most well-known examples for this are cubic crystals, which are isotropic in all three dimensions. But also hexagonal and tetragonal crystals show no birefringence when observed along their six- or four-fold axis, respectively. Thus, the birefringence of a sample proves it to be crystalline, while its absence does not disprove it.

Figure 1 Example of a crystal undergoing pseudomorphosis upon heating

A second very simple experiment is to put the sample, e.g. crystals from a solution crystallisation experiment, in suspension of the mother liquor on the microscopic slide and observe the effects of evaporation of the solvent on the crystals. If the observed crystal form is stable under ambient conditions, nothing will happen, while unstable crystal forms normally show a blackening of the particles and a change from a bright single coloured interference in dark-field to an indistinguishably grey interference colour. This change is called pseudomorphosis and normally accompanies crystal form changes and solvent loss from the crystals (Figure 1). Especially unstable hydrates and solvates will show this phenomenon upon solvent loss.

Thermomicroscopy is the temperature controlled variation of normal light microscopy. Most microscopes can be fitted with a hot-stage (Figure 2) and the temperature behaviour of crystalline and amorphous samples can be probed. The hot-stage can be chosen to fit the problem at hand, as different models have different characteristics concerning temperature range, heating rate, sample accessibility, possibility of gas purge, etc.

Figure 2 Microscope fitted with a Linkam hot stage

Heating programs of crystalline samples can show different characteristics in the crystals. Phase transitions are normally observed by either a front going through the crystal or by pseudomorphosis (see above). Desolvation and dehydration events are generally accompanied by pseudomorphosis or an amorphisation of the sample, which can be observed as a complete loss of birefringence and the rounding off of crystal edges. All different crystal forms (unsolvated polymorphs and solvates) can show melting without phase transition. The resulting melt can then either be stable or recrystallise as another crystal form.

Figure 3 Bubble formation of a hydrated sample in paraffin oil during heating

For solvated/hydrated crystal forms, the preparation of the sample in high-boiling point silicon oil gives information about volatile compounds given off during phase transitions. Since desolvation events are the release of solvent from the crystal lattice, these solvents will be observable as bubbles in the surrounding oil. In addition to desolvation events, volatile compounds, such as water or carbondioxide, can be given off during decomposition which will result in bubble formation.

For the observation of the melting point, the stoichiometric ratio of the solvate has to be kept stable, i.e. the desolvation has to be inhibited. The easiest way to do this is a preparation of the sample in a high-viscosity matrix, usually silicon oil. The vapour pressure of any volatile compound has to overcome the viscosity of the oil before bubble formation can occur to transport the volatile component from the sample. In the case of hydrates, the vapour pressure of the incorporated water needed to overcome the viscosity, and thus the dehydration temperature of the hydrate, can be tuned to be only achievable at temperatures higher than the melting point of the crystal form. For very stable hydrates (i.e. very high boiling points) the sample can be embedded in glue.

Microscopy can also be performed under controlled humidity conditions. The normal setup for these experiments comprises a chamber in which the sample is placed. This chamber is then purged with a dry and a wet gas, while by changing the relative ratio of the two adjusts the relative humidity in the cell. A humidity sensor then reports back the actual humidity in the chamber. Interesting applications for this kind of analysis are (amongst others) hydration/dehydration of hydrated crystal forms, desolvation of solvates under higher humidity, moisture induced crystallisation of amorphous samples and liquefaction of highly hygroscopic materials.