Diesel engines are known for their excellent efficiency and are therefore used in a variety of applications. However, they are also one of the main sources of hazardous emissions such as nitrogen oxides (NO_x) and smoke. Water-in-Diesel Emulsion (WiDE) is an alternative fuel that can possibly reduce some of the pollutant emissions without compromising engine performance. The surfactant formulation for WiDE usually follows the one used in water-in-oil (w/o) emulsions, where low hydrophilic–lipophilic balance (HLB) emulsifiers are preferred for better solubility in the diesel phase and stabilization at storage temperatures. However, by using a hydrophilic blend with a non-ionic surfactant, it is possible to develop an optimized formulation at higher fuel temperatures, which occur during an engine’s operating condition, achieving possibly higher benefits. Across the different speeds, the results for the emulsion show 7.57% mean improvement in specific fuel consumption (SFC), 19.14% mean improvement in thermal efficiency (TE), 5.54% mean reduction in carbon dioxide (CO₂), 20.50% mean reduction in nitric oxide (NO) and 75.19% mean reduction in smoke levels. However, carbon monoxide (CO) and hydrocarbons (HC) emissions were higher, with a mean increase of 81.09% and 93.83%, respectively.

The goal of the present study is to implement an algorithm capable of designing linear aerospike contours and develop a numerical investigation of the altitude adaptive behavior and performance of high enthalpy flow expanded with such contoured nozzle. The contour was designed using Angelino,s rapid method, targeting an exit Mach number of 4.16. An inviscid model and viscous model (employing the realizable k- $$\epsilon$$ϵ for turbulence) were considered and the results compared, showing that turbulent effects must be taken into account to properly simulate important flowfield aspects such as static pressure, static temperature, entropy and viscous losses. The specific impulse is used as the performance criterion and it was observed that it increased with altitude, with only a 2 $$\%$$% decrease from an ideal bell nozzle for an underexpanding operation, situation where a nearly axial exhaust plume is obtained. Conversely, when turbulent effects are taken into account, overexpanded conditions have the lowest specific impulse, being lower than a quasi-1D analysis and the inviscid case. The findings summit the advantages of the aerospike nozzle over conventional nozzles. Conventional nozzles not only typically struggle in underexpansion scenarios, but also have high side-loads in overexpansion, due to flow detaching from the nozzle’s wall. A further conclusion is that the realizable k- $$\epsilon$$ϵ turbulent model proved a reliable and fast tool to evaluate turbulent effects that characterize the aerospike flowfield.