The Power of XRPD to Understand Processes of Hydration
PDF Print E-mail

Stoichiometric hydrates are not necessarily considered detrimental to the development of a drug substance but they often exist in different stoichiometric ratios, which may interconvert, making these hydrates more challenging for development.  Insights, at an early stage of development, of the interconversions that may take place between hemi, mono and higher hydrates will direct formulation strategy and avoid delays at a later stage of development.  Although several analytical techniques are available for the characterization of hydrates, the structural characterization by X-ray powder diffraction (XRPD) is considered one of the most informative methods.  XRPD will not only provide insights of the intermolecular interactions between the water molecules and the API but also provide information of the relative stabilities of the different hydrated forms.

4-hydroxybenzamide is an excellent model compound to demonstrate the use of XRPD to understand the processes of hydration.  In the solid state 4-hydroxbenzamide can exist as an anhydrous form and as a hemi and monohydrated crystalline solid.  In 2007 Perlovich (Perlovich, G. L., Hansen, L. K., Volkova, T. V., Mirza, S., Manin, A. N. & Bauer-Brandt, A. 2007, Cryst. Growth Dec. 7, 2643) postulated that the monohydrate form may be formed either directly from the anhydrous form or from the hemihydrate as an intermediary.  In the same publication the author proposed that dehydration of the monohydrate to the anhydrate is always achieved via the hemihydrate.  Perlovich used DSC analysis and thermodynamic calculations to support his hypothesis.

At Crystallics we investigated the mechanism underlying the hydration of 4-hydroxybenzamide.  Obtaining 4-hydroxybenzamide monohydrate is not straightforward.  Incubation at 50°C and 85% relative humidity fails to deliver the monohydrate.  Only slurry conversion experiments in water or mixtures of water and organic solvents will result in recrystallization of anhydrous 4-hydroxybenzamide to the pure monohydrate.  DSC analysis of the monohydrate shows that the first thermal event observed at 50°C is the loss of water.  To fully understand the process variable temperature XRPD experiments were conducted.   The variable temperature XRPD analysis, shown in Figure 1, clearly demonstrates that already at 45°C the monohydrate starts to convert into the anhydrous crystalline form of 4-hydroxybenzamide.

Figure 1. XRPD diffractograms of freshly prepared 4-hydroxybenzamide monohydrate (left) and the same material after 10 minutes at 45 C (right).

Rietveld calculations indicate that at this temperature nearly half of the monohydrate converts into the anhydrous form.  Interestingly, the calculations also show that there are no traces of the hemihydrate.  This result suggests that conversion from monohydrate to anhydrous is not achieved via the hemihydrate form.  Analysis of the crystal structure of all three solid forms would shed light on the dehydration process.  The crystal structure of the hemihydrate was not available and Crystallics had to crystallize and refine the structure of the hemihydrate.  With all three structures solved we were able to explain the hydration and dehydration process of 4-hydroxybenzamide.

The explanation of both processes is found in the hydrogen bond pattern.  In the anhydrous crystals the hydrogen atom from the hydroxyl group is bonded to the oxygen atom in the amide moiety.  In both hydrates the water molecules are in between these two functional groups and serve as a bridge.  However, in the monohydrate the bridge is nearly linear while in the hemihydrate the bridge is at an angle.  The angle is significantly less than 180° as illustrated in Figure 2.

Figure 2.  Hydrogen bonds geometry between amide and hydroxy group 4-hydroxybenzamide crystals. Top left presents the monohydrate form, while top right is the hemihydrate and bottom structure is the anhydrous form. The arrows show possible dehydration process of both hydrate forms.

In addition to the hydrogen bond pattern, the density of the crystals is another contributing factor to the dehydration process.  For 4-hydroxybenzamide the density of the anhydrous, monohydrate and hemihydrate is 1.418 g/cm3, 1.381 g/cm3 and 1.368 g/cm3, respectively.  Therefore, the hydrogen bond geometry and the crystal density together explain why the dehydration of the monohydrate goes straight to the anhydrous form.  This observation is in contrast to the calculations made by Perlovich, the lattice energy of hemihydrate is the lowest and, therefore, the hemihydrate should be considered the least stable form of the three.

This example illustrates that analytical techniques such as DSC, and calculations based upon these, can be informative but structural information obtained through XRPD provides a significantly better understanding of solid-state processes.