Sérgio Silva

	Submitted: May 17, 2019. Accepted: June 13, 2019

Within the past few years, much has been learned about characterizing the particles used in inhalation therapies, with the goal of improving drug delivery to the lungs. Optimizing the manufacturing of these therapies requires avoiding particle interactions during size reducing, blending, and capsule filling, because these can affect the final product’s quality. 

	In an inhalation formulation, the physicochemical properties of size-reduced drug particles strongly affect the end product’s stability and performance. Previous studies reported that size-reduced APIs that had been milled using different techniques may present significant differences in terms of morphological and interfacial properties (1,2). Characterizing the particles’ surface properties is key to understanding API/excipient interactions and their impact on the final formulation performance. 

	Selecting the micronization technique is crucial for particles that are to be used in inhalation therapies, because it will determine the API’s properties. For inhalation delivery, a narrow and controlled particle size distribution (PSD) is key to improving and consistently delivering the aerodynamic performance. With a narrow PSD and Dv90<5µm, the fraction of particles that reach the lungs (FPF) will be higher.

	These interactions depend on physicochemical characteristics of the particles, such as morphology, electrostatic charge, contact area, surface energy, carrier surface roughness, and relative humidity. Thus, the characterization of the particles’ surface properties becomes the key to understanding their surface-based phenomena, offering insights into interactiveforces and adhesion affecting the API, carrier and device. Several studies (2,3,4) have shown the importance of the particle’s properties to the efficacy of the formulation and its aerodynamic performance.

	The objective of this work was to analyze and compare particle attributes (e.g., size, morphology, polymorphism, and surface properties) of two different corticosteroids that are intended for inhalation formulations. The materials were processed by both wet and dry milling. Different size reduction approaches were evaluated.

Micronization is a high energy-driven process that may induce changes in the crystallinity of materials and form amorphous domains on the particles’ surface. These amorphous domains, even when present in very small amounts on the particles’ surface, have a significant impact on the material’s physicochemical nature. In addition, they may affect particle-particle interactions.

	It is hypothesized that the milling process changes the orientation of molecules on the surface of the powder particles and changes surface energy (2). Additionally, the amorphous regions of the crystal tend to be converted into a more energetically stable crystalline form. The recrystallization of amorphous domains of the particles often leads to particle size growth and agglomeration as a result of molecular rearrangement phenomena occurring at the surface of the particles and, over time, can strongly affect the aerodynamic performance of a pharmaceutical formulation.

 Jet milling (JM)is widely used in the pharmaceutical industry and remains the leading technology for particle size reduction and to obtain powders within the inhalation size range. In JM, the powder to be size-reduced is fed into a milling chamber where compressed air or nitrogen, usually in a vortex motion, promotes particle-particle collisions.

	Particle classification is made by inertia, following reduction via impaction and abrasion. JM is a solvent-free and cost-effective technique capable of yielding an appropriate PSD for simple inhalation applications. During JM process, however, large quantities of energy are employed that may lead to undesired morphological changes, amorphization, and/or conversion in a different crystalline form of the APIs, often requiring post-conditioning steps.  The conditions under which the particles are processed to attain lower PSD influence particle morphology and the surface energy of the products. 

 Wet polishing (WP) is an alternative wet milling technique that circumvents the limitations associated with JM and is capable of generating stable crystalline material. Unlike JM, this technique uses an appropriate anti-solvent to produce a suspension during the milling phase that is later removed during a subsequent drying step (e.g., spray drying).

	The WP process is easily scalable, more reproducible than JM, and offers a much higher control over PSD, enabling finer PSD customization with narrower spans. Unlike JM micronization, which requires higher levels of energy and may result in a polymorphic form changing, WP requires less energy and maintains the polymorphic form, and, thus, is often considered preferable to JM.

	To determine and compare the effects of various micronization processes, several batches of two APIs that are widely used in inhalation therapies, were produced by wet and dry milling.

	The following methods were used to evaluate the impact of size reduction techniques on the particles’ properties:

		Scanning electron microscopy was used to study morphological properties 

		X-ray powder diffraction was used to characterize polymorphic forms

		Dynamic vapor sorption (DVS), Brunauer-Emmett-Teller (BET) methods, and PSD were used to study physical properties.

	PSD was measured using a laser diffraction analyzer (Malvern, Mastersizer 2000); morphology was studied via the scanning electronic microscope (Phenom ProX); specific area was determined using the BET method (Micromeritics, TriStar II 3020); water sorption value was measured using Dynamic Vapor Sorption (Surface Measurement Systems, DVS Intrinsic), and polymorphism was analyzed using a x-ray powder diffraction device (PANalytical, X’Pert PRO).

presents SEM images of particles that had been previously micronized by jet milling and by wet polishing for the two APIs evaluated. After micronization, the particles exhibited different morphologies based on technique employed. For both APIs, the particles obtained by wet polishing exhibited a smoother surface and a higher degree of homogeneity than those obtained by jet milling. After the micronization process, API-1 particles had a plate-shaped appearance, where API 2 particles were of an irregular, rounded shape.

, both APIs, regardless of the micronization technique employed, exhibited a high degree of crystallinity and maintained the same correspondent polymorphic form. Nonetheless, for several APIs processed via JM a post-production conditioning step was usually needed to convert the amorphous regions and to achieve a stable crystalline form.

 presents the BET and PSD results for both APIs. BET values for API-1 were similar, regardless of the milling technique employed. In contrast, for API-2, the BET value was higher for the particles milled by JM than it was for those milled by WP. In addition, the BET values were higher for API-2 than they were for API-1. These findings can be explained by the particle morphology shown in 

			Table I: Brunauer-Emmett-Teller (BET) and particle size distribution (PSD) of API-1 and API-2.			

	For API-1 (for both JM and WP particles), a smooth surface translated into a lower BET surface area. In contrast, for API-2, the particles processed by JM exhibited a rougher surface that translates into a higher BET value compared to the API particles micronized by WP, which presented a smoother and more polished surface and, consequently, a lower BET value.

	Regarding the PSD, results showed that it is possible to use both methods to produce particles within the inhalation range. However, WP showed a number of clear benefits when compared with JM: not only can it enable a smaller PSD, it reduces variability between batches. 

 shows the typical variability obtained by JM and WP processes when using the same process conditions. The standard deviation for the Dv90 of JM process is 0.32, while, for the WP process, it is 0.10. For highly dependent PSD formulations, therefore, WP presents a clear advantage to JM.

 show that, in the case of API-1, both samples behaved similarly during the analysis: they gained and released water reversibly without the formation of a hydrate form or the crystallization of an important amount of amorphous phase. 

			Table II: Dynamic vapor sorption (DVS) results of API-1 and API-2.			

	In the case of API-2, the maximum sorption value (at 90%RH) for both samples was quite similar (0.19%w/w for WP and 0.17%/w/w for JM). During the desorption phase, however, the sample processed by JM lost approximately all the water it had gained, while the sample processed by WP retained part of the water gained. This difference is not significant, though, and both samples are classified as non-hygroscopic.

	The particle size-reduction method chosen to prepare inhalable drugs can strongly affect the PSD, particle morphology, and surface area of the formulation. The choice of micronization method should depend on the critical quality attributes that are desired for the particles in the given formulation.

	WP enables a finer level of control, reducing variability in the micronized API’s physical properties. This results in more uniform material, improving both batch homogeneity and batch-to-batch consistency. These API properties, in turn, can result in better final product performance (i.e., a formulation that delivers more API to the lungs) and a more stable formulation.  

, 19(3), pp 239–245 (2002).
	2. A. Boshhiha, N.A. Urbanetz, 

3 (10) pp. 8-15 (2018).
	3. D. Williams, “Particle Engineering in Pharmaceutical Solids Processing: Surface Energy Considerations,”

Chapter 21, pp.2677-2694 (2015).
	4. V. N. P. Le, T. H. Hoang Thi, et al., 

Articles

Research compared the properties of corticosteroids processed by wet and dry milling for use in inhalation formulations.