Two main abnormal combustions are observed in spark-ignition engines: knock and low-speed pre-ignition. Controlling these abnormal processes requires understanding how auto-ignition is triggered at the “hot spot” but also how it propagates inside the combustion chamber. The original theory regarding the auto-ignition propagation modes was defined by Zeldovich and developed by Bradley who highlighted different modes by considering various hot spot characteristics and thermodynamic conditions around the hot spot. Two dimensionless parameters (ε, ξ) were then defined to classify these modes and a so-called detonation peninsula was obtained for H2-CO-air mixtures. Similar simulations as those performed by Bradley et al. are undertaken to check the relevancy of the original detonation peninsula when considering realistic fuels used in modern gasoline engines. First, chemical kinetics calculations in homogeneous reactor are performed to determine the auto-ignition delay time τi, and the excitation time τe of E10-air mixtures in various conditions (calculations for a RON 95 TRF surrogate with 42.8% isooctane, 13.7% n-heptane, 43.5% toluene, and using the LLNL kinetic mechanism considering 1388 species and 5935 reactions). Results point out that H2-CO-air mixtures are much more reactive than E10-air mixtures featuring much lower excitation times τe. The resulting maximal hot spot reactivity ε is thus limited which also restrains the use of the detonation peninsula for the analysis of practical occurrences of auto-ignition in gasoline engines. The tabulated (τi, τe) values are then used to perform 1D LES of auto-ignition propagation considering different hot spots and thermodynamic conditions around them. The detailed analysis of the coupling conditions between the reaction and pressure waves shows thus that the different propagation modes can appear with gasoline and that the original detonation peninsula can be reproduced, confirming for the first time that the propagation mode can be well defined by the two non-dimensional parameters for more realistic fuels.
Downsized spark ignition (SI) engines running under high loads have become more and more attractive for car manufacturers because of their increased thermal efficiency and lower CO2 emissions. However, the occurrence of abnormal combustions promoted by the thermodynamic conditions encountered in such engines limits their practical operating range, especially in high efficiency and low fuel consumption regions. One of the main abnormal combustion is knock, which corresponds to an autoignition of end gases during the flame propagation initiated by the spark plug. Knock generates pressure waves which can have long term damages on the engine. The aim for car manufacturers is to better understand and predict knock appearance. However an experimental study of such recurrent but non-cyclic phenomena is very complex, and these difficulties motivate the use of CFD for better understanding them.
In the present paper, Large-Eddy Simulation (LES) is used as it is able to represent the instantaneous engine behavior and thus to quantitatively capture cyclic variability and knock. The proposed study focuses on the LES analysis of knock for a direct injection SI engine. A spark timing sweep available in the experimental database is simulated, and 15 LES cycles were performed for each spark timing. Wall temperatures, which are a first order parameter for knock prediction, are obtained using a conjugate heat transfer study. This study shows that LES is able to describe the in-cylinder pressure envelope whatever the spark timing, even if the sample of LES cycles is limited compared to the 500 cycles recorded in the engine test bench. The influence of direct injection and equivalence ratio stratifications on combustion is also analyzed. Finally, focusing on knock, a MAPO (Maximum Amplitude Pressure Oscillation) analysis is conducted for both experimental and numerical pressure traces pointing out that LES well reproduces experimental knock tendencies.
Cycle simulation plays increasingly important roles on either conceptual designs or operating parametric optimization in development of internal combustion engines. Knock is one of the major constraints in improving spark-ignition (SI) engines, and determination of the knock limited spark advance (KLSA) is very crucial for a successful simulation. In the experimental calibration of a spark ignition engine, KLSA is usually defined as a spark angle where slight knock may occur with a statistical intensity (i.e. maximum amplitude of pressure oscillation, MAPO) distribution. However, the state-of-the-art knock models are mostly limited to prediction of onset of auto-ignition of end gas and fail to predict the stochastic nature of knock, causing deviation of the predicted KLSA from the experimental result. In this paper, a phenomenological model for knock onset, taking multiple variables including the pressure, temperature, excess air ratio and EGR ratio into accounts, is firstly formulated and validated against the experimental results of an SI engine. Based on statistical analysis of the experimental data with the spark sweep around the KLSA, the knock intensities are found to follow the lognormal distribution, and then a knock factor (KF) using the ratio of knock intensities corresponding to 95% and 25% probabilities in the cumulative lognormal function is proposed to determine the KLSA for the engine cycle simulation. To obtain the expectation and standard deviation of the lognormal distribution function, a two-factor model, which considers several factors, including the hot spot size, energy density and heat release rate in the end gas, is developed. Finally, the newly developed knock model is evaluated and better performance is testified, compared to the other knock models used in the engine cycle simulation.
Soot emissions from internal combustion (IC) engines represent a major challenge to engine manufactures with ever most stringent emission regulations, not only in soot mass yielded but also in soot particle number. For example, a particulate number (PN) standard has been introduced in 2011 with Euro 5b for diesel engines and in 2014 with Euro 6 for petrol engines (a limit of 6x1011/km). Soot models provide a detailed insight into soot evolution processes and are thus an essential tool in today's advanced engine designs. Therefore, continuous efforts are made to develop more physically-based engine soot models and improve the prediction accuracy. The primary objective of this work is to identify and demonstrate the critical parameters for accurate soot predictions in IC engine applications using the high-fidelity detailed soot model from an engineering point of view. A detailed soot model based on sectional method was used to solve the soot process in diesel and spark ignition direct injection (SIDI) gasoline engines. A series of sensitivity analyses were carried out to evaluate the importance and significance of wall boundary conditions, wall film formation and vaporization, multi-component fuel surrogate, and soot transport process in engine exhaust on soot predictions. The predicted results were compared in details to engine-out measurements in terms of soot mass, number density and size distributions under various operating conditions. The model results demonstrate that the correct description of the spray-wall interaction and wall film vaporization, as well as the soot transport processes in full engine cycle, is critical for achieving reliable predictive capabilities in engine simulations, especially for SIDI gasoline engines. The findings should help engineers in this field for more accurate soot predictions in engine simulations.
Diesel spray combustion is prone to lead to high soot and nitrogen oxides (NOx) emissions. Regulations impose lowering the emission limits drastically because of the severe impact on health and environment. Apart from health and environmental issues, soot production also results in an efficiency loss due to the heat loss through radiation and late oxidation of the soot particles. Thermal NOx and soot modeling for practical engine applications is challenging. In this study, a two-equation multi-step phenomenological (MSP) soot model is fully coupled with the gas phase chemistry to solve the 1D detailed flame (flamelet) equations for the so-called Spray-A conditions from the Engine Combustion Network (https://ecn.sandia.gov/). Two different soot kinetics models are used that either utilize C2H2 or PAH as the soot-precursor. For the gas phase, a chemical reaction mechanism of n-dodecane comprising of 253 species and 1437 reactions is utilized. The resulting flamelet database includes emission information, NOx and soot carbon concentrations, and the soot particle number density. Detailed flamelet results are inspected in the mixture fraction (Z) domain to characterize the soot formation and oxidation, and NOx formation. The other important controlling variable, the progress variable (PV), is used for the tabulation of the flamelet results. The definition of PV is optimized for key parameters such as soot and NOx. Furthermore, the detailed solutions are stored in a low-dimensional manifold, i.e. a Flamelet Generated Manifold (FGM), with respect to the controlling variables (Z and PV). Here, the FGM tabulation is extended with soot and NOx emissions for non-premixed combustion. Finally, FGM is coupled with CFD and validated by studying an igniting diesel spray and comparing the simulation results with the Spray-A experiments.
A modified 2-D Flamelet model was further developed and validated under various engine operating conditions. It was extended from 2-D flamelet model (C. Hasse, 2004) by simplifying the calculation procedures in extremely rich and lean region which can reduce the CPU time from the original one. Additionally, the collapsing method (C. Felsch, 2009) was introduced, and the model was extended to quadruple injection by applying the collapsing method twice. The simulation cases cover multiple injection strategies including triple and quadruple injection with different EGR rate, and the simulations were carried out by a commercial light-duty diesel engine with n-heptane skeletal chemical mechanism of which considers 29 species and 52 reactions. Simulation results show that the model can capture auto-ignition, mass and heat transfer of each fuel stream, in-cylinder pressure, heat release rate and NOx emissions under multiple fuel stream conditions. Particularly, the model could analyze the combustion process by the interaction of multiple fuel stream, which gives important information on the emissions reduction by multiple injection strategies. Based on the model, the effect on the advanced injection strategies to the reduction of emissions was quantitatively investigated. The quantitatively investigated results could be specified by analyzing how the change of injection timing, duration and the number of injection event can affect the solution of governing equation and corresponding mixture fraction domain.
It is much more important these days to utilize CAE techniques to shorten an engine development period and reduce manpower. These advantages surely bring competitiveness for developing new engines. The authors have been developing a diesel combustion calculation code based on KIVA-Waseda. So far, the calculation code is already equipped with the Hiroyasu-NSC(Nagle and Strickland-Constable) model for soot emission and the extended Zeldovich mechanism for NO(nitrogen monoxide) emission. Then aiming to utilize the calculation code on engine adaptation works to meet exhaust emission standards, following four items were studied. The first was that the detailed reaction model was modified in order to improve ignition delay estimations. The second was that the chemical reaction calculation speed was increased with two types of unique algorithm. The third was that a calculation system with auto-calculation software and analysis software was prepared to obtain NOx-soot trade-off curves easily, and the fourth was that a prompt-NO model was added to the detailed chemical reaction model to improve NO estimation at low power and high EGR operation conditions. The details of these four studies are described and a series of the calculation measures to obtain a NOx-soot trade-off curve is shown in this paper.
By front-loading of the conventional vehicle testing to engine test bench or even further forward to off-line simulations, it is possible to assess a comprehensive variation of powertrain design parameters and testing maneuvers in the early development stages. This entails a substantial cost reduction compared to physical vehicle testing, and hence an optimization of the modern powertrain development process. This approach is often referred to as Road-to-Rig-to-Desktop (R2R2D). To demonstrate the potential of this R2R2D methodology as a seamless development process, a crankangle resolved real-time engine model for a turbocharged gasoline engine was built with the simulation tool GT-Power®. The model was calibrated with measurement data from an engine test bench and integrated into a vehicle co-simulation, which also includes a dual clutch transmission, the chassis, the environment and the automated driver. The most relevant functions of the engine and the transmission control systems were implemented in a Simulink®-based software control unit. To verify the engine model in the transient vehicle simulation, two 900 s time windows from a 2 hours Real Driving Emission (RDE) test, representing urban and motorway conditions, are simulated using the developed co-simulation platform. The simulation results are compared with the respective vehicle measurement data. The fuel consumption deviation caused by the combustion engine model is within 5 %. The system transient behavior and the dominant engine operation points could be predicted with a satisfying accuracy.
Recently, internal combustion (IC) engine systems for automobiles have been required to improve the whole efficiency in a real world. Above all, the turbocharged engine system is attracting attentions. Therefore, it is quite important to consider efficiencies of machine elements such as a turbocharger as well as those related to the combustion process. However, the methods for estimating thermal and mechanical losses of the turbocharger separately and precisely have not been established. In this research, to propose mathematical model capable of predicting mechanical loss induced in a turbocharger, we started with deriving governing equation of the friction loss by a journal bearing and a thrust bearing under operating condition and finally compared with reported data to validate the proposed method. In addition, sensitivity analysis based on the proposed model is performed to investigate the influence of individual physical factors such as rotational speed of the turbocharger, lubrication oil temperature, flow rate and the thrust force on the friction work.
Combustion noise has become an aspect of great importance for engine manufacturers due to its impact on the customers' comfort and health. Furthermore, the increasingly important contribution of turbocharger noise in downsized engines aggravates this problem due to the increment of the compression ratios required. Limitations on the experimental techniques have led to develop numerical methodologies for analyzing and understanding both noise sources and their effects on the acoustic field. However, the complex phenomena associated to the noise generation compromise the accuracy of these approaches, and represent a challenge in the field of the Computational Fluid Dynamics (CFD) modelling. This paper presents a CFD methodology for assessing Diesel combustion as a noise source in compression ignition engines. The model is validated by simulating a steady operation condition at medium speed and medium load, and positive results are obtained in both temporal and frequency domains. Moreover, it allows predicting the main combustion related parameters and the metrics associated to the external acoustic field as well. The simulation results shed some light on the most dominant processes in the noise generation and its propagation inside the combustion chamber. The paper also includes a sensitivity study of the main operation settings (split between pilot injections, exhaust gas recirculation, intake and injection pressure) obtained by using a Design of Experiments (DoE) approach. In addition to determining the most convenient operation strategy, this technique is extremely useful to establish cause/effect relationships between the inputs (combustion settings) and outputs (noise source). Results show that by increasing the split between the two pilots and decreasing the gap between the second pilot and the main injection, it is possible to reduce the noise emissions, while improving the indicated efficiency. Additionally, the increase of the intake pressure improves the traditional trade-off trend between combustion noise and efficiency.
Wall wetting phenomenon in direct-injection spark-ignition (DISI) engine has been demonstrated to increase both the fuel consumption and soot emissions. The drawbacks of impinging spray become more severe under cold start conditions. However, the accuracy of the existing fuel impingement model is limited under cold condition, as the current models were only validated under high and room temperature. Therefore, this paper presents a parametric study of the spray impingement model with an aim to improve its accuracy under cold conditions. Experimental validation was performed under ultra-low plate temperature to simulate the cold start condition in cold areas. The spray structure was characterized with 2D particle image velocimetry (PIV) system. Also, laser induced fluorescence (LIF) technique was employed to measure the thickness of fuel film deposited on the impinging plate. Simulations were performed using a commercial software CONVERGE. The free spray pattern, velocity distribution of the spray and the wall film thickness are compared qualitatively and quantitatively to assess the current impingement model.
For the design of fuel-efficient reciprocating engines, it is necessary to estimate heat transfer loss from the cylinder under flow pulsation. However, conventional turbulence models based on a standard wall function and constant turbulent Prandtl number assumption for turbulent heat flux, which are often used for parametric study in engine design, may not be able to take into account the effect of pulsating flows on unsteady wall heat transfer. In this study, DNS of pulsating turbulent flows and heat transfer on the flat plate was conducted for validation of the conventional models and to check the applicability of the standard wall function and analogy between momentum transfer and heat transfer. The results indicated that prediction capability of heat transfer may be deteriorated by using the standard wall-function and the analogy between momentum and heat transfer.
Heat flux sensors for internal combustion engines were developed with MEMS technologies. Firstly, a plug shape MEMS heat flux sensor was produced as prototype, and its performance was tested in an open chamber and a rapid compression and expansion machine. The sensor measured heat fluxes with noise of 23.7 kW/m2 at a sampling frequency of 25 kHz and endured a harsh environment with pressure of 9 MPa and heat flux load of 26 MW/m2. Then, a new MEMS sensor for measuring heat flux vector was developed. This sensor has three thin film RTDs of a square 300 μm on a side in rotational symmetry, and these RTDs measure wall surface temperatures simultaneously in order to detect heat fluxes of not only a vertical direction component against the sensor surface but also a planer direction one. In the measurement test in the open chamber, the sensor could measure the heat flux vector reflecting a gas flow with noise of 8.3-9.0 kW/m2 at a sampling frequency of 10 kHz.
Simultaneous measurements of the cylinder wall temperature at different locations are required to elucidate the heat transfer mechanisms on the cylinder wall of IC engines. In this paper, we present the development of a flexible wireless wall temperature sensor and its performance evaluation using a 4-stroke engine with a glass cylinder. The sensor, corresponding to an LCR resonant circuit, is attached on the inner surface of the cylinder, and its coil is inductively coupled with an external read-out coil that is located outside of the cylinder to measure the resonant frequency wirelessly. A Cu-laminated-polyimide film is used as the substrate of the sensor, on which a 1 mm-wide sensing resistor is sputtered. The resonant frequency is measured in a steady state from the room temperature to 200 °C, showing the mean sensitivity of 6.22 kHz/°C. Instantaneous measurement synchronized with the piston movement with a time interval of 1.2 ms is carried out for an optical engine operated at around 700 rpm. Change of permeability with the piston motion is compensated, and the crank angle-resolved cylinder temperature measurement has been made. Near the ignition timing, temperature increase by 44 °C is observed, which corresponds to the oncoming flame to the wall. Preliminary simultaneous multipoint measurement is also made with three sensors on the engine cylinder.
Recently, reduction of wall heat flux in internal combustion engine has been recognized as one of the key issues to improve thermal efficiency. For maximizing thermal efficiency by suppressing wall heat transfer under the wide range of engine operation, comprehensive understanding and modeling of wall heat transfer are definitely necessary. In general, fully-developed turbulent flow is assumed in existing wall heat transfer models (wall function) used in CFD. However, there are a few studies reported that undeveloped flow was dominated in engine combustion chamber. In undeveloped flow, turbulent Reynolds number that represents the relationship between turbulent production and dissipation is varied in wall boundary layer according to flow condition. This behavior impacts on wall heat transfer. In this study, velocity distribution in wall boundary layer and wall heat flux were measured under engine-like condition with RCEM (Rapid Compression and Expansion Machine). From the experiment, we revealed that it was important for predicting wall heat flux without increasing calculation cost to take into account the difference of spatially averaged turbulent Reynolds number. Considering this finding, the new model was formulated introducing the ratio of turbulent Reynolds number to that in fully-developed turbulent flow. As a result, it was found that our new model was able to predict wall heat flux in internal combustion engine quantitatively with high accuracy.
To develop higher energy-efficiency and lower emission internal combustion (IC) engines, it is important to understand heat loss characteristics and flame-wall interactions since heat transfer through the wall during combustion greatly affects the energy-efficiency and combustion products. Heat loss from the wall depends significantly on velocity boundary layer of IC engine, which has not been fully understood yet. In this study, a micro particle image velocimetry (PIV) has been performed to investigate tumble enhanced flow characteristics near piston top surface of a motored IC engine for three inlet valve timing (-30, -15, 0 crank angle degrees (CAD)). PIV was conducted at 340, 350 and 360 CAD of the end of the compression stroke at the constant motored speed of 2000 rpm. The measurement region is 3.2 mm x 1.5 mm on the piston top including central axis of the cylinder. The spatial resolution of PIV defined by the interrogation region is about 75 micrometers and the vector spacing is about 37.5 micrometers. The first velocity vector is located about 60 micrometers from the piston top surface. The high spatial resolution PIV revealed that the mean flow near the piston top is not close to the turbulent boundary layer, and rather has tendency of the Blasius theorem, whereas turbulent intensity near the wall is not low. This tendency is considered to be possible from the viewpoints of the shortage of flow time and the laminarization due to the adverse pressure gradient through the compression stroke. This result shows that revision of a wall heat transfer model based on an assumption of the proper characteristics of flow field near the piston top is required for more accurate prediction of heat flux in gasoline engines.
Water injection has attracted attention as one of the techniques for improving thermal efficiency of SI engine. When liquid water is injected in cylinder or inlet manifold, water absorbs the heat of fuel/air mixture due to its evaporation and reduce temperature of mixture. Consequently, it is expected that water injection can reduce cooling loss and suppress knock. The water injection in SI engine so far has been attempted mainly in inlet manifold. Port injection of water forms uniform water vapor distribution in cylinder. On the other hand, this research proposes in-cylinder direct water injection for reduction of cooling loss and suppression of knock effectively with small amount of water by forming low temperature water vapor layer on the wall. In the present paper, in order to investigate the influence of in-cylinder direct water injection on combustion and heat transfer in SI engine, measurements of in-cylinder pressure and wall heat flux and combustion visualization are conducted simultaneously by using rapid compression and expansion machine(RCEM). In order to separate the thermodynamic and chemical effects of water addition, homogeneous mixture of fuel/air/water is introduced into RCEM, and the amount of water and in-cylinder gas temperature are changed independently. Firstly, in order to investigate the chemical effect of water, the amount of water is changed with a constant in-cylinder gas temperature. Secondly, in order to investigate the thermodynamic effect of water, the in-cylinder gas temperature is reduced for simulating the water absorbing heat due to evaporation in cylinder. The obtained results show that addition of water vapor to in-cylinder gas uniformly leads to reduction of wall heat flux and suppression of knock, whereas flame propagation speed remarkably decreases.
From the global environmental point of view, drastic fuel efficiency improvement is required for engines. Cooling heat loss is one of the most dominant losses among the various engine losses to be reduced. Although many attempts to reduce it by insulating the combustion chamber wall have been carried out, most of them have not been successful. Charge air heating by the constantly high temperature insulating wall is a significant issue of these attempts, because it deteriorates charging efficiency, soot and NOx emission in diesel engines, and the tendency of knock occurrence in gasoline engines. Authors have developed a new concept heat insulation methodology, which can reduce cooling heat loss without heating the charging air. Surface temperature of insulation coat changes rapidly, according with the quickly changing in-cylinder gas temperature in each engine stroke. Reduced temperature differences between them lead to lower heat transfer. During the intake stroke, surface temperature of the insulation coat goes down rapidly, and prevents intake air heating. To realize the above mentioned functionality, a thin coating layer with low thermal conductivity and low heat capacity was developed. It was applied on the pistons of diesel engines, and showed improvement of fuel efficiency, and increased exhaust gas temperature which contributes to earlier light off of the aftertreatment catalyst. As the result of energy balance analysis, cooling heat loss was reduced, on the other hand, break power and exhaust loss was increased. In addition, unburnt fuel emission reduction in the low temperature starting was observed.
Recently, fuel efficiency requirements has been stricter, and then diesel engines with high efficiency are focused in automotive industry. However, about 30% of the input heat quantity is lost as cooling loss inside the combustion chamber, therefore reduction of this is required to further improve the thermal efficiency in diesel engines. Cooling loss is mainly due to wall heat loss caused by convective heat transfer of the spray flame impinging on the chamber wall. From Newton's cooling law, wall heat loss is determined by physical factors, which are heat transfer coefficient, flame temperature, wall temperature, flame contact area and flame contact time on the wall. And these physical factors are affected by control factors. Therefore it is important to investigate the correlation among physical factors, control factors and the wall heat transfer with impinging flame on the wall for reducing cooling loss. In this study, it is investigated that the mechanism of the wall heat transfer under the low flow field with a rapid compression and expansion machine for reduction of wall heat loss in diesel engines. And in this paper, it is focused on that the correlation between diesel spray flame and wall heat loss by changing fuel temperature. In this experiment, temperatures on the wall surface and the wall inside are simultaneously measured with coaxial thermocouples, which are embedded in the piston, and heat flux is assessed from the difference of these temperatures. In addition, image correlation method and two-color method were performed from luminous flame images to calculate flame velocity and flame temperature. From the measurement result of the heat flux, it was found that wall heat loss is reduced with increasing fuel temperature.
Empirical models instead of physical knowledge mostly describe the heat transfer between combusting diesel sprays and piston walls in IC engines. Many influences on the wall heat transfer interfere with each other. This makes it difficult to investigate single parameters, which would be necessary for physical descriptions. Optical spray chambers offer test conditions suitable for such investigations. Thus, pervious work describes the heat transfer of impinging diesel sprays on flat walls under diesel relevant conditions or the influence of the impingement on the combustion of diesel fuel. Parameters like ambient conditions, wall distance and injection pressure were studied. Due to increasing power densities of IC engines, the wall material is a new parameter that has to be addressed. The increasing thermal and mechanical loads onto the piston demand changes in the used material or the surface after treatment. Accordingly, the resulting surface temperatures and heat fluxes are interdependent with the combustion. In this work, we show the temporal evolution of surface temperatures of different wall materials under impinging diesel flames in a constant pressure combustion chamber. The aluminum configuration reaches higher surface temperatures, than the steel configuration. The flames are planar characterized by their soot temperature (KL-method) and their OH* radical appearance. Due to higher surface temperatures a higher amount of rich fuel burns close to the wall in the aluminum configuration. In contrast the steel configuration shows a high ratio of premixed combustion in the wall jet vortex with less soot formation. The findings offer a detailed view on approved engine research, where aluminum The results of the current work count for the moment and place of flame wall interaction and do not average over an engine cycle. The local and temporal temperature peaks on the wall surface can influence the emission and soot formation as well lead to crack initialization and mechanic fatigue operating close the critical material temperatures.
Ignition delay times of 2,5-dimethylfuran(2,5-DMF)/n-heptane/oxygen/argon mixtures were measured in this study. Experiments were performed at the stoichiometric mixtures in the temperature range of 1200-1800K. A blend model for 2,5-DMF/n-heptane blends was developed by combining two selected models for each individual fuel component. According to the combined model, the ignition delay time increases with an increase in the 2,5-DMF addition level, non-linearly. Kinetic analysis was conducted to investigate interaction effects between two fuels. At relatively low blending ratios of 2,5-DMF, the two fuels have negligible impacts on the reaction path of each other. As the 2,5-DMF addition increases to relatively higher levels, decomposition path of n-heptane is significantly changed due to the competition for small radical.
The ignition delay times of 2-methylfuran (MF), the primary reference fuel (PRF), and the MF-PRF90 blend were measured using a rapid compression machine at equivalence ratios of 1.0 and 0.5. The oxygen concentration was 16.4% and the temperature and pressure ranges were 655-888 K and 1.97-2.78 MPa, respectively. The ignition delay times of the MF-PRF90 blend were shorter than those of MF and longer than those of PRF90 under all experimental conditions. A detailed chemical kinetic model for the MF-PRF90 blend was constructed on the basis of the existing model. Further modifications of the rate constants were carried out to improve the agreement with the experimental ignition delay times. The simulated ignition delay times were in good agreement with those in the experiments. A sensitivity analysis of the ignition delay times was performed to investigate the important reactions in the ignition of the MF-PRF90 blend. It was found that the reaction of the HO2 radical with the 2-furanylmethyl radical to form the OH radical plays a key role in the oxidation of the MF-PRF90 blend, as do the H abstraction reactions from isooctane. It was considered that the increase in the ignition delay times of the MF-PRF90 blend was caused by the consumption of the OH radical by MF.
A validation study on the chemical reaction models for heptane-isooctane mixing fuel as a primary reference fuel of gasoline (PRF) and heptane-toluene mixing fuel (TRF) was performed by using repetitive compression-expansion reactor (RCER). The mechanism of RCER is similar to a motoring engine, however, the concentration of sample gas in the reactor is regulated like a conventional rapid compression machine (RCM). Experimental conditions were set as follows. The values for initial pressure and temperature were set to 30, 40 kPa and 360 K, respectively. Engine speed was set to 600 rpm. Maximum pressure in the reactor reached about 1 MPa from an initial pressure as 30 kPa by compression of air. The dimension of RCER is as follows. Displacement volume is 382 cm3 and the compression ratio is 27.5. The initial gas mixtures are charged homogeneously in the reactor and compressed for auto-ignition, that is, the HCCI condition. The ignition timing was delayed by increasing the mixing fraction of isooctane or toluene. Comparing the results and simple chemical kinetics simulations was also performed with two reaction models in the literature. Both models showed similar trends of the delay times as well as the experimental results. However, the retarding effect of a higher fraction of toluene in the fuel mixtures was overestimated in one model despite good agreement each other in the case of heptane alone. It suggests that the toluene reaction model has some problems for the conditions in this study.
A vertical-type micro flow reactor with a controlled temperature profile was used to investigate the effect of the equivalence ratio on low and high temperature reactions of ultra-lean gasoline surrogate/air weak flames. Surrogates of research octane number 90 and 100 were used at equivalence ratios between 0.50 and 1.00. In the experiments, leaner mixtures showed a higher reactivity as indicated by hot flames stabilizing at lower wall temperatures. In 1D steady computations by PREMIX only LLNL mechanism was able to reproduce this trend. This was accounted to difference in the reaction parameters of the chain termination reaction H+O2(+M)=HO2(+M) in the KUCRS mechanisms. Once LLNL's parameters were introduced, results by KUCRS and SIP mechanisms were able to reproduce the experimental results with higher accuracy.
In the gasoline combustion team of Cross-ministerial Strategic Innovation Promotion Program (SIP), “Innovative Combustion Technology,” gasoline surrogate fuels composed of isooctane, normalheptane, toluene, isooctene, and methylcyclohexane, has been employed to standardize fuel properties. A proto-type of detailed reaction mechanism for the five-component gasoline surrogate fuels, consisting of 1759 species and 5799 reactions, has been developed. Japanese regular gasoline with the research octane number of 90 has been modeled with the mole fractions of the five components of 0.238247, 0.199032, 0.38830, 0.121247, and 0.38830, respectively. In the present study, based on knowledge derived from reaction path analysis for each component, a small-scale reaction mechanism for the regular gasoline surrogate fuel was developed. Species and reactions were reduced to 55 and 92, respectively. This mechanism can represent not only the relationships between initial temperature and ignition delay time, but also the profiles of heat release rate during ignition process with the initial temperatures between 600 K and 1200 K, and the equivalence ratios between 0.5 and 1, presented using the detailed mechanism, accurately except for low-temperature oxidation induction times with fuel, O2, and N2 concentrations decreased.
Chemical kinetics of natural gas combustion receives much attention for the improvement of gas turbine engines with the difficulties of flashback, blow-off and combustion instabilities. This study aimed to develop a skeletal reduction mechanism for the combustion of natural gas. By using a detailed chemical kinetics mechanism of acyclic saturated hydrocarbons composed of up to C7 species, we found that methane, ethane, propane, and n-butane are surrogate components for the prediction of ignition delay time and laminar flame velocity of natural gas. Although minor components from C5 to C7 enhance the reactivity at lower temperatures, the mixture in which these minor species are replaced into n-butane can reproduce the ignition delay time of original natural gas within 25% error. Then, we performed the reaction path and sensitivity analysis for the combustion of surrogate components by using a detailed chemical kinetics mechanism, and identified the unnecessary species based on the 5% of branching ratio and sensitivity. The present skeletal reduction mechanism which includes 77 species and 334 reactions can reproduce the ignition delay times and laminar flame velocities predicted by detailed chemical mechanism.
A detailed chemical kinetic mechanism has been developed to describe the pyrolysis and oxidation of the hydrogen/NOx and syngas/NOx systems. The thermodynamic data of nitrogenous compounds have been updated based on the study of Bugler et al. [J. Bugler, K.P. Somers, J.M. Simmie, F. Güthe, H.J. Curran. J. Phys. Chem. A, 2016, 120(36):7192-7197.]. The rate constants of individual elementary reactions associated with the Zeldovich mechanism, the N/O sub-mechanism (NO2, N2O and NO3), the H/N/O sub-mechanism (HNO/HON, HNO2/HONO and HONO2) and the NH3 mechanism (NNH and NH2OH) have been selected through a synthetic comparison of the data available in the literature and the adoption of the latest available published rate constant data. The proposed mechanism has been validated against a large number of experimental data including pyrolysis histories, ignition delay time data, species profile versus time and temperature and flame speed measurements over a wide range of initial combustion conditions and various experimental devices including shock tubes, flow reactors, jet-stirred reactors and spherical combustion bombs. The simulations of the proposed model have also been compared to those from five recently published kinetic models available in the literature. It was found that although these mechanisms generally reproduced well the data for which they were validated, they did not globally capture the combustion characteristics of all of the hydrogen/NOx and syngas/NOx systems. Finally, the proposed model has been used to simulate the formation of NO at practical gas-turbine relevant conditions. A detailed flux analysis has been performed to kinetically explore the NO formation mechanism under various combustion conditions.
The conversion effects of a three-way catalyst are simulated in previous works using single and multiple representative channel approaches with detailed surface kinetic models. In addition, this article introduces global gas phase chemistry to the model. This allows reflecting ongoing reactions due to incomplete combustion products in low temperature regime. The 1D single-channel model representative for the catalyst is used here. Next to the comparison of the catalyst outlet emissions with and without gas phase chemistry, the transient temperature increase is simulated in order to model the catalysts light off temperature. Additionally, the transient inlet emissions are enhanced to show the influence of water and hydrogen on the modeling results. The heat transfer is modeled by wall heat losses to provide proper heat dissipation out of the catalyst. The modeling results show a good agreement to the experimental data with low computational cost.
In this study, a detailed chemical kinetic database of NO/ CO/ O2 surface reaction on Rh/ Al2O3 has been constructed based on the measurements of gaseous / surface species. The gaseous species at the upstream and downstream of the monolithic catalyst were identified by FTIR, while the surface species on the powder catalyst were directly measured by FTIR. Based on those experimental results, detailed surface reaction kinetic database has been constructed. As the result of numerical simulation with 1-D code, it was confirmed that the gaseous conversion rates of NO and CO were quantitatively reproduced with the database.
The permanently tightening emission legislations especially for NOx pollutants mean new challenges to car manufacturers. A very effective method for reducing NOx emissions in the exhaust gas is available in the framework of selective catalytic reactors, originally developed for coal power plants. Excellent CFD based methods to predict AdBlue dosing, ammonia uniformity and catalytic reactions have been presented in the past. The aim of this paper is to depict a powerful and efficient method for the simulation of urea-water decomposition. Therefore the two main points discussed here are a detailed decomposition model and efficiency improvements to reduce simulation turnaround time. It will be concluded that a mathematical model to consider solid urea and the deposition is a key point for the successful simulation of an exhaust aftertreatment system. Furthermore, with appropriate efficiency enhancements, simulating minutes of real time becomes affordable. Simulation results are shown to display both the quality of simulation models and methods. The benefits are highlighted and an outlook is given about potential future developments in the field of numerical model development for SCR systems.
The urea-SCR system has been used commercially due to the high performance of NOx reduction, and much research has been conducted to invent low-noble metal catalysts effective at low exhaust temperatures. The CFD (computational fluid dynamics) with a reaction model is effective for SCR catalyst development and many SCR simulation models have been proposed. However, experimental feedback of surrogate gas tests is required in the simulations and fitting parameters in the models are adjusted by the experiment results. This is a cause of the low versatility of the reaction models. To develop a versatile ammonia-SCR reaction model, the reaction rate of Standard-SCR reactions, molecular diffusion in catalyst coated layers, and structural parameters of honeycomb catalysts are introduced in the simulation in the report. These chemical and material parameters are not based on engine emission test data, and the reaction rate coefficients and order of reactions are measured with the catalyst in powder form by flow reactor experiments, and the mass-transport is described by the mass transfer coefficient and Knudsen diffusion in the model. The NOx conversion in the model is calculated for several copper zeolite catalysts, ZSM5, Beta, SSZ13, and P-CHA at an 88,000 h-1 space velocity (SV) condition, and the calculated results are compared with the NOx conversion data from engine emission tests and surrogate gas (mixture of nitrogen oxide [NO], oxygen [O2], water vapor [H2O], and nitrogen [N2] as the balance) tests. The calculated results of the ammonia SCR model correlate well with the surrogate gas tests up to 350°C with all catalysts, but the correlation is poor above 400°C because the ammonia oxidation is not considered in the simulation.
A Hydrocarbon is supplied to a diesel oxidation catalyst installed in an upstream of the diesel particulate filter for burning particulate matter deposited on the diesel particulate filter. In this study, we developed a thermal fluid numerical computation fluid dynamics code including a chemical reaction rate model which shows the influence of a hydrocarbon concentration and a type on the oxidation reaction of a hydrocarbon on a diesel oxidation catalyst. First, oxidation characteristics of four types of hydrocarbon fuels (Decane, Hexadecane, Eicosane, and 1-Methylnaphthalene) were investigated using a straight flow substrate carrying a platinum palladium catalyst. 1-methylnaphthalene greatly deteriorated the oxidizing ability, but there was no significant difference in the three kinds of alkane fuels. The difference of these oxidation characteristics could be reproduced by constructing a sub model expressing the difference between the adsorption characteristics of each hydrocarbon fuel and the oxidation inhibition effect due to adsorption. Further, the oxidation characteristics of the zeolite-containing catalyst were evaluated. As a result of the evaluation, the oxidation characteristics of the decane of the linear hydrocarbon did not change when compared with the catalyst which did not contain the zeolite, but the oxidation characteristics of the aromatic hydrocarbon 1-methylnaphthalene deteriorated. It was assumed that the zeolite had high adsorption capacity of 1-methylnaphthalene and a further inhibition of a reaction by adsorption was accelerated.
Diesel Particulate Filters are the most complex component of today’s emission control systems as they need to incorporate different and often conflicting functionalities such as high soot nanoparticle filtration efficiency, low pressure drop behavior, direct catalytic soot oxidation activity, high oxidation activity for Carbon Monoxide, Hydrocarbons and Nitrogen Oxide, as well as ability to reduce Nitrogen Oxides. Significant progress has been made employing both fundamental research as well as applications-oriented approaches. These have led to an improved understanding of the coupled physicochemical transport and reaction phenomena occurring in DPFs. We have researched new materials, designs and control approaches for enhancing the design and reliability of future Diesel Particulate Emission Control systems. An enabling technology has been the development and synthesis of nanostructured catalysts and their incorporation into a multifunctional reactor for Diesel Emission. Apart from the composition, the material morphological characteristics also contribute to the catalytic activity. As an example we present different nanoparticle catalysts characterized with respect to their physical and morphological properties as well as with respect to their catalytic soot oxidation activity. The detailed kinetic data obtained are shown to correlate very well to a composite morphological parameter, derived here for the first time, combining the catalyst particle size, surface area, crystallite size, and porosity. Likewise when sub-micron/micron sized, dense catalyst particles were obtained by milling for various time intervals, resulting in a poly-disperse multimodal distribution, the kinetic data correlate well with the total surface-weighted mean particle size. This study establishes to our knowledge for the first time a quantitative mechanistic link between catalyst particle structure and kinetic parameters, facilitating much the rational estimation of kinetic parameters in simulation studies of catalytic soot oxidation.
DPF (Diesel Particulate Filter) is effective for trapping diesel soot from diesel engines. The trapped diesel soot is removed by combustion above 600 °C using excess fuels, which results in reduction of fuel economy. An effective design of DPF is strongly required to improve DPF regeneration efficiency and fuel economy; however, the internal phenomenon of diesel soot combustion on DPF is still not clarified. For an effective design of DPF, in this study, we investigated the effect of diesel soot structure on combustion temperature using real diesel soot. Soot samples were characterized by means of various physicochemical techniques: XPS (X-ray photoelectron spectroscopy), EELS (Electron Energy Loss Spectroscopy), Raman spectroscopy, TEM (transmission electron microscopy), and elemental analysis. TEM images and Raman spectra of real diesel soot showed that the difference in the crystallinity of soot was small. XPS, EELS, and elemental analysis revealed that oxygen content, which is derived from surface functional groups of -COOH, -OH, and -C=O at defect sites, was strongly affected by the engine operating conditions. The ignition temperature of diesel soot was dependent on O/C ratio measured by elemental analysis. With the increase in oxygen content of diesel soot, combustion temperatures of soot became lower. It was concluded that the concentration of oxygen containing functional groups is an essential factor for the control of the ignition temperature of diesel soot.
The soot trapping process on a diesel particulate filter (DPF) occurs in three phases: bridge formation, surface pore filtration, and cake filtration. The accumulation of soot in the surface pores, which is initiated by bridge formation at the constricted area inside a DPF porous channel, causes remarkable increase in pressure drop and negatively affects engine efficiency. An accurate prediction of accumulation depth and location of the bridge facilitates accurate estimation of the accumulated soot mass based on the pressure drop. In this study, soot was introduced into a DPF sample, wherein superficial gas flow velocity and soot size were varied in order to observe their effects on the soot penetration depth. The extracted pressure drop and soot accumulated number for the bridge formation and surface pore filtration were investigated, and the soot packing fraction distribution was obtained via post-observation of the wall cross-section. While the location of the bridge formation is a function of the porous configuration, a smaller soot amount is observed to accumulate inside the surface pores under conditions of lower superficial gas flow velocity or smaller soot size. The variation in soot accumulation with respect to the variation in the condition was a result of the timing of bridge formations, which occurred competitively at various bridging sites. Soot accumulation in deeper areas is prevented by high efficiency of the bridge formation, which is a result of the contribution of Brownian motion, which in turn depends on the flow velocity and soot size.
Diesel engine vehicles are equipped with a particulate matter (PM) sensor for the on-board diagnostics (OBD) of the diesel particulate filter (DPF). It has been previously reported that a fraction of the ash generated by diesel engines can pass through the DPF, causing malfunction of the PM sensor. However, the slip mechanism of the ash accumulated in the DPF is still not well understood. In this study, the slip mechanism of ash stored in a DPF has been elucidated using a DPF, in which surrogate ash was stored. Quantitative measurements of the component of ash passing through the DPF were performed using an HPLC, by varying the factors that influence the mass of the ash passing through the DPF, such as the mass flow of water, inlet gas temperature, exhaust gas flow rate, and the mass of ash stored in the DPF. The main contributing factor for the occurrence of ash slip was found to be the mass of liquid water passing through the DPF. The components of ash passing through the DPF were mainly Ca2+ and SO42-, which indicates that the CaSO4 in the ash dissolves in the condensed water in the exhaust. Using experimental data, the estimation method for the mass of ash passing through the DPF was also investigated.
This study investigates effects of evaporation and mixture formation of diesel spray on ignition and initial flame development. Spray structures injected from a single-hole injector were observed by a shadowgraph technique using a constant-volume spray chamber. The spray chamber was set under the condition of temperature at 900K and pressure at 4MPa similar to the atmosphere in a real diesel engine. Analysis of the observed spray images shows that combustible mixture tends to be formed early at the midstream region. Evaporation at the inside of the spray is promoted at the midstream as well as the spray boundary including spray tip. Direct photography measurement reveals that OH radical first appears at the midstream region. High injection pressure enlarges the fuel vapor area and moves the OH radical early to the downstream region. Initial luminous flame development that is thought to be luminescence of soot can be observed at the center region of the spray downstream under the condition of low injection pressure of 100MPa; on the other hand, the initial luminous flame tends to develop at the outer region of the spray for high injection pressure of 140MPa. The equivalence ratio distribution obtained by the tracer-LAS measurement can explain the relation between the mixture formation and the initial luminous flame development. Analysis of the micro-scale spray images provides that the difference of the initial flame development depending on the injection pressure is partly caused by the droplet distribution of the spray. Large-size droplet tends to be distributed at outer of the spray boundary for higher injection pressure. Fuel vapor area is easy to expand outward for higher injection pressure. Under the condition of lower injection pressure, mixture formation at the inside of the spray is delayed; accordingly initial luminous flame develops at the inside of the spray.
L2F (Laser 2-Focus Velocimeter) has been utilized for the measurement of velocity and size of droplets in diesel fuel spray injected from a single-hole nozzle. The L2F had a micro-scale probe which consists of two foci. The focal diameter was about 3 μm, and the distance between two foci is 20 μm. Diesel fuel sprays were injected intermittently by using a common rail injector. The diameter of the nozzle orifice was 0.125 mm. The injection pressure was 100 MPa, and the ambient pressure was changed from 0.1 to 0.7 MPa. The spray image was taken by a CMOS camera with a resolution of 1.62 million pixels in the xenon lamp lighting with a period of 180 ns. The L2F measurement was conducted at 13 mm downstream from the nozzle exit. The velocity of droplets near the spray center was higher than the one at the spray periphery. The change of droplet velocity was small with the increase of ambient pressure at the spray center, and the velocity of droplets at spray periphery decreased with the increase of ambient pressure. The size of droplets at the spray center was larger than the one at the spray periphery. The size of droplets decreased with the increase of ambient pressure inside the spray, especially, the amount of decrease in size was larger at the spray center. The characteristic velocity of the Weber number was the difference between the droplets velocity and the air flow velocity which was defined as the velocity of droplets less than 3 μm. It is understood that the increase in the Weber number due to the increase in the ambient density enhances the atomization of large droplets. These findings are thought to be applied to the initial condition and evaluation of numerical simulation of atomization.