Friday, 18 December 2020 14:08

Investigating the complex phase behavior and drug release of ternary solid dispersions based on ethyl cellulose

Written by Melissa Everaerts, Lennert Cools and Guy Van den Mooter
Rate this item
(0 votes)

In this work, the insoluble hydrophobic polymer ethyl cellulose (EC) was investigated as a potential carrier for the formulation of amorphous solid dispersions (ASDs). Insoluble carriers have recently gained more and more attention due to their unique drug release mechanism which is based on a diffusion process [1,2]. As a result of this diffusion process, a more gradual build-up towards supersaturation is created compared to ASDs based on a water-soluble polymer and ideally, the ASDs based on an insoluble carrier are able to maintain supersaturation for a longer period of time.

Figure 1: Composition of the ternary ASDs based on EC as an insoluble carrier. Focus of this study was to assess the effect of the miscibility and the ratio of EC/PIA on the pharmaceutical performance of the ASDs. Indomethacin was used as a model drug.


EC is a well-known polymer in the pharmaceutical industry for coating applications, but often requires an additional component to increase its porosity [3]. Next to that, little is known on how altering the porosity of an insoluble carrier system can affect the drug release from an ASD. Hence an additional water-soluble polymer, defined as a porosity increasing agent (PIA), was included in the EC-based ASDs. Whether or not a PIA is miscible with EC can affect the porosity of the formulation upon contact with water and therefore also the drug release (Figure 1). In the case of a miscible system, drug molecules will be distributed in a homogeneous polymer network of PIA and EC while in the case an immiscible system, discrete regions of PIA will be formed. For this reason, assessment of the miscibility of various PIA with EC was considered as a crucial part for this study, prior to the formulation of ternary ASDs. For the ternary ASDs, various ratios of EC/PIA were prepared to assess if there was an optimum in the composition of the polymer blend for the drug release (Figure 1).

Miscibility study and solid state of ASDs

Miscibility study of EC and PIA

Modulated differential scanning calorimetry (mDSC) was used to investigate the miscibility of the following PIA with EC: hydroxypropyl cellulose (HPC), polyvinylpyrrolidone /vinyl acetate (PVP/VA), polyvinylpyrrolidone (PVP) K12 and PVP K25 (Figure 2). For the first three systems, it was difficult to draw any conclusions with mDSC since HPC displayed no glass transition temperature (Tg) and the Tgs of pure EC, PVP/VA and PVP K12 were located so close to each other, that it was difficult to distinguish if the polymer blends demonstrated one or two Tgs. EC/PVP K25 showed two Tgs for the blends of 5 to 50% PVP K25 and this partial miscibility was also previously described in literature by Dereymaker et al [4]. However, caution is needed with this interpretation since EC also displayed an endothermic event around the Tg of PVP K25.

Figure 2: Reversing heat flow of polymer blends EC/HPC, EC/PVP VA, EC/PVP K12 and EC/PVP K25. The endothermic event of EC is indicated with an arrow. Legend from top to bottom (m% PIA): green 0% PIA, cyan 5% PIA, blue 10% PIA, black 20% PIA, grey 50% PIA, dark blue 80% PIA, violet 90% PIA, purple 95% PIA and maroon 100% PIA.

Solid state of binary and ternary ASDs with EC

After the miscibility study, binary and ternary ASDs were prepared via spray drying with indomethacin (IND) as a model drug and PVP K12 or K25 as a PIA. The drug loading was set at 30% (m%) for all formulations while the ratio of EC/PVP was varied from 0/100 to 100/0 (m/m). All ASDs had a single Tg, meaning that a miscible system was obtained based on mDSC. The IND/EC/PVP K12 ASDs demonstrated an exothermic event in all nine formulations (Figure 3). For the formulations with a high EC/PVP ratio, this exothermic event can be attributed to cold crystallization of EC since it was also observed in pure EC. However, the formulations which consisted of a higher amount of PVP K12 or only consisted of PVP K12 as a carrier showed an exothermic event as well at a higer temperature. This observation was considered to be peculiar since PVP K12 is not a semicrystalline polymer [5] and to our knowledge does not form ordered regions upon heating. The investigation of the phase behavior of these ternary ASDs is still ongoing and remains one of the major challenges of this research. Nevertheless, it is clear that when working with ternary ASDs, the elucidation of the phase behavior becomes more complex compared to binary systems.

Figure 3: Total heat flow of binary and ternary ASDs based on IND/EC/PVP K12. Left figure (a) shows the formulations with an EC/PVP K12 ratio between 100/0 to 50/50 (m/m). Legend from top to bottom for the ratio of EC/PVP K12: green 100/0, cyan 95/5, blue 90/10, black 80/20 and grey 50/50. The right figure (b) shows EC/PVP K12 formulations with a ratio between 50/50 to 0/100 (m/m). Legend from top to bottom for the ratio of EC/PVP K12: grey 50/50, dark blue 20/80 PIA, violet 10/90, purple 5/95 and maroon 0/100.

Matrix structure and drug release

The formation of a porous network after the release of PVP was confirmed for both PIA with scanning electron microscopy (SEM) (Figure 4). After the investigation of the porous network, drug release was monitored for both binary and ternary ASDs in acidic medium which was adjusted to a neutral pH after one hour (Figure 5). When the ratio of EC/PVP K12 (m/m) was decreased from 100/0 to 80/20, the release of IND improved greatly. However, when the ratio was further decreased towards 20/80, drug release was significantly reduced and performed even worse than both binary ASDs. This phenomenon was not expected as our hypothesis was that more water-soluble PVP would create more free space for the drug to diffuse through and therefore lead to a better drug release. A possible explanation for this trend is the occurrence of an interaction between EC and PVP, but this is currently still under investigation. A similar trend was observed for the ternary systems with PVP K25. In addition to that, ASDs with EC/PVP K25 (m/m) ratios between 95/5 and 80/20 showed a slower release compared to the EC/PVP K12 systems with the same ratio. The opposite observation was done for EC/PVP K25 ASDs with ratios between 80/20 and 5/95. For those formulations the drug release was also limited, but not to the same extent as for the ASDs of EC/PVP K12.

Figure 4: SEM images of spray dried a) EC/PVP K12 (20/80, m/m) and b) EC/PVP K25 (20/80, m/m) after dissolution in water. A porous network was created after the PVP had diffused out of the EC matrix.

Figure 5: Drug release of binary and ternary ASDs of IND/EC/PVP K12. Left figure (a) shows the formulations with an EC/PVP K12 ratio of 100/0 to 50/50 (m/m). Legend from top to bottom for the ratio of EC/PVP K12: green 100/0, blue 95/5, black 90/10, cyan 80/20 and grey 50/50. The right figure (b) shows the release of all ASDs with an EC/PVP K12 ratio of 20/80 to 0/100 (m/m). Legend from top to bottom for the ratio of EC/PVP K12: green 20/80, blue 10/90, black 5/95, cyan 0/100. The pH value of the medium is indicated above each graph.

Conclusion and challenges

The use of EC as a hydrophobic insoluble carrier in combination with a PIA seems to be a promising strategy for the formulation of ASDs with a delayed yet sufficient drug release as demonstrated by the SEM images and release results. However, some challenges were faced with the miscibility study and the analysis of the solid state behavior. First of all, mDSC appeared to be insufficient as stand-alone technique to investigate the miscibility. Next to that, working with a ternary ASD system makes the elucidation of the phase behavior more complex compared to a binary system. Finally, it is clear that by adjusting the PIA and the ratio of EC/PIA drug release can easily be altered, but the factors that influence the release mechanism next to diffusion such as a possible interaction between EC and PIA, distribution of the pores, wettability and local viscosity created by the release of the PIA, are not yet fully understood.


  1. Sun, D.D.; Lee, P.I. Probing the mechanisms of drug release from amorphous solid dispersions in medium-soluble and medium-insoluble carriers. J. Control. Release 2015, 211, 85–93.
  2. Everaerts, M.; Van den Mooter, G. Complex amorphous solid dispersions based on poly (2-hydroxyethyl methacrylate): study of drug release from a hydrophilic insoluble polymeric carrier in the presence and absence of a porosity increasing agent. Int. J. Pharm. 2019, 566, 77–88.
  3. Ye, Z.; Rombout, P.; Remon, J.P.; Vervaet, C.; Van den Mooter, G. Correlation between the permeability of metoprolol tartrate through plasticized isolated ethylcellulose / hydroxypropyl methylcellulose films and drug release from reservoir pellets. Eur. J. Pharm. Biopharm. 2007, 67, 485–490.
  4. Dereymaker, A.; Scurr, D.J.; Steer, E.D.; Roberts, C.J.; Van den Mooter, G. Controlling the release of indomethacin from glass solutions layered with a rate controlling membrane using fluid-bed processing. Part 1: Surface and cross-sectional chemical analysis. Mol. Pharm. 2017, 14, 959–973.
  5. Shi, N.Q.; Lei, Y.S.; Song, L.M.; Yao, J.; Zhang, X.B.; Wang, X.L. Impact of amorphous and semicrystalline polymers on the dissolution and crystallization inhibition of pioglitazone solid dispersions. Powder Technol. 2013, 247, 211–221.



SEM images were recorded by Dr. P. Baatsen of the Department of Neurosciences KULeuven Chemistry (VIB – KU Leuven, Leuven, Belgium).



Melissa Everaerts 1, Lennert Cools 1 and Guy Van den Mooter 1

Drug Delivery and Disposition, KU Leuven, Department of Pharmaceutical and Pharmacological Sciences, Campus Gasthuisberg ON2, Herestraat 49 b921, 3000 Leuven, Belgium (This email address is being protected from spambots. You need JavaScript enabled to view it.This email address is being protected from spambots. You need JavaScript enabled to view it.)

Read 301 times