Laser-induced breakdown spectroscopy (LIBS) is a powerful and popular analytical atomic spectroscopy technique. The technique is most often used to analyse solid samples, although it is compatible with gases and liquids as well. However, the direct and sensitive LIBS analysis of (bulk) liquid samples is still challenging, due to multiple problems associated with liquids hindering focusing, plasma formation and light collection, which altogether result in high detection limits, poor reproducibility, and the need for using quite high laser fluences [1]. Some approaches aiming at tackling or circumventing these problems have been already presented in the LIBS literature (mostly based on liquid-to-solid conversion), but they are either not sensitive enough or are not really practical [2-4]. The goal of the present project is to develop such nano- or microstructured sample holders that would allow the direct and sensitive LIBS analysis of microliter volumes of aqueous solutions. Our core idea is that such structured sample holders would now only allow for a better laser-substrate coupling, thus better utilization of the pulse energy, but their solid material could also serve as “sacrificial” targets. The latter condition is important as the optical breakdown threshold of solids is known to be orders of magnitude lower than that of liquids. Thus, finely structured sample holders could theoretically facilitate plasma ignition which could then also atomize a nearby, microscopic volume liquid sample. We employ two approaches for the preparation of the sample holders. One is the use of commercially available clean, nanoporous materials, whereas the other one is to use micrometer-resolution 3D-printing to fabricate plastic structures (Figure 1.). We employ numerical simulations and experimental methods to test various geometrical patterns, porosities as well as the wettability of the materials. Various microscopy techniques are also utilized to follow the laser ablation behaviour of these sample holders. Our LIBS analytical results clearly show that both of these approaches can produce sample holders that facilitate the practical, direct, reproducible LIBS analysis of 1-10 µL aqueous solutions with trace level (ppm or lower) limits of detection.