Tribovoltaic effect in metal-semiconductor couplings: design and optimization of the set-up and investigation of the Al-Si tribo-pair.

Triboelectric nanogenerators operate on the mechanisms of contact electrification and friction. Most previous studies focused on alternate voltage scenarios, consequently needing rectification prior to utilization. On the other hand, tribovoltaic nanogenerators (TVNGs) produce DC voltage outputs on many material couplings. On the basis of the most reliable theories, a TVNG operates on the mechanism of electron-hole pairs created as a result of the energy released during sliding. In literature, various material couplings have been investigated to understand the mechanisms, but until now the details on the electronic changes occurring at the atomic and molecular level, as well as their relation with the energy produced during friction, are still debated. Thus, there is the need to further investigate this topic to develop a robust method for the exploitation of the phenomenon. One of the important part of this thesis is related to the design and optimization of the experimental set-up. A standard tribometer has been modified to collect in real-time the tribovoltaic signal generated in sliding contact during reciprocating motion. For this purpose, the existing reciprocating ball-on-plate module has been combined with a circuit consisting of a low noise pre-amplifier and a digital oscilloscope, connected to the moving sample and the static counterpart. In particular, a dynamic Schottky junction is obtained by sliding an Aluminum ball against a Silicon substrate 2 mm thick (Al-Si pair), varying contact pressure, sliding frequency and humidity. Since the tribological properties of the mentioned pair are expected to be strictly connected to the outcoming tribovoltage, friction and wear of the Si substrate and of the Al ball have been estimated, and compared with the corresponding generated voltage. In the initial tests, the friction coefficient (CoF) exhibits a clear evolution through three distinct regimes, due to the progressive wear and the subsequent transfer of material between the mating surfaces: running-in, low-friction, and high-friction. These regimes are mostly unaffected by tribological parameters, except humidity, which lowers CoF at low levels. The measured voltage signal shows a less distinct variation than CoF throughout the whole test, with average value almost stable around 0.2 mV. It was observed that the tribovoltaic measurement dramatically depends on the wiring set-up, thus different procedures have been developed to optimize the acquisition procedure and the data analysis. To deepen this phenomenon, the tribovoltaic data have been combined with the study of the wear by means of topographical (profilometry, optical imaging, SEM) and chemical analyses (EDX) of the counterparts. The topographical analysis shows variations in the wear rate exclusively at higher normal loads, while the chemical analysis confirms the presence of material transferred between the two counterparts. Despite this, no clear correlations are pointed out within the limits of the accuracy offered by the current system.