467271 Modeling Spray Dried Dispersion: From Droplet to Particle

Friday, November 18, 2016: 9:55 AM
Continental 5 (Hilton San Francisco Union Square)
Jaime Curtis-Fisk1, Priti Jain2, Jodi Mecca3, Bart Rijksen4, Shrikant Khot1 and William Porter III1, (1)Dow Pharma and Food Solutions, The Dow Chemical Co., Midland, MI, (2)Dow Chemical International Private Limited, Mumbai, India, (3)Formulation Science, The Dow Chemical Company, Midland, MI, (4)Dow Benelux BV, Terneuzen, Netherlands

Modeling Spray Dried Dispersion: From Droplet to Particle

Jaime L. Curtis-Fisk1, Priti Jain2, Jodi Mecca1, Bart Rijksen, 3 Shrikant Khot1,  William W. Porter III1
1The Dow Chemical Company, Midland, MI USA

2Dow Chemical International Private Limited, Mumbai, India

3 Dow Benelux BV, Terneuzen, Netherlands


Poorly soluble APIs are often formulated as a solid amorphous dispersion in combination with an excipient polymer to enhance their solubility. A common route to producing the dispersion is through spray drying. This process involves spraying a solution of polymer, active pharmaceutical ingredient (API) and solvent into a heated drying gas where the liquid spray droplets dry into solid particles of the amorphous dispersion.  Hydroxypropyl methylcellulose acetate succinate (HPMCAS), such as AFFINISOL™ HPMCAS, is one polymer shown to significantly enhance API solubility when in an amorphous dispersions.  The purpose of this work is to develop a droplet drying model connecting the fundamental properties of HPMCAS polymer solutions to the spray drying particle formation kinetics and the resulting particles’ morphology and compositional uniformity. With this modeling capability, it may be possible to predict and gain insight into the observed properties of amorphous spray dried dispersions.


Approach and Methods:

A particle formation model was developed considering the balance between solvent evaporation and solute diffusion to predict particle morphology and composition. The model, as illustrated in Figure 1, considers the diffusion of API and excipient towards the core of the droplet and surface recession of the solvent as the two main counteracting mechanisms in the drying of droplets. Methods for predicting particle formation are known, but do not fully address the complexities of producing spray dried dispersions from a solution containing API and polymer.[i] This model framework builds upon these references and addresses the non-ideality of the liquid phase due to the use of polymeric excipients, including the interaction between solutes and the change in evaporation rate due to non-ideal vapour liquid equilibrium. This added complexity will provide a more accurate route to predicting the impact of formulation and spray drying conditions on particle morphology.


To provide inputs to the model, fundamental properties including viscosity and diffusivity of HPMCAS-API solutions were characterized in a wide range of solvents and solvent blends.  Additionally methods to characterize HPMCAS-API interactions were used to understand potential effects on mass transport within the solution droplet. 




Figure 1. The model framework combines process conditions, solution properties, and evaporation rate in an Athena framework to generate predictions of particle morphological properties


This particle formation model was shown to compare well against experimental results in the literature. Figure 2 illustrates a comparison of the model predictions of spray dried trehalose against results from Vehring et al.[ii] The samples evaluated cover a low Peclet number range of 0.44 to 1, with an initial droplet diameter of 20 µm resulting in a dry particle diameter of 1.7 to 1.8 µm. The model accurately predicts the measured dry particle density, which is near the density of amorphous trehalose indicating solid particles.



Figure 2. Comparison of experimental results from Vehring et al and model predictions of trehalose particles formed through spray drying.ii


It was also found to predict well the experimental results from Vehring et al.ii for spray drying of glycoprotein solution as shown in Figure 3.  As the Peclet number increases, the particle diameter increases while the density decreases. Both the experimental and modeled particle densities (0.1 to 0.5 g/cc) are significantly lower than the density of solid glycoprotein. This indicates hollow particles with void space, a result confirmed through microscopy of the generated particles.

Figure 3. Comparison of experimental results from Vehring et al and model predictions of glycoprotein particles formed through spray drying.ii

To apply the droplet drying model to spray drying API-HPMCAS dispersions, fundamental solution properties were characterized.  The viscosities of such solutions were found to vary strongly based on the specific HPMCAS polymer and spray drying solvent.  As shown in Figure 4, at similar concentrations a wide range of viscosities were observed indicating very different polymer configurations in solution. Similar measurements were conducted to quantify the diffusivity of polymer and API solutes in solution and the surface tension of these solutions.  Having quantified these properties, predictions around the droplet drying process and resulting particle properties will be shared.  It will be shown that depending on the solvents selected for spray drying, diffusion controlled or surface recession controlled drying kinetics can occur resulting in very different particle morphologies and compositional uniformity.

Figure 4: Viscosity of Polymer/Solvent Solutions that serve as inputs for particle formation modeling


Experimental analysis of formulation solution properties and polymer chemistry provide a solid framework for building modeling capabilities to predict the formation of solid particles from liquid droplets in the spray dried dispersion process. The two examples presented highlight the dramatically different potential outcomes from a spray drying process, from solid particles of high density to low density hollow particles. The ability to predict the nature of the particles formed based on formulation and process properties provides the formulator with a valuable tool to optimize this process to hone in on a targeted outcome prior to conducting time consuming experiments. Particle morphology can play a critical role in downstream processing and ultimately the application performance. Combining polymer chemistry and formulation expertise with modeling capabilities provides formulators with a streamlined route to product development.






[i] Chen, Xiao Dong; Sidhu, Harvinder; Nelson, Mark; “Theoretical probing of the phenomenon of the formation of the outermost surface layer of a multi-component particle, and the surface chemical composition after the rapid removal of water in spray drying,” Chem. Eng. Sci. 66 (2011) 6375-6384.

[ii] Vehring, Reinhard; Foss, Wiillard R.; Lechuga-Ballesteros, David; “Particle formation in spray drying,” Aerosol Science 38 (2007) 728-746.

Extended Abstract: File Not Uploaded