[en] Highlights: • B20 and diesel exhibit similar spray tip penetration and angle. • Change in orientation of spray shapes observed with different fuels. • B100 shows poor air fuel mixing compared to B20 and diesel. • Diesel shows higher equivalence ratio compared to B20 and B100. - Abstract: In this study, the fuel spray characteristics and air-fuel mixing process of waste cooking oil biodiesel (B100) and its blend with diesel (B20) were investigated and compared with diesel fuel. Spray characteristics such as spray tip penetration, spray angle, spray velocity and spray morphology were investigated under high injection and ambient pressure conditions using a constant volume spray chamber. The air-fuel mixing process was analysed using empirical relations like fuel volume, mass of air entrained within the spray and equivalence ratio. The results shows that B100 has higher spray tip penetration and velocity but narrow spray angles due to high viscosity and large momentum possessed by B100 compared to B20 and diesel fuels. The deviation in spray tip penetration reduces under high ambient pressure. The spray angle shows no change under various injection pressures; however it increases significantly under high ambient pressure. The spray shape is affected by the cavitation inside the injector nozzle holes. The fuel volume and amount of air entrainment within the spray showed that B100 exhibits poor air-fuel mixing compared to B20 and diesel fuels. Nevertheless, the equivalence ratio along the axial direction of spray reveals that the B100 has lean equivalence ratio compared to B20 and diesel fuel due to the presence of inherent oxygen content in its structure. A numerical simulation was conducted using new hybrid spray model implemented in KIVA4 and found that the results obtained from the simulation were in good agreement with the empirical results calculated from the experiments

[en] Highlights: • A compact multi-component skeletal reaction mechanism was developed. • Combined bio-diesel and PRF mechanism was proposed. • The mechanism consists of 68 species and 183 reactions. • Well validated against ignition delay times, flame speed and engine results. - Abstract: A new coupled bio-diesel surrogate and primary reference fuel (PRF) oxidation skeletal mechanism has been developed. The bio-diesel surrogate sub-mechanism consists of oxidation sub-mechanisms of Methyl decanoate (MD), Methyl 9-decenoate (MD9D) and n-Heptane fuel components. The MD and MD9D are chosen to represent the saturated and unsaturated methyl esters respectively in bio-diesel fuels. Then, a reduced iso-Octane oxidation sub-mechanism is added to the bio-diesel surrogate sub-mechanism. Then, all the sub-mechanisms are integrated to a reduced C_2–C_3 mechanism, detailed H_2/CO/C_1 mechanism and reduced NO_x mechanism based on decoupling methodology. The final mechanism consisted of 68 species and 183 reactions. The mechanism was well validated with shock-tube ignition delay times, laminar flame speed and 3D engine simulations.

[en] Highlights: • Effect of injection rate-shaping on heat-release is significant with less turbulence. • Two peak heat-releases are seen for the shallow-depth re-entrant piston. • Significant combustion phasing occurs with kerosene usage and high turbulence. - Abstract: In this work, the combustion characteristics of a direct injection compression ignition (DICI) engine fueled with kerosene-diesel blends, using different piston bowl geometries together with varying injection rate-shapes were investigated. A total of three combustion bowl geometries, namely the omega combustion chamber (OCC), the shallow-depth combustion chamber (SCC) and the shallow-depth re-entrant combustion chamber (SRCC), were used together with six different ramp injection rate-shapes and pure diesel, kerosene-diesel and pure kerosene fuels. It is seen that the SRCC geometry, which has the shortest throat length, gives the highest turbulence kinetic energy (TKE) and this resulted in two peak heat-releases, with a primary peak heat-release during the premixed combustion phase and a secondary peak heat-release during the mixing-controlled combustion phase. In addition, the SCC geometry gives rather distinct premixed combustion and mixing-controlled combustion phases due to the fact that combustion is predominantly controlled by the injected fuel spray itself because of less turbulence. Also, when kerosene is used in place of diesel, the heat-release during the premixed combustion phase increases and diminishes during the mixing-controlled and late combustion phases. It is interesting to note that the effect of injection rate-shaping on the heat-release rate is more obvious for bowl geometries that generate less TKE. Moreover, bowl geometries that generate higher TKEs as well as fuels with lower viscosities generally give lower carbon monoxide (CO) emissions and higher nitrogen oxide (NO) emissions. More importantly, it is possible to achieve low NO and CO emissions simultaneously by using the appropriate bowl geometry, injection rate-shape and fuel, although a slight decrease in power is inevitable.

[en] Highlights: • An approach is used to develop a robust kerosene–diesel reaction mechanism. • Ignition delay of the kerosene sub-mechanism is well validated with experiments. • The kerosene sub-mechanism reproduces the flame lift-off lengths of Jet-A reasonably well. • The kerosene sub-mechanism performs reasonably well under engine conditions. - Abstract: The use of kerosene fuels in internal combustion engines is getting more widespread. The North Atlantic Treaty Organization military is pushing for the use of a single fuel on the battlefield in order to reduce logistical issues. Moreover, in some countries, fuel adulteration is a serious matter where kerosene is blended with diesel and used in diesel engines. So far, most investigations done regarding the use of kerosene fuels in diesel engines are experimental and there is negligible simulation work done in this area possibly because of the lack of a robust and compact kerosene reaction mechanism. This work focuses on the development of a small but reliable kerosene–diesel reaction mechanism, suitable to be used for diesel engine simulations. The new kerosene–diesel reaction mechanism consists only of 48 species and 152 reactions. Furthermore, the kerosene sub-mechanism in this new mechanism is well validated for its ignition delay times and has proven to replicate kerosene combustion well in a constant volume combustion chamber and an optical engine. Overall, this new kerosene–diesel reaction mechanism is proven to be robust and practical for diesel engine simulations.

[en] Highlights: • Thermo-physical properties of liquid DME and DEE are reported. • Ether fuels tend to cavitate higher compared to that of diesel fuel. • Spray tip penetration and SMD are found to be lesser for ether fuels. • Ether fuels shows excellent atomization behavior. - Abstract: In this work, the spray characteristics of ether fuels such as dimethyl ether (DME) and diethyl ether (DEE) have been numerically investigated using KIVA-4 CFD code. A new hybrid spray model developed by coupling the standard KHRT model to cavitation sub model was used. The detailed thermo-physical properties of ether fuels have been predicted and validated with experimental results available from literature. The cavitation inception inside the injector nozzle hole has been studied for ether fuels in comparison with diesel fuel. It was found that ether fuels cavitates higher compared to that of conventional diesel fuel because of its low viscosity. The spray tip penetration of diesel fuel was longer than that of ether fuels due to high viscosity and density of diesel fuel. Ether fuels characterized by low Ohnesorge number and high Reynolds number showed better atomization behavior compared to that of the diesel fuel.

[en] Highlights: • A Polyoxymethylene Dimethyl Ether 3 reaction mechanism is proposed. • The proposed mechanism is highly compact (61 species and 190 reactions). • The major reaction pathway consists of only 11 species and 17 reactions. • Negative temperature coefficient behavior is successfully captured. • Well validated against ignition delay, flame speed, species and engine results. -- Abstract: With high oxygen content, Polyoxymethylene Dimethyl Ether 3 (PODE₃) is a potential fuel additive to reduce soot emissions. However, reaction mechanisms to describe PODE₃ combustion are not yet compact enough for 3-D numerical simulations. Therefore, the current work aims to develop a small yet reliable PODE₃ reaction mechanism. Based on sensitivity analysis, the major reaction pathway is identified to construct the PODE₃ sub-mechanism. Thereafter, it is integrated with a primary reference fuel (PRF) mechanism to create a PRF-PODE₃ mechanism containing 61 species and 190 reactions. The major reaction pathway of the PODE₃ sub-mechanism consists of only 11 species and 17 reactions. Furthermore, the new mechanism has been well validated with experimental results in terms of ignition delay times (rapid compression machine at pressures = 1.0 MPa and 1.5 MPa, equivalence ratios = 0.5, 1.0 and 1.5), laminar flame speeds (P_in = 1 atm, T_in = 408 K), flame species concentrations (pressure = 33.3 mbar, equivalence ratio = 1) and homogeneous charge compression ignition (HCCI) combustion (equivalence ratios = 0.18 and 0.34). Overall, this highly compact yet robust PRF-PODE₃ mechanism is demonstrated to be suitable for internal combustion engine simulations. In addition, with good agreement in terms of fundamental combustion validations, the proposed PRF-PODE₃ mechanism can reasonably be applied to other practical applications such as simulations in jet engines, pulse detonation engines, boilers and furnaces.

[en] Highlights: • Engine combustion was numerical studied with a quad-component biodiesel mechanism. • Less soot exhausted from biodiesel fuel engine combustion than that fueled with diesel fuel. • Biodiesel fuel with higher fraction of unsaturated FAMEs produces more soot precursor. - Abstract: In this study, numerical analysis of fuel structures on engine soot particles’ mass and size were done by CFD combustion modelling using diesel and different levels of unsaturated biodiesel fuels through the KIVA4-CHEMKIN platform. The proposed numerical approach, with a quad-component skeletal mechanism of biodiesel blend surrogates along with a multi-step phenomenological soot particle model, could capture the soot particle characteristics of test fuels with acceptable accuracy under engine combustion conditions. The reduction of exhaust soot from biodiesel combustion, compared to diesel fuel, was attributed to the suppressed soot precursors formation and lower number of particles in total. However, it was concluded that the biodiesel fuel with a higher fraction of unsaturated FAMEs (more double carbon bonds CC) contributed more to the formation of soot precursors, thus producing a higher amount of soot particles in mass and numbers as a consequence of accelerated soot particle nucleation and soot surface growth.

[en] Highlights: • A hybrid model based on the classical CTC model and CHEMKIN model was proposed. • The proposed hybrid model is able to model RCCI combustion with detailed chemistry. • The hybrid model is robust and efficient for RCCI combustion simulations. This study proposed a hybrid model consisting of a characteristic time combustion (CTC) model and a closed reactor model for the combustion modelling with detailed chemistry in RCCI engines. In the light of the basic idea of the CTC model of achieving chemical equilibrium in high temperature, this hybrid model uses the CTC model to solve the species conversion and heat release in the diffusion flame. Except for the diffusion flame, the auto-ignition in RCCI combustion is computed by a closed reactor model with the CHEMKIN library by assuming that the computational cells are closed reactors. The border of the transition between the CTC model and closed reactor model is determined by two criteria, a critical temperature and a critical Damköhler number. On the formulation of this hybrid model, emphasis is placed on coupling detailed chemistry into this hybrid model. A CEQ solver for species equilibrium calculations at certain temperature, pressure was embedded with CTC for detailed chemistry calculation. Then this combustion model was integrated with the CFD framework KIVA4 and the chemical library CHEMKIN-II and validated in a RCCI engine. The predicted in-cylinder pressure and heat release rate (HRR) show a good consistency with the data from the experiment and better accuracy than that computed from the sole closed reactor model. More importantly, it is observed that this model could save computational time compared with closed reactor model due to less stiff ordinary differential equations (ODEs) computation. A sensitivity analysis of the critical temperature and critical Damköhler number was conducted to demonstrate the effect of these two parameters in the current model.

[en] Highlights: • A clustered dynamic adaptive chemistry (CDAC) method is proposed and validated for engine combustion simulations. • Boot injection rate shapes in DICI engines are investigated fuelled with kerosene, diesel and their blending. • The effect of start of injection of boot injection on the combustion and emission characteristics is investigated. • CDAC is able to reduce the computational time by more than 60% while maintaining good accuracy. In this study, we conducted a numerical investigation on the effect of injection timing of boot injection rate shape on the combustion and emission characteristics in a direct injection compression ignition (DICI) engine fueled with kerosene/diesel blending. Considering the complex surrogate in kerosene chemical mechanisms and the huge computational workload in multi-dimensional engine simulations, we employed a clustered dynamic adaptive chemistry method (CDAC) to accelerate the chemistry integration process. This study firstly specified the user-defined parameters in this CDAC method by sensitivity analysis in a HCCI and DICI engine with different user-defined parameter combinations. With these specified parameters, CDAC is then validated by comparing its predicted in-cylinder pressure with the full chemistry ones. It is found that the current CDAC method could reduce the computational time by more than 60% compared with the full chemistry CPU time. CDAC, subsequently, is used to conduct the numerical investigation on the injection timing of boot injection rate shapes. Four different boot injection rate shapes are simulated and compared with the normal rectangular injection. The effect injection timing of the boot injection rate on the engine performance and combustion/emission characteristic is then analyzed in detail. It is found that the change of start of injection (SOI) in boot injection has little influence of the ignition delay in the DICI engine fuelled with diesel and kerosene blending due to the high cetane number of diesel and better volatility of kerosene. In addition, with kerosene addition into the diesel combustion, it is observed that the CO emission could be reduced at all the varied SOI.