Nanoclusters, aggregates of several to hundreds of atoms, have been one of the central issues of nanomaterials sciences owing to their unique structures and properties, which could be found neither in nanoparticles with several nanometer diameters nor in organometallic complexes. Along with the chemical nature of each element, properties of nanoclusters change dramatically with size parameters, making nanoclusters strong potential candidates for future tailor-made materials; these nanoclusters are expected to have attractive properties such as redox activity, catalysis, and magnetism. Alloying of nanoclusters additionally gives designer functionality by fine control of their electronic structures in addition to size parameters. Among binary nanoclusters, binary cage superatoms (BCSs) composed of transition metal (M) encapsulating silicon cages, M@Si16, have unique cage structures of 16 silicon atoms, which have not been found in elemental silicon nanoclusters, organosilicon compounds, and silicon based clathrates. The unique composition of these BCSs originates from the simultaneous satisfaction of geometric and electronic shell-closings in terms of cage geometry and valence electron filling, where a total of 68 valence electrons occupy the superatomic orbitals of (1S)2(1P)6(1D)10(1F)14(2S)2(1G)18(2P)6(2D)10 for M = group 4 elements in neutral ground state. The most important issue for M@Si16 BCSs is fine-tuning of their characters by replacement of the central metal atoms, M, based on one-by-one adjustment of valence electron counts in the same structure framework of Si16 cage; the replacement of M yields a series of M@Si16 BCSs, based on their superatomic characteristics. So far, despite these unique features probed in the gas-phase molecular beam and predicted by quantum chemical calculations, M@Si16 have not yet been isolated. In this Account, we have focused on recent advances in synthesis and characterizations of M@Si16 BCSs (M = Ti and Ta). A series of M@Si16 BCSs (M = groups 3 to 5)was found in gas-phase molecular beam experiments by photoelectron spectroscopy and mass spectrometry: formation of halogen-, rare-gas-, and alkali-like superatomswas identified through one-by-one tuning of number of total valence electrons. Toward future functional materials in the solid state, we have developed an intensive, size-selected nanocluster source based on high-power impulse magnetron sputtering coupled with a mass spectrometer and a soft-landing apparatus. With scanning probe microscopy and photoelectron spectroscopy, the structure of surface-immobilized BCSs has been elucidated; BCSs can be dispersed in an isolated form using C60 fullerene decoration of the substrate. The intensive nanocluster source also enables the synthesis of BCSs in the 100-mg scale by coupling with a direct liquid-embedded trapping method into organic dispersants, enabling their structure characterization as a highly symmetric "metal-encapsulating tetrahedral silicon-cage" (METS) structure with Frank-Kasper geometry.