Today, a variety of devices for dry powder inhalers (DPIs) is available and many different formulations for optimized deposition in the lung are developed. However, during the production of powder inhalers, processing steps may induce changes to both, the carrier and active pharmaceutical ingredients (APIs). It is well known that standard pharmaceutical operations may lead to structural changes, crystal defects and amorphous regions. Especially operations such as milling, blending and even sieving generate these effects. These disorders may induce re-crystallization and particle size changes post-production which have a huge influence on drug delivery and product stability.In this study, pilot tests with a polar solvent (water) and hydrophilic drug (Salbutamol sulfate) were performed to receive a first impression on further possible implementation of hydrophobic samples with organic solvents. Thereafter, a reliable method for the accurate detection of low amounts of amorphous content is described up to a limit of quantification (LOQ) of 0.5% for a hydrophobic model API (Ciclesonide). The organic vapor sorption method which is a gravimetric method quantifies exactly these low amounts of amorphous content in the hydrophobic powder once the suitable solvent (isopropanol), the correct p/p0 value (0.1) and the exact temperature (25°C) have been found. Afterward it was possible to quantitate low amorphous amounts in jet-milled powders (0.5–17.0%).In summary, the data of the study led to a clearer understanding in what quantity amorphous parts were generated in single production steps and how variable these parts behave to fully crystalline material. Nevertheless it showed how difficult it was to re-crystallize hydrophobic material with water vapor over a short period. For the individual samples it was possible to determine the single humidity at which the material starts to re-crystallize, the behavior against different nonpolar solvents and the calculation of the reduction of the glass transition temperature (Tg) according to the Gordon–Taylor equation.