Graphene-based materials (GBMs) have sparked intensive research towards their commercial translation, due to their unique properties. GBMs can already be found in commercial products such as tennis rackets and sports shoes, and novel applications such as membranes for water filtration, textiles and inks are close to become reality. Such applications demand for large sheets with lateral dimensions of several micrometres. Considering the expected increase of human exposure in the near future, we hypothesised that large GBMs could pose a new risk to the respiratory system upon inhalation of aerosolised GBMs. Current literature has failed to establish a definitive safety profile of GBMs due to inconsistent reports, mostly due to incomplete material characterisation. Therefore, the main aim of this study was to prepare well-characterised graphene oxide (GO) dispersions with controlled lateral dimensions, in order to compare their effects in pulmonary response. Here, we developed a reproducible method to synthesise 3 types of endotoxin-free GO sheets with controlled lateral dimensions. An extensive panel of physicochemical characterisation techniques identified different lateral size ranges for l-GO (1 â 30 Î¼m), s-GO (50 nm â 2 Î¼m) and us-GO (10 â 300 nm) and confirmed their similar surface properties. Hence, the effects of lateral dimensions could be confidently analysed in vivo. We first studied whether lateral dimensions of GO would play a similar role to length of carbon nanotubes (CNTs). Long CNTs may induce mesothelioma in the pleural and peritoneal cavities, in association with material persistence. Neither l-GO nor s-GO elicited granulomatous response to the mesothelium after intraperitoneal injection. Whole-body SPECT/CT imaging showed that both materials exhibited favourable clearance profiles. We further investigated whether proteins in the administered dispersion could interact with the surface of GO and affect their biological interactions. When dispersed in a protein-free solution (5% dextrose), s-GO sheets enhanced monocyte recruitment to the peritoneal cavity. This study highlighted that multiple parameters may determine the biological response to GBMs, such as biokinetics and surface reactivity. Under the light of the previous results with the peritoneal model, we hypothesised that biokinetics of GO in the respiratory tract could affect the pulmonary response. Using radiolabelled GO, we evidenced a size-dependent biodistribution of GO sheets after intranasal instillation. Nevertheless, the reduced translocation of l-GO to the lungs did not impede its enhanced acute infiltration of interstitial macrophages and inflammatory dendritic cells, which ultimately contributed to the formation of persistent granulomas up to 90 days after exposure, in line with the secretion of Th1/Th17 cytokines. On the other hand, us-GO induced the mildest response, suggesting its safer profile. Finally, the effects of GO nanosheets (s-GO and us-GO) were evaluated in mice infected with influenza A virus. We hypothesised that exposure to GO could exacerbate viral pathology. This was only verified when GO was administered prior to infection, with s-GO eliciting acute interstitial infiltration with neutrophil recruitment. On the other hand, us-GO modulated inflammatory response with transient anti-inflammatory effects. Overall, this thesis highlights the role of lateral dimensions in the pulmonary response to GO. The results obtained herein support the potential application of us-GO as a vector for pulmonary delivery of medicines and biomolecules (e.g. vaccines). Further investigation on the interactions of GO with myeloid cells is required to disclose potential applications in nanomedicine.