High-resolution imaging of organic pharmaceutical crystals by transmission electron microscopy and scanning moiré fringes.

Abstract

SummaryFormulation processing of organic crystalline compounds can have a significant effect on drug properties, such as dissolution rate or tablet strength/hardness. Transmission electron microscopy (TEM) has the potential to resolve the atomic lattice of these crystalline compounds and, for example, identify the defect density on a particular crystal face, provided that the sensitivity of these crystals to irradiation by high‐energy electrons can be overcome. Here, we acquire high‐resolution (HR) lattice images of the compound furosemide using two different methods: low‐dose HRTEM and bright‐field (BF) scanning TEM (STEM) scanning moiré fringes (SMFs). Before acquiring HRTEM images of furosemide, a model system of crocidolite (asbestos) was used to determine the electron flux/fluence limits of low‐dose HR imaging for our scintillator‐based, complementary metal‐oxide semiconductor (CMOS) electron camera by testing a variety of electron flux and total electron fluence regimes. An electron flux of 10 e−/(Å2 s) and total fluence of 10 e−/Å2 was shown to provide sufficient contrast and signal‐to‐noise ratio to resolve 0.30 nm lattice spacings in crocidolite at 300 kV. These parameters were then used to image furosemide which has a critical electron fluence for damage of ≥10 e−/Å2 at 300 kV. The resulting HRTEM image of a furosemide crystal shows only a small portion of the total crystal exhibiting lattice fringes, likely due to irradiation damage during acquisition close to the compound's critical fluence. BF‐STEM SMF images of furosemide were acquired at a lower electron fluence (1.8 e−/Å2), while still indirectly resolving HR details of the (001) lattice. Several different SMFs were observed with minor variations in the size and angle, suggesting strain due to defects within the crystal. Overall BF‐STEM SMFs appear to be more useful than BF‐STEM or HRTEM (with a CMOS camera) for imaging the crystal lattice of very beam‐sensitive materials since a lower electron fluence is required to reveal the lattice. BF‐STEM SMFs may thus prove useful in improving the understanding of crystallization pathways in organic compounds, degradation in pharmaceutical formulations and the effect of defects on the dissolution rate of different crystal faces. Further work is, however, required to quantitatively determine properties such as the defect density or the amount of relative strain from a BF‐STEM SMF image.