Most applications are currently powered by compression ignition (CI) engines due to their reliability and superior torque, power output and fuel economy. Unfortunately, CI engines cause social and environmental harms by emitting high levels of pollutants and greenhouse gases (GHGs), thus the use of alternative, zero-carbon fuels (e.g. hydrogen and ammonia) under dual-fuel (DF) combustion in CI engines has recently drawn attention, offering the potential to burn cleaner gaseous fuel at a thermal efficiency comparable to a diesel-only engine but with substantially reduced emissions. The aim of this Thesis is to develop a comprehensive, physically based numerical modelling framework to accurately predict the combustion and emissions characteristics of hydrogen-blended DF combustion. Dual-fuel flames comprise premixed and non-premixed combustions, hence investigating their characteristics in a laminar environment clarifies their coupled nature in DF combustion. This thesis first investigates the one-dimensional laminar hybrid DF flames of various hydrogen fuel blends relevant to DF combustion by setting the conditions so that their combination represents a DF flame. The aim is to better understand the fundamental characteristics of hydrogen-blended laminar DF flames through intensive parametric study to identify the effects of diverse parameters, such as preferential diffusion and elevated pressure, on various hydrogen fuel blends. The results reveal that preferential diffusion effects via hydrogen addition greatly enhance the reaction rate by expanding the concentrations, oxidisations and formations of highly reactive species in one-dimensional laminar flame calculations. The accurate prediction of nitric oxides (NOx) emissions requires implementing a thermal and prompt NOx formation sub-model. The second part of the Thesis is focused on developing a novel hybrid combustion model based on flamelet generated manifold (FGM) incorporating preferential diffusion effects. The model development was achieved by coupling non-premixed and premixed flamelets databases to accurately predict the multistage combustion process in DF technology. The preferential diffusion effects were incorporated using a two-step correction to better capture the auto-ignition process, flame propagation and heat release rate. The hybrid combustion model employs three control variables—mixture fraction, reaction progress variable and enthalpy—and was thoroughly validated against the experimental data of high hydrogen content DF engine combustion. The results demonstrate that the novel hybrid combustion model can capture the multistage processes of hydrogen-blended DF combustion. The final part of the Thesis performs a detailed parametric study to achieve greener DF combustion in a DF combustion engine by using alternative gaseous fuels (ammonia and hydrogen) and liquid fuel, replacing diesel with hydro-treated vegetable-oil (HVO). The results reveal that the improved in-cylinder parameters and thermal efficiency of hydrogen-blended ammonia DF combustion over ammonia DF combustion. The parametric study also shows that HVO can be used as a replacement for diesel pilot fuel in hydrogen-blended DF combustion engines without compromising engine thermal efficiency, demonstrating HVO’s suitability as a clean pilot fuel for hydrogen-blended DF internal combustion engine applications.