1wileyonlinelibrary.com www.particle-journal.com www.MaterialsViews.com Characterization of core-shell type nanoparticles in 3D by transmission electron microscopy (TEM) can be very challenging. Especially when both heavy and light elements coexist within the same nanostructure, artifacts in the 3D reconstruction are often present. A representative example would be a particle comprising an anisotropic metallic (Au) nanoparticle coated with a (mesoporous) silica shell. To obtain a reliable 3D characterization of such an object, a dose-effi cient strategy is proposed to simultaneously acquire high-angle annular dark-fi eld scanning TEM and annular dark-fi eld tilt series for tomography. The 3D reconstruction is further improved by applying an advanced masking and interpolation approach to the acquired data. This new methodology enables us to obtain high-quality reconstructions from which also quantitative information can be extracted. This approach is broadly applicable to investigate hybrid core-shell materials.of an object from its 2D projections, a monotonic relationship between image intensity and mass thickness is required. [ 1 ] When applying electron tomography to crystalline solids, diffraction contrast must therefore be avoided. [ 7 ] Indeed, diffraction contrast depends on the orientation of the object with respect to the electron beam and therefore the projection requirement for electron tomography is not fulfi lled. [ 1 ] As a consequence, artifacts will occur in the fi nal 3D reconstruction. [ 8,9 ] This is one of the main reasons why electron tomography in materials science is mostly based on high angle annular dark fi eld scanning transmission electron microscopy (HAADF-STEM). [ 1,10 ] For each composition, the intensity of the HAADF-STEM images changes monotonically with specimen thickness thus fulfi lling the projection requirement. To obtain 3D reconstructions from HAADF-STEM tilt series, reconstruction algorithms such as the simultaneously iterative reconstruction technique (SIRT), [ 11 ] total variation minimization, [ 12,13 ] and the discrete algebraic reconstruction technique [ 14 ] are currently used. Also for samples that consist of more than one type of element, HAADF-STEM tomography is very valuable since not only morphological but also chemical information can be extracted. [ 10 ] However, for samples in which elements with a high and low atomic number ( Z ) are simultaneously present, data acquisition becomes even more challenging. Since the intensity is strongly dependent on the atomic number, [ 10 ] a high difference in Z in the material leads to a large difference in the collected signal that is not always possible to cover with detector's dynamic range. Therefore, some areas of the image can be either overor undersaturated. This situation is often encountered when investigating core-shell hybrid materials such as nanoparticles encapsulated by a lighter matrix. In this paper, we propose an improved approach to obtain reliable 3D reconstructions for these systems, which requires optim...