The direct writing technology known as ion-beam-induced deposition (IBID) has been attracting attention mainly because of its high degree of flexibility of locally prototyping three-dimensional (3D) nanostructures. These high-resolution nanostructures have various research applications. However, no systematic study of the capability of IBID to fabricate 3D nanostructures has been published to date. This is partly caused by the lack of suitable methods to monitor and to access the numerous time-varying process parameters and our lacking overview of the interplay between the relevant parameters. This thesis partially aims to fill this gap. This thesis mainly includes three parts: (1) Exploration of the limits of IBID to fabricate nanopillars. Firstly to fabricate IBID pillars in a controllable and reproducible manner, we have studied the optimization of the pillar growth conditions. With the conventional Ga+ FIB and the novel He+ FIB approaches, the influence of precursor surface density and of the ion beam interaction have been investigated, respectively. Moreover, relevant simulation work is discussed to explain the interplay between vertical and lateral growth and their dependence on precursor depletion and replenishment. Combining these results, a comparison between Ga+ and He+ IBID pillar growth is made. Secondly, to improve the quality of IBID pillars, we have studied the formation of the irregular sidewall surface and the halo viz. the deposits around the bases of a typical Ga+ IBID pillar by comparing pillars grown on either an insulating Si3 N4 membrane or on a semiconducting Si wafer. Thirdly, by changing the substrate properties and the distance between neighboring pillars, we have studied the proximity effect in IBID pillar growth. This proximity effect is important when fabricating dense pillar arrays. The proximity effect of He+ IBID is similar to that of Ga+ IBID, though the trend is much less pronounced. (2) Exploration of the limits of IBID to fabricate nanopores in thin membranes. We have demonstrated that sub-10-nm-diameter nanopores in a Si3 N4 membrane can be fabricated in a single Ga+ IBID step by carefully adjusting the ion beam and gas exposure conditions. This is accomplished by exploiting the competition between sputtering and deposition processes during IBID. Apart from the simplicity and the speed, another advantage is a broad choice of material for the deposit and the membrane. At various stages of pore formation we have studied the chemical composition and the shape of the pore, which are the factors that determine the functionalization of the nanopores. For this purpose, energy dispersive x-ray (EDX), electron energy loss spectroscopy (EELS) analysis have been used for determining the chemical composition, and 3D electron tomography for determining the shape of the pore. It is found that the chemical structure in the rim of the pore depends on the properties of the precursor gas. Furthermore, simulation shows that the forward and the backward sputtering depend differently on membrane thickness. This difference can also play a role in the pore formation and shrinkage. (3) Study of the IBID process mechanisms. We have done a series of experiments to distinguish the roles of different mechanisms involved in IBID. Firstly we have found a significant contribution of secondary particles to Ga+ IBID. This result was obtained by comparing the volume of a deposited box with that of the material deposited onto a nearby sidewall. Subsequently we have investigated two models that describe IBID in terms of the impact of secondary electrons and of sputtered atoms, respectively. For this purpose, the yields of deposition, sputtering, and secondary electron emission as well as the energy spectra of the secondary electrons were measured in situ during Ga+ IBID as functions of ion incident angle and energy. The results indicate that the sputtered atom model describes Ga+ IBID better than the secondary electron model. I also briefly discuss the contribution of primary ions. Based on these results, we review the studies on the mechanisms of IBID with Ga+ or He+ ion beams and EBID mechanisms reported in the literature. I conclude that IBID has to be described by multiple mechanisms. The dominating mechanism is in Ga+ IBID related to sputtering, while in He+ IBID and EBID to secondary electron emission. In this thesis work, we have studied the capability of IBID to grow 3D nanostructures. Future efforts, for instance improvement of the purity of the deposits, will be necessary to functionalize IBID nanostructures.