The potential of amorphous solid dispersions to improve the solubility, dissolution rate and bioavailability of poorly water soluble drugs is well known. However, the number of formulations that have made it through to the market is limited because of the unstable nature of the amorphous form, which often results in recrystallization of the drug with the subsequent loss of the solubility and dissolution advantages. Thus, ensuring the stability constitutes a major challenge in the development of amorphous solid dispersions. 

Published in Research Highlights


The mechanism of plastic deformation in amorphous polymers is still obscure due to lack of experimental techniques to probe molecular mobility during active deformation. Lee et al. showed an increase in a segmental mobility by several decades using a specific active deformation protocol (uniaxial tensile creep) by directly measuring the molecular mobility using an optical photobleaching approach [1]. Our previous studies using ATR-FTIR spectroscopy showed that compression of amorphous solid dispersions containing naproxen and PVP or PVP-VA appeared to affect the drug-polymer interactions, the drug-polymer mixing and the physical stability of the system [2, 3]. Deformation can affect the molecular dynamics of amorphous polymers that include local bonds, segmental and whole chain mobility. It potentially leads to conformational changes and also enhances or affects local bond motions of polymers which may be involved in intermolecular interactions with drugs in solid dispersions. The objective of this study is to understand the effect of compression on the local and segmental mobility of PVP-VA to understand the molecular origin of the differences in drug-polymer interactions between the compressed and the uncompressed solid dispersions. The impact of pharmaceutical tabletting on the molecular dynamics of amorphous solid dispersions can be critical in terms of physical stability of the drug during storage.


High resolution broadband dielectric spectroscopy (ALPHA Analyzer, Novocontrol Technologies) was used to measure the complex dielectric responses (Eqn. \ref{eq:1}, ε' and ε'' are the real and the imaginary part) of the compressed (565 MPa) (comp), the slightly compressed (scomp) and the powder (pow) PVP-VA 64 and powder PVP K25 samples in the frequency range from 10-1 to 106 Hz.

\begin{equation}\varepsilon (\omega)=\varepsilon' (\omega) - i\varepsilon'' (\omega)\label{eq:1} \end{equation}

The “conduction free” dielectric loss was computed from the real part of the dielectric spectra using the Kramers-Kronig relation to suppress the Ohmic conduction interference in the imaginary part of the dielectric response (Eqn. \ref{eq:2}) [4].

\begin{equation}\varepsilon'' (\omega_0)=\frac{\sigma_{dc}}{\varepsilon_\nu\omega_0}+\frac{2}{\pi}\int_0^\inf \varepsilon'(\omega)\frac{\omega_0}{\omega^2-\omega_0^2}\text{d}\omega\label{eq:2} \end{equation}

Generally, the activation energy according to the Arrhenius-law is usually considered a true energetic barrier, however, processes that involve cooperativity might additionally show an entropic “barrier”. The Starkweather analysis allows to separate the activation energy into its enthalpic and entropic part to assess the cooperative nature of the secondary relaxation of PVP-VA (Eqns. \ref{eq:3} and \ref{eq:4}) (for the details of the equations refer to [5]).

\begin{equation}f=\frac{k_B T}{2\pi h}\exp \left( -\frac{\Delta H^{\ddagger}}{RT} \right)\exp\left(\frac{\Delta S^{\ddagger}}{R} \right)\label{eq:3} \end{equation}

\begin{equation}E^{*}_{a}=RT\left( 1+ \ln \frac{k_B}{2\pi h} \ln \frac{T}{f} \right)\label{eq:4} \end{equation}

Results and Discussion

The secondary relaxation of compressed PVP-VA shifts synchronously with the primary relaxation to lower temperature compared to the powder and slightly compressed samples even after annealing in the supercooled liquid state (Fig. 1). The source of the secondary relaxation process is likely originating from the vinyl pyrrolidone moiety of PVP-VA. Thus, compression appeared to enhance the localised motion which likely originates from the conformational transition involving the N-C(H) bond in PVP-VA and also the segmental mobility due to dynamic Tg [6].

The non-zero activation entropy of compressed PVP-VA was higher than the powder sample which indicates the presence of a higher entropic barrier during conformational reorientation for the compressed sample (Table 1). Moreover, it may also indicate a higher cooperative nature of the β relaxation process for the compressed sample as the result of enhanced conformational rotations.

The primary relaxation process starts at a very low temperature for the compressed PVP-VA and an additional wing was also identified with a mean relaxation time smaller than the powder and the slightly compressed PVP-VA (Fig. 2).


Conclusion and Outlook

The molecular dynamics results support the idea that severe compression affects the local and the segmental mobility of PVP-VA. Compression of PVP-VA appeared to lead to a shorter time scale for the secondary (β) relaxation process. Moreover, the primary relaxation process starts at a very low temperature for the compressed PVP-VA and an additional wing was also identified with a mean relaxation time smaller than the powder and the slightly compressed PVP-VA due to the heterogeneity in segmental mobility imparted by compression. The observation can be very useful for drug containing amorphous solid dispersions which are stabilised by specific drug-polymer interactions. The current study on PVP-VA should also be extended to one with binary solid dispersions stabilised by a strong and a weak drug-polymer hydrogen bonding further to probe the effect of compression in the molecular dynamics of both the drug, the polymer and their specific interactions.


[1]    Lee, H. N.; Paeng, K.; Swallen, S. F.; Ediger, M. Science 2009, 323, 231.
[2]    Ayenew, Z.; Paudel, A.; Van den Mooter, G. Eur. J. Pharm. Biopharm. 2012, 81, 207.
[3]    Worku, Z. A.; Aarts, J.; Van den Mooter, G. Mol. Pharmaceutics 2014, 11, 1102.
[4]    Wübbenhorst, M.; van Turnhout, J. J. Non-Cryst. Solids 2002, 305, 40.
[5]    Starkweather Jr, H. W. Polymer 1991, 32, 2443.
[6]    Tonelli, A. Polymer 1982, 23, 676.

PSSRC Facilities

The Research group of Prof. Guy Van den Mooter focuses on the study of the physical chemistry of solid (molecular) dispersions prepared by hot melt extrusion, spray drying, bead coating and spray congealing. It is the aim to correlate the physical structure of the drug-polymer dispersions to their pharmaceutical performance and stability profile, and to correlate formulation and processing parameters to the resulting physical structure. Analytical techniques such as thermal analysis (DSC, MTDSC, TGA, hot-stage microscopy, isothermal microcalorimetry, solution calorimetry), X-ray powder diffraction, infrared spectroscopy, solid state NMR, broadband dielectric spectroscopy and in vitro (intrinsic) dissolution testing are being used for this purpose. Other (solid state) analytical techniques that are available are (powder) rheology, He-pycnometry, instrumented compression testing, SEM, TEM, coulter counter and Laser diffraction.

As part of the Center for Drug Delivery and Analysis of KU Leuven, this research group is also involved in formulation development and preformulation studies (a.o. study of polymorphism) for pharmaceutical companies.

Published in Research Highlights
Saturday, 07 September 2013 08:01

Surface characterization of solid dispersions


Solid dispersions are an intensively investigated enabling technology to formulate poorly soluble drugs. Many contributions already studied their higher solubility and resulting dissolution rate as well as the challenges at the level of physical stability due to their high intrinsic energy. Whereas the vast majority of these studies focus on the bulk characteristics of the samples, we are convinced that the (often distinct) properties of the sample surface should not be overlooked.

Published in Research Highlights