Computational Study of the Laminar Reaction Front Properties of

Subscriber access provided by UNIVERSITY OF THE SUNSHINE COAST

Article

A Computational Study of the Laminar Reaction Front Properties of Diluted Methanol-Air Flames Enriched by the Fuel Reforming Product Duc-Khanh Nguyen, and Sebastian Verhelst Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.7b00691 • Publication Date (Web): 25 Jul 2017 Downloaded from http://pubs.acs.org on July 30, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

A Computational Study of the Laminar Reaction Front Properties of Diluted Methanol-Air Flames Enriched by the Fuel Reforming Product Duc-Khanh Nguyen*, Sebastian Verhelst Department of Flow, Heat and Combustion Mechanics, Ghent University, SintPietersnieuwstraat 41, B-9000 Gent, Belgium

ACS Paragon Plus Environment

1

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 48

ABSTRACT

The current work investigates the flame structure and the propagation of premixed laminar flame fronts for mixtures of diluted methanol-air enriched by fuel reforming products at spark ignition engine conditions. Two engine concepts were investigated, one with external fuel reforming (EFR) and one with reformed exhaust gas recirculation (R-EGR). Here, the fuel reformate (or syngas) is a mixture of H2, CO and CO2 with a CO selectivity of 6.5%, which is used to represent the products of methanol steam reforming over a Cu-Mn/Al foam catalyst. The simulations were exercised over a wide range of dilution level and unburned gas temperature at 40 bar with a skeletal chemical kinetic mechanism using 0/1-dimensional codes. Two types of dilution, air and EGR, were compared at the same fuel-to-charge equivalence ratio ϕ’. The results showed that the knock limit for spark-ignited operation is extended with rising dilution levels, especially when diluted by an R-EGR mixture. Syngas addition also leads to reduce the knock tendency. At the knock limit, the fraction of heat released by auto-ignition is greater at a higher dilution rate. The ringing intensity during knock increases with a higher concentration of diluted gases, especially diluted by the EGR gases. At the lower flammability limit, the stable flame propagation range is expanded to a lower unburned gas temperature when using air dilution. At stoichiometric conditions, dilution with an R-EGR mixture is recommended for part load operation because it provides a stable flame at high dilution levels. The influence of gas properties were also investigated, where a shorter ignition delay after compression was observed with a leaner mixture. The ignition delay increased with a higher EGR ratio, especially in the case of the REGR mixture. Therefore, the knock tendency increases with a leaner mixture. However, the knock ringing intensity decreases if the mixture is diluted by air. A mixture diluted by R-EGR


2

Page 3 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

can reduce the knock tendency compared to the two EGR dilution cases, and can decrease the knock ringing intensity compared to the two air dilution cases at high dilution ratios.

INTRODUCTION Although internal combustion engines (ICEs) had a big contribution to the development of the modern society, some governments have been planning to reduce the number of ICE powered vehicles because they consume around 42 million barrels of crude oil each day 1 and produce a huge amount of polluting emissions. Recently, four big cities (Paris, Mexico City, Madrid and Athens) have decided to ban cars and trucks that run on diesel from 2025 2. Although diesel engines have a higher fuel efficiency (less CO2) than gasoline engines (or spark ignition engines), a larger quantity of particulate matter (PM) and nitrogen oxides (NOx) are the key reasons for that decision. PM, NOx and other pollutants have contributed to the deaths of over three million people each year 3. In spark ignition (SI) engines, fuel and air are premixed at stoichiometric condition before the ignition, a homogeneous mixture which results in less PM than the diesel engines. Although the SI engine produces more NOx than the diesel engine, that emission can be easily treated by a cheap after-treatment system, the three-way catalyst (TWC). However, low fuel efficiency is a challenge of the SI engines. With high throttling losses and low compression ratio, the SI engine is less efficient than the diesel engine. The compression ratio is limited by the auto-ignition of the end gas (knock phenomena) during the flame propagation. Diluting the mixtures or using alternative fuels with high research octane number (RON) and high burning velocity are solutions to prevent knock.


3

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 48

Both air and exhaust gas recirculation (EGR) dilutions are promising approaches to mitigate knock 4 as well as to enhance the efficiency of SI engines 5. The efficiency improvement can be explained by the decrease of combustion temperature, thus lower heat losses, reduced pumping work and increased specific heat ratio. With the enhancement of knock resistance, a higher compression ratio can be used. However, the cycle-by-cycle variation increases with both types of dilution 4. Another challenge of lean operation is the NOx emission due to the low conversion efficiency of the TWC in that condition. Regarding alternative fuels, methanol (CH3OH) currently is regarded as an excellent fuel for future SI engines 6. Compared to the conventional fuel of SI engines, gasoline, methanol has a faster laminar burning velocity, higher RON and higher heat of vaporization. These interesting properties support the operation with high dilution levels in a higher compression ratio engine. Furthermore, it is an oxygenated fuel, so the combustion can be cleaner. A large number of studies have shown the potential of methanol in increasing engine efficiency and reducing pollutant emissions 7-12. Hydrogen is also considered as a viable solution for future transportation 13. However, the biggest challenges for hydrogen include the production source, distribution infrastructure, and onboard storage 14. Due to the low energy density, hydrogen is compressed to 700 bar or liquefied before distribution 13. In order to solve the energy density problem, fuel reforming of liquid hydrocarbons has been investigated for onboard hydrogen production, mostly for fuel cell application. Compared to other fuels, thanks to high H/C ratio and low reaction temperature, methanol is recognized as the leading candidate for onboard generation of hydrogen for automotive propulsion 15. The onboard generation reactions are usually endothermic. Waste exhaust heat can be used to drive these reactions. There are two main approaches which are normally used to produce hydrogen to recover the waste heat, methanol thermal decomposition


4

Page 5 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

(or dissociated methanol) and methanol steam reforming. Table 1 compares two reforming approaches in global reaction, lower heating value (LHV) ratio of products to reactants, exergy ratio, and exergy-to-LHV ratio of the reformate products. The exergy X is the energy that is available to produce work, it is used as the basis of the Second Law analysis 16. The exergy-toLHV ratio is used to predict the capability of producing more work compared to the indicated LHV. Subscripts p and r represent products and reactants, respectively. Table 1. Comparison of two methanol reforming approaches: thermal decomposition vs. steam reforming Reforming approach

Global reaction

Thermal decomposition CH3OH CO + 2H2 Steam reforming

CH3OH + H2O CO2 + 3H2

LHVp/LHVr

Xp/Xr

Xp/LHVp*

1.197

1.0375

0.9743

1.132

1.0093

1.0034

* based on 1 mole of methanol. As can be seen in the LHV ratio of product to reactant, the LHV increases by about 20% with the thermal decomposition method, which is higher than that of steam reforming (where it is around 13%). However, the exergy only improves by around 4% and 1% in thermal decomposition and steam reforming, respectively. Based on one mole of methanol, the ratio of exergy-to-LHV of the steam reforming products is greater than unity, which means the steam reformate gases are capable to perform more work than that indicated by their LHV. During the 1980s-1990s, some studies about the use of decomposed methanol as a fuel for SI engines were done at the Royal Institute of Technology – KTH. They concluded that the engine


5

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 48

could start at -30 oC and the engine efficiency increases 15-20% relative to neat methanol with the presence of hydrogen rich gas (syngas) 17-19. Brinkman and Stebar 20 did the experimental study to compare the engine efficiency and fuel consumption with methanol and decomposed methanol at a varied compression ratio and equivalence ratio. At the same compression ratio and equivalence ratio, although the LHV increased by around 13%, the reduction of fuel consumption for decomposed methanol was around 3-7% compared to methanol. However, the indicated efficiency of decomposed methanol is around 0.89-0.55 times that of methanol due to the increase of compression work (lower molar weight) and higher heat losses. In their research, the authors did not take the efficiency of fuel decomposition into account, they considered only the engine efficiency, not the efficiency of the system. In general, the use of decomposed methanol results in higher engine efficiency 21. However, using 100% decomposed methanol is not recommended because the liquid fuel is needed for a fast response at transient conditions, for increasing the volumetric efficiency, and for operating at high load modes 21-22. McCall et al. 23 compared the performance of dissociated methanol, methanol steam reformate and liquid methanol on an unmodified SI engine. The maximum power at 2000 rpm was lower by around 50% for decomposed methanol and by 65% for steam reformate compared to methanol. However, there was an insignificant difference in thermal efficiency between dissociated and steam reformed methanol, both approximately 25% higher than liquid methanol. Due to the reduction of engine power at high load, a research group at Israel Institute of Technology – Technion recently has converted a carbureted gasoline SI engine to a direct injection system operating with gaseous methanol steam reformates to maintain the peak power output at its original level 24-26. With direct injection of methanol steam reforming products, the thermal efficiency was improved by 18-39%, with a dramatic reduction of CO, HC and NOx,


6

Page 7 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

compared to gasoline 25. The combustion duration was shorter and the lean burn limit was significant extended compared to gasoline due to the high burning velocity and wide flammability range of the hydrogen in the methanol reformate 26. Syngas has a significantly faster laminar burning velocity compared to methanol 27 and the knock tendency in SI engines reduces when syngas is added 28. Faster burning fuel leads to a reduction of combustion duration, a shorter time for heat transfer, and the combustion phasing can be optimized, resulting in a higher efficiency 29. With a higher burning velocity fuel, the lean limit or the dilution tolerance can be increased 30, lowering heat loss due to a lower combustion temperature. Because the lean or EGR tolerance is extended, a leaner mixture or a higher EGR rate can be used, so that the throttling loss at part load significantly decreases. Furthermore, because fuel reforming is an endothermic process, syngas has a higher heating value than the input fuel, methanol. These reasons explain the increase in engine efficiency. In this research, the steam reforming technology is chosen because the reforming process reacts faster 31 and the deactivation of the catalyst is less severe with the presence of water 32. The reformate (syngas) is a mixture of H2, CO and CO2 with a CO selectivity of 6.5% which represents the products of methanol steam reforming over a Cu-Mn/Al catalyst at a temperature of ~320 oC and a water to methanol molar ratio of 1.5 33. This catalyst will be used in future experiments by the authors, hence its characteristics have been chosen to fix these conditions. In previous research by the authors 34, the external fuel reforming concept (EFR) (see Figure 1a) and the reformed exhaust gas recirculation concept (R-EGR) (see Figure 1b) were compared. In the first concept, a small amount of fuel (methanol) is injected into the catalyst which is heated up by the hot exhaust gases and reacts with water from another tank to form syngas. Because the reaction temperature of methanol steam reforming is much lower than the engine exhaust gas


7

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 48

temperature, the methanol conversion is assumed to be 100%, i.e. there is no methanol left in the product. The volumetric flow rate of water is manipulated to meet the required water to methanol molar ratio, 1.5. In the steam reforming, to convert one mole of methanol, one mole of water is needed. Because the water to methanol molar ratio is 1.5, there is water vapor left in the product after reforming. The wet syngas then is condensed, after which the dried mixture is recirculated to the intake and mixed with the fresh air before entering the combustion chamber. Methanol is also injected directly into the combustion chamber. The molar ratio of methanol to syngas in the chamber is controlled by the mass fraction of methanol which is supplied to the catalyst. The EGR system is added to dilute the combustion. The second concept, R-EGR, was proposed and investigated with other fuels such as bio-ethanol, gasoline, dimethyl ether, diesel, biodiesel and gas-to-liquid fuel 35-45. Here, the water vapor contained in the EGR is utilized to react with fuel (methanol) which is injected and vaporized upstream the catalyst to produce syngas. The catalyst is installed inside the EGR system. The amount of methanol is controlled on the basis of water vapor concentration in the exhaust gas (or the EGR rate), with the water to methanol molar ratio similar to the EFR concept. In this research, the dry reforming (CO2 reforming) of methanol is neglected. The mixture of syngas, excess water vapor and inert gases (CO2 and N2) in the combustion products then follows the EGR path to recirculate back to the intake manifold. In this concept, an extra water tank is not needed. In the EFR concept, in order to make a mixture of methanol-syngas with a 50%-50% ratio by volume (designated as SG50), around 20% of the fuel mass is supplied to the catalyst (with the remainder being injected directly into the combustion chambers). In the R-EGR concept, the amount of methanol that is injected into the EGR system is constrained by the concentration of


8

Page 9 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

water vapor in the exhaust gas. Around 20% of fuel mass is injected into the EGR system if the EGR ratio is 13%. Previous research showed that at stoichiometric condition, with the same EGR ratio of 13% and mass fraction of methanol supplied to the catalyst (~20% by mass), the laminar burning velocity in the R-EGR concept is faster than that of the EFR concept 34. So far, Nguyen and Verhelst 34 developed laminar burning velocity correlations for methanol-air mixtures with the enrichment of syngas for both the EFR and R-EGR concepts. However, the properties of the flame front like the knock behavior and flame structure at auto-ignition and lower flammability limit have not yet been studied. The primary objective of this research is to analyze the reaction front characteristics of diluted methanol-air flames with the addition of fuel reforming products. The stable combustion regime, knock and flame limit of diluted methanolsyngas-air flames will be discussed. Here, if knock is mentioned, the auto-ignition of the end gas after spark ignition is meant. METHODOLOGY In order to analyze the flame structures, chemical kinetic calculations were done using CHEM1D. This is a one-dimensional flame code developed by the Combustion Technology group at Eindhoven University of Technology 46. It solves the conservation equations of mass, momentum, species and enthalpy 47 and uses the EGLIB complex transport model 48. In each case, a stationary simulation was performed with an exponential differential scheme and free flame type using 200 grid points. CHEM1D was also used to calculate the flame thickness δF which was determined based on the temperature profile 49. For analyzing the stable combustion regime with different dilution methods, the auto-ignition delay time τID is needed. The ignition delay time was calculated using Cantera 50. The simulation is performed with a homogeneous


9

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 48

mixture in a constant volume reactor, and the ignition delay is defined as the time when the maximum gradient of temperature in the reactor is observed. All the simulations were performed with the kinetic mechanism developed by Li et al. 51. In the previous research 34, the laminar burning velocity of methanol and syngas (50% H2 – 50% CO) was validated with Li’s mechanism considered to be the best mechanism to predict the laminar burning velocity of methanol-syngas blends. The comparison of ignition delay of methanol and syngas between simulation and experiments is presented in Figure 2. The ignition delay of both methanol and syngas were measured at the National University of Ireland - Galway, in the studies of Burke et al. 52 and Kéromnès et al. 53 respectively. As can be seen in this Figure, the ignition delay of methanol is predicted almost perfectly with Li’s mechanism. For syngas, the simulated data does not fit that well with the measured ones, especially at low temperatures. However, according to Olm et al. 54, Li’s mechanism predicts the ignition delay of syngas better than the USC-II 2007 mechanism (the second best mechanism to predict the laminar burning velocity of methanolsyngas blends). Therefore, this mechanism is employed in this investigation. In the first parts of this paper, all simulations were done at a constant pressure of 40 bar, similar to previous studies 55-56. This is close to the end of compression pressure of a stoichiometric methanol-air mixture corresponding to an end gas temperature of 850 K if the pressure and temperature of the initial mixture are 1 bar and 343 K respectively. The temperature, equivalence ratio and EGR rate have been varied in a range of 600 – 1000 K, 0.6 – 1.0, and 0 – 40% by mass respectively. The impact of the gas properties is analyzed separately in the last part, as the end gas pressure and temperature after the compression will change for different mixtures or dilution ratio. Two assumptions of the initial mixture condition at intake valve closure were assumed, leading to a significant change in the end gas state, which will be explained later.


10

Page 11 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

To compare the air and EGR dilution, the fuel-to-charge equivalence ratio ϕ’ is used, given by 57:

φ'=

(F / ( A + R )) = φ 1 − X EGR ≈ φ (1 − X EGR ) (F / A)st 1 + φX EGR (F A)st

(1)

where, F, A and R are the masses of fuel, air and residual gas, respectively. The subscript st indicates the stoichiometric condition. For example, at ϕ = 1, if the EGR ratio is 20% by mass, the fuel-to-charge equivalence ratio is 0.8. In this research, the flame structure analyses were performed at a constant fuel-to-charge equivalence ratio, ϕ’ = 0.87. Five mixtures were investigated in this research, MeOH-Air, MeOH-EGR, SG50-Air, SG50-EGR and R-EGR. Methanol and methanol-syngas blend (50%-50% by volume) is designated respectively as MeOH and SG50. The following characters, Air and EGR, are presented for the air and EGR dilution, respectively. In the air dilution cases, it is a lean combustion without EGR, so ϕ = ϕ’ = 0.87. In contrast, the EGR dilution is done at stoichiometric condition only, which is the preferred operating condition in SI engines to achieve the highest conversion efficiency of the three-way catalyst. In the R-EGR case, because of the complexity of the reforming process in the presence of oxygen left in the exhaust gas when the engine is operated under lean combustion, this concept is investigated at stoichiometric condition only. STABLE REGIME OF DILUTED COMBUSTION In the research of Lavoie et al. 57, a multi-mode combustion diagram was proposed, based on the constraints of each region. For instance, the SI region is restricted by ϕ’ ≤ 1, the flame and the knock limits. The flame limit is a constant flame speed contour that passes through Tb = 1900 K, based on findings from experimental engine data of Flynn et al. 58. That limiting temperature is associated with the reaction rate balance within the flame front, which is driven by fundamental


11

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 48

hydrocarbon kinetics, presenting a boundary for a sufficiently fast combustion 58. When Tb falls below 1900 K, flame quenching or local quenching will occur. This value for Tb was observed with hydrocarbon fuels, and most likely does not represent the lower limit for the combustion of oxygenated fuels like methanol and a non-carbon fuel like hydrogen. Therefore, different Tb values should be used for each mixture, requiring fundamental studies for these fuels. For simplicity, instead of a lower temperature limit, a lower limit for laminar burning velocity uL is assumed here. A value of uL between 10 – 15 cm/s is considered as the flame limit for iso-octane 57

. With a faster laminar burning of methanol compared to iso-octane 59-60, a constant value of

laminar burning velocity uL = 15 cm/s, is employed to represent the lower limit of the SI region. For the upper limit of the SI region, a knock limit line is defined, as the temperature of unburned gas reaches ~1000 K 57. That temperature is considered as the knock limit of gasoline (isooctane), and that value will also be changed when a new fuel is used. Other research used the residence time, τRES, as an indicator of knock occurrence. If τID < τRES, the end gas auto-ignites before the flame fully consumes the charge 61. This is a very rough approach to define the knock limit in SI engines, as knock is strongly influenced by the end gas pressure, temperature, and mixture composition but is used here for its simplicity. The residence time is a function of engine speed and residence period in crank angle degrees, and is defined by

τ RES (ms) = 13.5 × 1000 /(6 × RPM )

(2)

The residence time is 1.125 ms at an engine speed of 2000 rpm. This value is in the range of ignition delay when knock occurs, as in research of Kalghatgi et al. 62. The speed of 2000 rpm is selected because the engine which will be used for the experimental part of this research reaches


12

Page 13 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

maximum output torque at that speed. Therefore, this value of ignition delay will be used here as the upper boundary for the SI region. Figure 3 shows the simulated ignition delay τID of five mixtures versus ϕ’ at 40 bar and Tu = 850 K. The unburned temperature of 850 K is selected because that value is the minimum temperature to set off auto-ignition of methanol 63. The calculated ignition delay of iso-octane (iC8H18) and ethanol (EtOH) at lean conditions, using Arrhenius type correlations 64-65, were also plotted in the Figure. The trends of τID of iso-octane and ethanol which are calculated using Arrhenius correlations are similar to the simulated data of methanol and SG50, i.e. a longer ignition delay with a leaner mixture. The ignition delay in the EGR diluted cases also present a similar trend, the ignition delay increases with rising EGR ratio 66. Only the R-EGR mixture shows a significant improvement of τID with a reduction of ϕ’ due to the important growth of the H2 mole fraction in the reactant. SG50 has a longer τID than that of MeOH because of the presence of H2 in the reactant and a lower O2 concentration. Compared to the air diluted cases at the same ϕ’, the τID of EGR diluted mixtures are slightly shorter with both MeOH and SG50 fuels. Based on the relationship of τID and Tu, the required Tu to achieve the desired τID = 1.125 ms is found at each ϕ’. Figure 4 presents the contours of constant τID (the upper limit) and laminar burning velocity, which is derived from the developed correlation (the lower limit) 34 in the Tu-ϕ’ domain at 40 bar. Five mixtures were compared. As can be seen in this Figure, there is one trend that was observed for all mixtures: the knock limit is extended to a higher value of Tu when increasing the dilution level, therefore the auto-ignition of the end gas is inhibited with a diluted mixture at the same pressure. This can be explained by the increase of τID when ϕ’ decreases, as shown in Figure 3. Higher temperature leads to a reduction of ignition delay. Therefore, the fuel


13

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 48

which has a longer ignition delay can cope with a higher temperature before reaching the same value of ignition delay compared to high reactivity fuels. With the addition of syngas, the upper limit is also extended to a higher value of Tu. In the R-EGR case, at a higher EGR rate, the upper Tu limit can be increased significantly. Therefore, at a higher dilution level, R-EGR is a promising concept to decrease the knock tendency. Looking at the lower limit of the stable combustion regime, at low dilution levels, the unstable combustion does not matter with a homogeneous charge because the combustion only becomes unstable at a very low Tu. That temperature is almost impossible to reach when the charge is compressed to 40 bar. If the dilution level is larger, the required Tu to achieve a stable flame propagation is increased. Therefore, to avoid quenching or partial combustion during cold starts, it is recommended that the engine operates at stoichiometric condition without EGR. Furthermore, if the engine is operated with a stratified mixture and temperature inhomogeneities, there are unwanted phenomena at the locations which have a high dilution level or low local temperature. Therefore, if the EGR system is applied on turbocharged SI engines to dilute the combustion, low pressure loop EGR is recommended for a better mixing to avoid abnormal combustion during the flame propagation. From the lower and upper limits, air dilution has a wider range for Tu to achieve a stable flame propagation than EGR dilution. With the air dilution, stable combustion can be observed at a higher dilution rate (smaller ϕ’), therefore air dilution is suggested for part load operation. If NOx emissions are considered, operation at stoichiometric condition with EGR-type dilution is preferred to achieve a higher conversion efficiency of the TWC. In that case, R-EGR is a promising approach for part load operation to extend the dilution limit of the engine to get a


14

Page 15 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

higher fuel efficiency. At a higher load, 0.9 ≤ ϕ’ ≤ 1, the SG50-EGR case in the EFR concept has a wider range of Tu to achieve a stable flame propagation. KNOCK BEHAVIOR AT THE UPPER LIMIT In order to analyze the heat release at the knock limit, the ratio of ignition delay to reaction front time τID/τF when knock occurs is used. The reaction front time τF (or chemical time scale) is the time during which the flame propagates by a distance of once its flame thickness, and is calculated as

τF =

δF uL

(3)

where δF is the flame thickness, which is calculated using CHEM1D based on the flame temperature profile 49. The laminar burning velocity is derived from the previously developed correlation 34. This ratio presents the chemical time scale, i.e. a large τF value means a slow chemical reaction. As the chemical time scale increases, the flame front moves slower when crossing the flame thickness which separates the burned zone from the unburned zone 67. Schmid et al. 68 concluded that the turbulent burning velocity could be related to the molecular diffusivity a0 of the mixture and the chemical time scale: S t = 1.2 a0 τ F . In this research, the impact of turbulent combustion is not considered, therefore this parameter is used as a metric to present the speed of combustion for each mixture at different ϕ’. Figure 5 shows the τID/τF ratio (or τRES/τF ratio) at the knock limit, τRES = 1.125 ms. That ratio also gives the number of flame thicknesses over which the flame passed. A higher τID/τF ratio means the residence time is long enough for the flame development. Because τID is a constant, a lower


15

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 48

value of τF increases the τID/τF ratio. Schmid et al. 68 assumed that the rate of heat release is proportional to the inverse of τF, so more heat is released by the flame with a lower τF mixture (or higher τID/τF ratio). As can be seen in the Figure, when auto-ignition occurs, the τID/τF ratio for the air dilution cases is larger than for the EGR dilution cases. When the dilution level increases, the ratio decreases. Therefore, more fuel energy is consumed by the flame and less energy is released by the auto-ignition for air dilution. The fraction of heat release by autoignition increases with the decrease of ϕ’. When the end-gas spontaneously auto-ignites, a rapid release of the remaining fuel energy will generate high-amplitude pressure waves. Because the mixture energy is not identical between the cases, one cannot conclude that the ringing intensity increases for a higher fraction of the auto-ignition heat release. In the final part of this study, a comparison at the same energy content is done to compare the ringing intensity. FLAME STRUCTURES AT THE LOWER LIMIT In this section, the flame structure at the flammability limit (uL = 15 cm/s) was analyzed. The simulation was done at p = 40 bar, ϕ’ = 0.87, and a Tu which was adjusted to ensure a resulting flame speed of 15 cm/s (as shown in Figure 4). The required Tu to achieve a laminar flame speed of 15 cm/s for MeOH-Air, MeOH-EGR, R-EGR, SG50-Air and SG50-EGR cases respectively is 333 K, 398 K, 365 K, 317 K and 385 K. Figure 6 shows the temperature profile of the five mixtures as a function of the axial distance. The burned temperature Tb of the five mixtures is around 2120 – 2160 K. In the research of Flynn et al. 58, the end of combustion flame temperature is estimated using a constant pressure, adiabatic condition, at the point where 95% of the total heat release has occurred. The flame temperature at the point where 95% of the total heat released is calculated, was around 1983 K, 1952 K, 1942 K, 1965 K and 1936 K for MeOHAir, MeOH-EGR, R-EGR, SG50-Air and SG50-EGR, respectively (see symbols in Figure 6).


16

Page 17 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

The flame temperature is lower with the dilution of EGR (same fuel) because of a higher heat capacity of the burned gases. These temperatures are similar to the data cloud in the research of Flynn et al. 58. As can be seen, there is a very small difference between the temperature profiles of MeOH and SG50. Compared to the EGR-type dilution cases (R-EGR, MeOH-EGR and SG50-EGR), although the unburned gas temperature is lower, the air dilution cases have a higher burned gas temperature. The Tb of the R-EGR mixture is higher than that of the normal EGR dilution (MeOH-EGR and SG50-EGR) and lower than the air dilution cases. This can be explained by the concentration of burned products like carbon dioxide (CO2) and water vapor (H2O), species which have higher heat capacity values. In the air dilution cases, because no combustion products are present and there is a greater oxygen concentration, Tb is higher than for the EGR-type dilution cases. Compared to normal EGR dilution, the R-EGR case has a higher Tb due to a smaller concentration of water vapor in the reactant because it was consumed partly for the reforming. The concentrations of four main species (CH3OH, O2, CO2 and H2O) are compared between MeOH-Air and SG50-Air in Figure 7. Because of its replacement by syngas, the mole fraction of CH3OH in the unburned zone of the SG50-Air case is obviously smaller than for the MeOH-Air case. The stoichiometric air to fuel ratio (by mole) of syngas is much smaller than that of methanol (~1.81 compared to 7.14), so the required amount of O2 to have the same ϕ = 0.87 when syngas is added is lower than for pure methanol combustion. However, CH3OH is almost consumed at the same location. CO2 is a product of the fuel reforming process, so the initial CO2 concentration is higher with the addition of syngas. In this study, syngas is a product of steam reforming, therefore the mole fraction of H2O increases faster to a higher concentration for the


17

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 48

enrichment with syngas. With a higher heat capacity of H2O, the combustion temperature of SG50 is lower than methanol if EGR is used to dilute the combustion, as can be seen in Figure 6. To compare the air and EGR dilution, the mole fraction of the mentioned species of the SG50Air case was plotted together with the SG50-EGR case, as shown in Figure 8. In the EGR dilution case, the mole fraction of O2 in the unburned region is much lower than for the air dilution case. Although the simulation is done at a similar ϕ’, the mixture for the EGR dilution cases is stoichiometric (ϕ = 1), so the O2 in the burned zone is almost completely consumed. The combustion products concentrations, CO2 and H2O, are higher in both the unburned and burned zones, leading to a lower combustion temperature. The SG50-EGR case then is compared to the R-EGR one. At ϕ’ = 0.87, the mass fraction of methanol for fuel reforming and the EGR ratio for the SG50-EGR and R-EGR cases are identical. Therefore as can be seen in Figure 9, the molar concentrations of CH3OH, O2 and CO2 are related for these two cases. In the SG50-EGR case, a slightly lower concentration of CH3OH and O2 can be observed in the unburned mixture because of their displacement by H2O, which is not partly consumed for the fuel reforming as in the R-EGR concept. The H2O concentration is smaller in the R-EGR case both in the unburned and burned zones because the water vapor in the EGR mixture is consumed partly for the steam reforming of methanol. The higher H2O concentration in the SG50-EGR case leads to a decrease in the burned temperature (see Figure 6). Figure 10 shows the mole fraction of H2 and CO, two components of syngas, as a function of axial distance through the flame. In the combustion of methanol (MeOH-Air and MeOH-EGR cases), hydrogen only appears as an intermediate species. In these cases, the hydrogen


18

Page 19 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

concentration grows and reaches its peak at around the flame layer position. With the dilution of EGR-type mixture, the hydrogen concentration increases faster and reaches a peak at a higher concentration. The H2 mole fraction in the burned zone is also higher because the H atom is less consumed by H-abstraction reactions at a lower combustion temperature 69. In the R-EGR and SG50 cases, H2 is present as a fuel component, so going from left to right, its concentration falls and almost equals the concentration in the combustion of methanol. Three EGR-type dilution cases (R-EGR, MeOH-EGR and SG50-EGR) have a higher H2 concentration in the burned zone. As can be seen, the consumption rate of H2 in the R-EGR case is a bit faster than that of the SG50-EGR case, which might be due to a smaller concentration of water vapor in the reactant. The CO concentration is also plotted in Figure 10. Although CO is a component of the methanol steam reforming product, CO still is an intermediate species in the R-EGR and SG50 cases. Compared to the EGR-type dilution cases, the air dilution cases have a lower CO concentration, which agrees with the typical trend of CO emission with lambda. The difference in CO mole fraction is very small between methanol and SG50, in both the air and EGR dilution cases. This can be explained as follows. First, the higher gas temperature promotes the CO oxidation, so the case which has a lower Tb results in a higher concentration of CO. The burned temperature of methanol and SG50 are similar, so the CO emission is almost the same between the two fuels. Second, the ratio of hydrogen to methanol (or H/C ratio) increases when syngas is added, leading to the decrease of carbon-related emissions like CO 70. This is similar to the observation in the research of Han et al. 71, where the CO concentration in exhaust gases was higher with EGR dilution compared to air dilution (at the same ϕ’ = 0.8), and there was no clear change in CO emission with varied syngas fraction. INFLUENCE OF GAS PROPERTIES


19

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 48

In the previous parts, the comparison was done at a constant pressure. However, this is not a good basis for the comparison of different mixtures in real engines. In this part, to simulate the condition of the end gas after compression in SI engines, the influence of gas properties is considered. Two scenarios for the intake charge condition at intake valve closure (IVC) were considered in this study: the first case is the same initial pressure and temperature, a pressure of 1 bar and a temperature of 343 K, for all mixtures (constant inlet pressure scenario), the second one is the same initial temperature (343 K) and the same energy content (constant inlet energy scenario). In the first scenario, because the energy content varies with different mixtures as well as ϕ’, there is a penalty on engine performance when it works with a diluted mixture. In the second scenario, the initial pressure of the diluted mixtures at IVC was increased to result in the same amount of energy as for the undiluted mixture. The reference pressure is 1 bar for the stoichiometric methanol-air case without EGR. The specific heat ratio γ as a function of temperature for each mixture at different ϕ’ were calculated using NASA polynomial coefficients. Assuming the compression is adiabatic, the maximum compression is assumed to be reached when the end gas temperature reaches the autoignition temperature of a methanol-air mixture. With the auto-ignition temperature of methanol being 850 K 63, this “compression ratio” is then calculated, using the specific heat ratio γ of a stoichiometric methanol-air mixture at 850 K, as follows:

CR = (γ −1)

Tend = 17.74 TIVC

(4)

This compression ratio is used for all mixtures to calculate the end gas temperature and pressure using the γ value at the end of compression (γ at Tend, as shown in Figure 11). As can be seen in


20

Page 21 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

this Figure, γ decreases with the increase of the EGR ratio in two normal EGR dilution cases (MeOH-EGR and SG50-EGR). Although the mass fraction of methanol (which has the lowest γ) reduced with the increase of EGR ratio, the oxygen mass fraction is also decreased. Therefore, the overall γ of the mixture declines with a higher dilution ratio. This shows a different trend compared to the research of Alger et al. 72, where the cycle-averaged γ is used to compare at different EGR ratios. Due to a lower combustion temperature with EGR dilution, the cycleaveraged γ is increased with increasing EGR ratio. Here, the impact of combustion temperature is neglected, so that γ is a function of only compressive temperature and mixture composition. The end gas temperature and pressure calculated from isentropic compression were then used as the initial condition for the ignition delay prediction. In both scenarios, because the initial temperature and the compression ratio are the same, there is no difference in end gas temperature after compression. In the constant inlet pressure scenario, the end gas pressure follows the trend of γ: decreasing in the two normal EGR dilution cases (MeOH-EGR and SG50-EGR), and increasing in the air dilution and R-EGR cases with the reduction of ϕ’. However, in the constant inlet energy scenario, due to the difference in initial pressure, the end gas pressure increases with the increase in dilution ratio. At ϕ’ = 0.6, the maximum difference between the five mixtures in pressure and temperature is around 6 bar and 35 K respectively. Because of a higher end gas pressure in the second scenario with ϕ’ < 1, the ignition delay is shorter compared to the first scenario, as can be seen in Figure 12. However, the trend is similar in both scenarios: the ignition delay is shorter with lower ϕ’ in the two air dilution cases (MeOH-Air and SG50-Air) and vice versa in the three EGR type dilution cases (MeOH-EGR, R-EGR and SG50-EGR). The trends in the two air dilution cases are reversed versus the trends observed in Figure 3, the ignition delay decreases


21

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 48

instead of increases with a smaller value of ϕ’. This can be explained by the improvement of end gas temperature after compression with a leaner mixture, which is not a constant temperature like was assumed for Figure 3. Based on the ignition delay trends, we can conclude that the knock tendency decreases with EGR type dilution, with R-EGR having a lower knock tendency compared to the SG50-EGR case for ϕ’ ≤ ~0.85. However, the knock tendency increases with a leaner mixture. For the constant inlet energy scenario, this is similar to the conclusion of Topinka et al. 28, however it is opposite to the findings with a constant inlet pressure scenario. This can be explained by a lower peak combustion pressure due to a lower energy content, so the end gas temperature is lower in this scenario. Therefore a longer ignition delay will be observed, and the knock tendency decreases with lean operation 28. As in this research, the impact of flame propagation is neglected, the decrease of the peak pressure (as well as the end gas temperature) cannot be captured. Therefore the increase of ignition delay for a lean mixture is not predictable in the constant inlet pressure scenario. For the constant inlet energy scenario, the effect of flame propagation on peak combustion pressure can be accounted. The peak combustion pressure is dependent on the pressure at ignition timing as well as on the combustion speed. A diluted mixture has a higher pressure at spark timing and a slower burning velocity, based on this tradeoff and the same indicated mean effective pressure (due to the same energy content), we can assume that the peak combustion pressure is similar in all cases. Therefore, the next analyses will be done for the constant inlet energy scenario. Assuming the compression of the end gas is adiabatic, Figure 13 illustrates the end gas temperature versus end gas pressure at different ϕ’ and dilution methods (air vs. EGR) for the constant inlet energy scenario. Ignoring the change of γ during the compression, a specific value of γ for each mixture at 850 K was used to calculate the end gas conditions. Because the energy


22

Page 23 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

content of the diluted mixtures is less than of the undiluted one, a higher initial pressure needs to be introduced, leading to a higher end gas pressure. However, the end gas temperature is dependent on only compression ratio, γ and initial temperature, so the difference in end gas temperature is much smaller than the difference in end gas pressure. Furthermore, as the energy content of the EGR diluted mixture is lower than of the air diluted one (at same ϕ’), the initial pressure of the EGR diluted mixture needs to be higher, so the end gas pressure increases. Therefore, at a given pressure, the end gas temperature is lower with a higher dilution ratio (lower ϕ’) and the end gas temperature for the EGR dilution cases is lower than for the air dilution ones (same ϕ’). Figure 14 shows the pressure and temperature of the end gas when knock occurs (τID = 1.125 ms) with ϕ’ varying from 1 to 0.6 (∆ϕ’ = 0.05 for each step) of the five mixtures for the constant inlet energy scenario. As can be seen in the Figure, the case of methanol at ϕ = 1 experiences knock when the end gas is compressed to a pressure of ~ 72 bar and a temperature of ~ 955 K. For the stoichiometric SG50-air mixture this is ~ 73 bar and ~ 962 K, respectively. As the increase of end gas pressure is more significant than of the end gas temperature, for an EGR diluted mixture, that mixture can be auto-ignited at a lower temperature and higher pressure compared to an undiluted one, as can be seen in Figure 14. In the two air (MeOH-Air and SG50-Air) and the two EGR dilution cases (MeOH-EGR and SG50-EGR), the auto-ignition of the end gas is almost independent of the temperature, with only a very small change in the end gas temperature with a diluted mixture, at ~ 955±5 K and ~ 962±5 K respectively. Knock for the R-EGR case is only experienced at a higher end gas pressure and temperature when ϕ’ decreases, which is mainly due to the increase of the hydrogen concentration with increasing EGR ratio, which is a chemical effect.


23

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 48

Figure 15 shows the τID/τF ratio at the end gas conditions which were presented in Figure 14. The trend is similar to the comparison at 40 bar (see Figure 5), i.e. the ratio decreases with increasing dilution ratio (reducing ϕ’). In this scenario, the energy is similar for all mixtures, therefore knocking can be compared between the different fuels. The amount of heat released by autoignition reduces for a fuel which has a higher τID/τF ratio, therefore the knock intensity reduces. As can be seen, at ϕ’ = 0.87 (in this case, the SG50-EGR and R-EGR cases have the same mass fraction of methanol fed to the catalyst and the same EGR ratio), the τID/τF ratio of the R-EGR mixture is higher than of the SG50-EGR mixture, therefore the pressure oscillation due to knock is reduced for the R-EGR case. At 0.9 ≤ ϕ’ ≤ 1.0, the SG50-EGR case is better than R-EGR in terms of ringing intensity. Compared to lean combustion of methanol, dilution with R-EGR mixture can reduce the ringing intensity. Air dilution has a larger amount of fuel energy consumed by the flame and less energy released by the auto-ignition. Although the knock tendency for the air dilution case is higher, when the knock occurs, the EGR dilution case has a higher ringing intensity 73. The phenomenon is similar to what happens in SACI (spark assisted compression ignition) engines, it is more SI-like combustion with air dilution and more HCCIlike combustion with EGR dilution. When diluted with an R-EGR mixture, it is more SI-like combustion than other EGR dilution cases and the mixture is stoichiometric, so this strategy is recommended for the SI engines to enable an increase in dilution ratios as well as to achieve a high conversion efficiency of the three-way catalyst. Because SI engines have a higher specific energy content of the charge (ϕ’) than SACI engines (0.65 ≤ ϕ’SI ≤ 1 instead of 0.45 ≤ ϕ’SACI ≤ 0.65) 74, once knock develops, very strong knock will be observed. CONCLUSIONS


24

Page 25 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

This paper presents an analysis of the reaction front properties of diluted methanol-air flames when enriched with fuel reformates through 0D/1D simulation at a constant pressure, 40 bar. The simulation showed that the upper limit of the unburned gas temperature is extended with higher dilution levels, especially in the case of the R-EGR mixture. Therefore the R-EGR concept is a promising approach to extend the operating range of high EGR ratio engines. At the knock limit, the fraction of auto-ignition heat release is lower with the dilution of air. The knock propensity is reduced with the addition of syngas. At the flammability limit, compared to EGR dilution, the air dilution cases have a lower Tu limit to get a stable flame. The limit is also extended to a lower value of Tu when syngas is added. With the addition of syngas, the concentration of water vapor in the burned zone increases. It is further improved when EGR mixture is used to dilute the combustion. If the mixture is diluted by R-EGR mixture, the water vapor is reduced in both unburned and burned zones due to the consumption of part of the water for the fuel reforming. The concentration of H2 and CO in the combustion product when using EGR-type dilution is higher than for the air dilution cases. The influence of gas properties is also investigated for a constant inlet energy scenario. After the compression, the end gas condition is significantly changed between the five mixtures, especially at low ϕ’, resulting in a shorter ignition delay for a leaner mixture and a longer one for higher EGR rates. Therefore, the knock tendency increases with the decrease of equivalence ratio and it decreases with the increase of EGR ratio. In the R-EGR case, the knock occurs at a higher pressure and temperature when ϕ’ decreases from 1 to 0.6. Compared to the EGR dilution cases, air dilution has a higher knock tendency. However, the ringing intensity decreases for the dilution with air. At a higher dilution ratio (ϕ’ ≤ 0.85), R-EGR has a lower ringing intensity


25

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 48

compared to the two normal EGR dilution cases and a lower knock tendency compared to the two air dilution cases.

AUTHOR INFORMATION Corresponding Author * Tel.: +32 9 264 34 53. Fax: +32 9 264 35 75. Email: [email protected] Notes The authors declare no competing financial interest. ACKNOWLEDGMENT The Special Research Fund (BOF) of Ghent University is gratefully acknowledged for the starting grant (no. 01N03013). ABBREVIATIONS EFR, external fuel reforming; EGR, exhaust gas recirculation; HCCI, homogeneous charge compression ignition; ICE, internal combustion engine; PM, particulate matter; R-EGR, reformed exhaust gas recirculation; RON, research octane number; SACI, spark assisted compression ignition; SG, syngas; SI, spark ignition; TWC, three way catalyst. REFERENCES 1. Reitz, R. D., Grand Challenges in Engine and Automotive Engineering. Frontiers in Mechanical Engineering 2015, 1 (1), 1-3. 2. Condliffe, J. Four Huge Cities Are Banning Diesel Cars. https://www.technologyreview.com/s/603026/four-huge-cities-are-banning-diesel-cars/. 3. Reilly, M. Global Air Pollution Is Getting Worse, but Removing It Could Worsen Climate Change. https://www.technologyreview.com/s/601455/global-air-pollution-is-gettingworse-but-removing-it-could-worsen-climate-change/.


26

Page 27 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

4. Grandin, B.; Ångström, H.-E., Replacing Fuel Enrichment in a Turbo Charged SI Engine: Lean Burn or Cooled EGR. SAE Technical Paper 1999-01-3505, 1999. 5. Caton, J. A., A Comparison of Lean Operation and Exhaust Gas Recirculation: Thermodynamic Reasons for the Increases of Efficiency. SAE Technical Paper 2013-01-0266, 2013. 6. Verhelst, S., Future vehicles will be driven by electricity, but not as you think [Point of View]. Proceedings of the IEEE 2014, 102 (10), 1399-1403. 7. Balki, M. K.; Sayin, C.; Sarikaya, M., Optimization of the operating parameters based on Taguchi method in an SI engine used pure gasoline, ethanol and methanol. Fuel 2016, 180, 630637. 8. Sileghem, L.; Ickes, A.; Wallner, T.; Verhelst, S., Experimental Investigation of a DISI Production Engine Fuelled with Methanol, Ethanol, Butanol and ISO-Stoichiometric Alcohol Blends. SAE Technical Paper 2015-01-0768, 2015. 9. Vancoillie, J.; Demuynck, J.; Sileghem, L.; Van de Ginste, M.; Verhelst, S., Comparison of the renewable transportation fuels, hydrogen and methanol formed from hydrogen, with gasoline - Engine efficiency study. International Journal of Hydrogen Energy 2012, 37 (12), 9914-9924. 10. Vancoillie, J.; Demuynck, J.; Sileghem, L.; Van De Ginste, M.; Verhelst, S.; Brabant, L.; Van Hoorebeke, L., The potential of methanol as a fuel for flex-fuel and dedicated spark-ignition engines. Appl Energ 2013, 102, 140-149. 11. Vancoillie, J.; Sileghem, L.; Van de Ginste, M.; Demuynck, J.; Galle, J.; Verhelst, S., Experimental Evaluation of Lean-burn and EGR as Load Control Strategies for Methanol Engines. SAE Technical Paper 2012-01-1283, 2012. 12. Xie, F. X.; Li, X. P.; Wang, X. C.; Su, Y.; Hong, W., Research on using EGR and ignition timing to control load of a spark-ignition engine fueled with methanol. Appl Therm Eng 2013, 50 (1), 1084-1091. 13. Verhelst, S.; Wallner, T., Hydrogen-fueled internal combustion engines. Prog Energ Combust 2009, 35 (6), 490-527. 14. Tollefson, J., Hydrogen vehicles: Fuel of the future? Nature 2010, 464 (7293), 12621264. 15. F. Brown, L., A comparative study of fuels for on-board hydrogen production for fuelcell-powered automobiles. International Journal of Hydrogen Energy 2001, 26 (4), 381-397. 16. Szybist, J. P.; Chakravathy, K.; Daw, C. S., Analysis of the Impact of Selected Fuel Thermochemical Properties on Internal Combustion Engine Efficiency. Energ Fuel 2012, 26 (5), 2798-2810. 17. Lindner, B.; Sjöström, K., Operation of an internal combustion engine: lean conditions with hydrogen produced in an onboard methanol reforming unit. Fuel 1984, 63 (11), 1485-1490. 18. Pettersson, L.; Sjöström, K., An experimental and theoretical evaluation of the onboard decomposed methanol spark-ignition engine. Combustion Science and Technology 1990, 71 (13), 129-143. 19. Pettersson, L.; Sjöström, K., Onboard hydrogen generation by methanol decomposition for the cold start of neat methanol engines. International Journal of Hydrogen Energy 1991, 16 (10), 671-676. 20. Brinkman, N. D.; Stebar, R. F., A Comparison of Methanol and Dissociated Methanol Illustrating Effects of Fuel Properties on Engine Efficiency—Experiments and Thermodynamic Analyses. SAE Technical Paper 850217, 1985.


27

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 48

21. Jamal, Y.; Wyszynski, M. L., On-board generation of hydrogen-rich gaseous fuels—a review. International Journal of Hydrogen Energy 1994, 19 (7), 557-572. 22. Pettersson, L.; Sjöström, K., Decomposed Methanol as a Fuel—A review. Combustion Science and Technology 1991, 80 (4-6), 265-303. 23. McCall, D. M.; Lalk, T. R.; Davison, R. R.; Harris, W. B., Performance and Emissions Characteristics of a Spark Ignition Engine Fueled with Dissociated and Steam-Reformed Methanol. SAE Technical Paper 852106, 1985. 24. Tartakovsky, L.; Amiel, R.; Baibikov, V.; Fleischman, R.; Gutman, M.; Poran, A.; Veinblat, M., SI Engine with Direct Injection of Methanol Reforming Products–First Experimental Results. SAE Techincal Paper 2015-32-0712, 2015. 25. Poran, A.; Tartakovsky, L., Performance and emissions of a direct injection internal combustion engine devised for joint operation with a high-pressure thermochemical recuperation system. Energy 2017, 124, 214-226. 26. Poran, A.; Tartakovsky, L., Influence of methanol reformate injection strategy on performance, available exhaust gas enthalpy and emissions of a direct-injection spark ignition engine. International Journal of Hydrogen Energy 2017, 42 (23), 15652-15668. 27. Nguyen, D.-K.; Verhelst, S., The temperature dependence of laminar burning velocities of methanol-syngas-air flames. In FISITA 2016 World Automotive Congress, Busan, South Korea, 2016. 28. Topinka, J. A.; Gerty, M. D.; Heywood, J. B.; Keck, J. C., Knock behavior of a leanburn, H2 and CO enhanced, SI gasoline engine concept. SAE Technical Paper 2004-01-0975, 2004. 29. Cracknell, R.; Prakash, A.; Head, R., Influence of Laminar Burning Velocity on Performance of Gasoline Engines. SAE Technical Paper 2012-01-1742, 2012. 30. Kolodziej, C. P.; Pamminger, M.; Sevik, J.; Wallner, T.; Wagnon, S. W.; Pitz, W. J., Effects of Fuel Laminar Flame Speed Compared to Engine Tumble Ratio, Ignition Energy, and Injection Strategy on Lean and EGR Dilute Spark Ignition Combustion. SAE Int. J. Fuels Lubr. 2017, 10 (1), 82-94. 31. Peppley, B. A.; Amphlett, J. C.; Kearns, L. M.; Mann, R. F., Methanol–steam reforming on Cu/ZnO/Al2O3. Part 1: the reaction network. Applied Catalysis A: General 1999, 179 (1–2), 21-29. 32. Twigg, M. V.; Spencer, M. S., Deactivation of Copper Metal Catalysts for Methanol Decomposition, Methanol Steam Reforming and Methanol Synthesis. Topics in Catalysis 2003, 22 (3), 191-203. 33. Papavasiliou, J.; Avgouropoulos, G.; Ioannides, T., In situ combustion synthesis of structured Cu-Ce-O and Cu-Mn-O catalysts for the production and purification of hydrogen. Appl Catal B-Environ 2006, 66 (3-4), 168-174. 34. Nguyen, D.-K.; Verhelst, S., Development of laminar burning velocity correlation for the simulation of methanol fueled SI engines operated with onboard fuel reformer. SAE Technical Paper 2017-01-0539, 2017. 35. Leung, P.; Tsolakis, A.; Rodríguez-Fernández, J.; Golunski, S., Raising the fuel heating value and recovering exhaust heat by on-board oxidative reforming of bioethanol. Energy & Environmental Science 2010, 3 (6), 780-788. 36. Gomes, S. R.; Bion, N.; Blanchard, G.; Rousseau, S.; Duprez, D.; Epron, F., Study of the main reactions involved in reforming of exhaust gas recirculation (REGR) in gasoline engines. RSC Advances 2011, 1 (1), 109-116.


28

Page 29 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

37. Fennell, D.; Herreros, J.; Tsolakis, A., Improving gasoline direct injection (GDI) engine efficiency and emissions with hydrogen from exhaust gas fuel reforming. International Journal of Hydrogen Energy 2014, 39 (10), 5153-5162. 38. Gomes, S. R.; Bion, N.; Blanchard, G.; Rousseau, S.; Bellière-Baca, V.; Harlé, V.; Duprez, D.; Epron, F., Thermodynamic and experimental studies of catalytic reforming of exhaust gas recirculation in gasoline engines. Applied Catalysis B: Environmental 2011, 102 (1– 2), 44-53. 39. Zheng, Z.; Liu, C.; Zhang, X., Numerical study of effects of reformed exhaust gas recirculation (REGR) on dimethyl ether HCCI combustion. International Journal of Hydrogen Energy 2014, 39 (15), 8106-8117. 40. Ashida, K.; Maeda, H.; Araki, T.; Hoshino, M.; Hiraya, K.; Izumi, T.; Yasuoka, M., Study of an On-board Fuel Reformer and Hydrogen-Added EGR Combustion in a Gasoline Engine. SAE Int. J. Fuels Lubr. 2015, 8 (2), 358-366. 41. Tsolakis, A.; Megaritis, A.; Yap, D.; Abu-Jrai, A., Combustion Characteristics and Exhaust Gas Emissions of a Diesel Engine Supplied with Reformed EGR. SAE Technical Paper 2005-01-2087, 2005. 42. Abu-Jrai, A.; Rodríguez-Fernández, J.; Tsolakis, A.; Megaritis, A.; Theinnoi, K.; Cracknell, R. F.; Clark, R. H., Performance, combustion and emissions of a diesel engine operated with reformed EGR. Comparison of diesel and GTL fuelling. Fuel 2009, 88 (6), 10311041. 43. Tsolakis, A.; Megaritis, A.; Wyszynski, M. L., Application of Exhaust Gas Fuel Reforming in Compression Ignition Engines Fueled by Diesel and Biodiesel Fuel Mixtures. Energ Fuel 2003, 17 (6), 1464-1473. 44. Rodríguez-Fernández, J.; Tsolakis, A.; Cracknell, R. F.; Clark, R. H., Combining GTL fuel, reformed EGR and HC-SCR aftertreatment system to reduce diesel NOx emissions. A statistical approach. International Journal of Hydrogen Energy 2009, 34 (6), 2789-2799. 45. Jamal, Y.; Wagner, T.; Wyszynski, M. L., Exhaust gas reforming of gasoline at moderate temperatures. International Journal of Hydrogen Energy 1996, 21 (6), 507-519. 46. CHEM1D A one-dimensional laminar flame code, Eindhoven University of Technology. https://www.tue.nl/en/university/departments/mechanical-engineering/research/researchgroups/multiphase-and-reactive-flows/our-expertise/research-topics/chem1d/. 47. Hermanns, R. T. E. Laminar burning velocities of methane-hydrogen-air mixtures. PhD Thesis, Eindhovend University of Technology, 2007. 48. Alexandre Ern, V. G. EGLIB - A Multicomponent Transport Software for Fast and Accurate Evaluation Algorithms. http://www.cmap.polytechnique.fr/www.eglib/. 49. Jarosinski, J., The Thickness of Laminar Flames. Combust Flame 1984, 56 (3), 337-342. 50. Goodwin, D. G.; Moffat, H. K.; Speth, R. L. Cantera: An object- oriented software toolkit for chemical kinetics, thermodynamics, and transport processes. http://www.cantera.org. 51. Li, J.; Zhao, Z.; Kazakov, A.; Chaos, M.; Dryer, F. L.; Scire, J. J., A comprehensive kinetic mechanism for CO, CH2O, and CH3OH combustion. International Journal of Chemical Kinetics 2007, 39 (3), 109-136. 52. Burke, U.; Metcalfe, W. K.; Burke, S. M.; Heufer, K. A.; Dagaut, P.; Curran, H. J., A detailed chemical kinetic modeling, ignition delay time and jet-stirred reactor study of methanol oxidation. Combust Flame 2016, 165, 125-136. 53. Kéromnès, A.; Metcalfe, W. K.; Heufer, K. A.; Donohoe, N.; Das, A. K.; Sung, C.-J.; Herzler, J.; Naumann, C.; Griebel, P.; Mathieu, O.; Krejci, M. C.; Petersen, E. L.; Pitz, W. J.;


29

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 48

Curran, H. J., An experimental and detailed chemical kinetic modeling study of hydrogen and syngas mixture oxidation at elevated pressures. Combust Flame 2013, 160 (6), 995-1011. 54. Olm, C.; Zsely, I. G.; Varga, T.; Curran, H. J.; Turanyi, T., Comparison of the performance of several recent syngas combustion mechanisms. Combust Flame 2015, 162 (5), 1793-1812. 55. Martz, J. B.; Middleton, R. J.; Lavoie, G. A.; Babajimopoulos, A.; Assanis, D. N., A computational study and correlation of premixed isooctane-air laminar reaction front properties under spark ignited and spark assisted compression ignition engine conditions. Combust Flame 2011, 158 (6), 1089-1096. 56. Middleton, R. J.; Martz, J. B.; Lavoie, G. A.; Babajimopoulos, A.; Assanis, D. N., A computational study and correlation of premixed isooctane air laminar reaction fronts diluted with EGR. Combust Flame 2012, 159 (10), 3146-3157. 57. Lavoie, G. A.; Martz, J.; Wooldridge, M.; Assanis, D., A multi-mode combustion diagram for spark assisted compression ignition. Combust Flame 2010, 157 (6), 1106-1110. 58. Flynn, P. F.; Hunter, G. L.; Durrett, R. P.; Farrell, L. A.; Akinyemi, W. C., Minimum Engine Flame Temperature Impacts on Diesel and Spark-Ignition Engine NOx Production. SAE Technical Paper 2000-01-1177, 2000. 59. Sileghem, L.; Alekseev, V. A.; Vancoillie, J.; Nilsson, E. J. K.; Verhelst, S.; Konnov, A. A., Laminar burning velocities of primary reference fuels and simple alcohols. Fuel 2014, 115, 32-40. 60. Sileghem, L.; Alekseev, V. A.; Vancoillie, J.; Van Geem, K. M.; Nilsson, E. J. K.; Verhelst, S.; Konnov, A. A., Laminar burning velocity of gasoline and the gasoline surrogate components iso-octane, n-heptane and toluene. Fuel 2013, 112, 355-365. 61. Lewis, A.; Ortiz-Soto, E.; Lavoie, G.; Assanis, D. N., Scaling and dimensional methods to incorporate knock and flammability limits in models of high-efficiency gasoline and ethanol engines. International Journal of Engine Research 2015, 16 (2), 181-196. 62. Kalghatgi, G.; Algunaibet, I.; Morganti, K., On Knock Intensity and Superknock in SI Engines. SAE Int. J. Engines 2017, 10 (3). 63. Kumar, K.; Sung, C. J., Autoignition of methanol: experiments and computations. International Journal of Chemical Kinetics 2011, 43 (4), 175-184. 64. He, X.; Donovan, M. T.; Zigler, B. T.; Palmer, T. R.; Walton, S. M.; Wooldridge, M. S.; Atreya, A., An experimental and modeling study of iso-octane ignition delay times under homogeneous charge compression ignition conditions. Combust Flame 2005, 142 (3), 266-275. 65. Yates, A.; Bell, A.; Swarts, A., Insights relating to the autoignition characteristics of alcohol fuels. Fuel 2010, 89 (1), 83-93. 66. Gauthier, B. M.; Davidson, D. F.; Hanson, R. K., Shock tube determination of ignition delay times in full-blend and surrogate fuel mixtures. Combust Flame 2004, 139 (4), 300-311. 67. Strozzi, C.; Mura, A.; Sotton, J.; Bellenoue, M., Experimental analysis of propagation regimes during the autoignition of a fully premixed methane–air mixture in the presence of temperature inhomogeneities. Combust Flame 2012, 159 (11), 3323-3341. 68. Schmid, H.-P.; Habisreuther, P.; Leuckel, W., A Model for Calculating Heat Release in Premixed Turbulent Flames. Combust Flame 1998, 113 (1), 79-91. 69. Gu, X.; Huang, Z.; Cai, J.; Gong, J.; Wu, X.; Lee, C.-F., Emission characteristics of a spark-ignition engine fuelled with gasoline-n-butanol blends in combination with EGR. Fuel 2012, 93, 611-617.


30

Page 31 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

70. Hu, E.; Huang, Z.; Liu, B.; Zheng, J.; Gu, X.; Huang, B., Experimental investigation on performance and emissions of a spark-ignition engine fuelled with natural gas–hydrogen blends combined with EGR. International Journal of Hydrogen Energy 2009, 34 (1), 528-539. 71. Han, T.; Lavoie, G.; Wooldridge, M.; Boehman, A., Effect of Syngas (H2/CO) on SI Engine Knock under Boosted EGR and Lean Conditions. SAE Int. J. Engines 2017, 10 (3). 72. Alger, T.; Gingrich, J.; Roberts, C.; Mangold, B., Cooled exhaust-gas recirculation for fuel economy and emissions improvement in gasoline engines. International Journal of Engine Research 2011, 12 (3), 252-264. 73. Olesky, L. M.; Martz, J. B.; Lavoie, G. A.; Vavra, J.; Assanis, D. N.; Babajimopoulos, A., The effects of spark timing, unburned gas temperature, and negative valve overlap on the rates of stoichiometric spark assisted compression ignition combustion. Appl Energ 2013, 105, 407-417. 74. Lavoie, G. A.; Ortiz-Soto, E.; Babajimopoulos, A.; Martz, J. B.; Assanis, D. N., Thermodynamic sweet spot for high-efficiency, dilute, boosted gasoline engines. International Journal of Engine Research 2013, 14 (3), 260-278.


31

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 48

Figure captions Figure 1. Schematic diagram of two reforming engine concepts. a) external fuel reforming (EFR), b) reformed exhaust gas recirculation (R-EGR). Figure 2. Comparison of ignition delay of methanol and syngas (50% H2-50% CO) from experimental studies (symbols) of methanol 52 and syngas 53 vs. the simulated data (lines). Figure 3. The τID versus ϕ’ at p = 40 bar, Tu = 850 K. Simulated data with Li’s mechanism 51 (lines) vs. calculated ignition delay of iso-octane and ethanol at lean condition using Arrhenius correlations 64-65 (symbols). Figure 4. The stable combustion diagram of five mixtures in Tu-ϕ’ domain at 40 bar. The upper limit lines separate the knock and normal combustion, and the lower limit lines separate the stable (normal) and unstable combustion (quenching). Figure 5. The τID/τF ratio versus ϕ’ when knock occurs (τID = 1.125 ms) at 40 bar. Figure 6. The temperature profiles at lower flammability limit, p = 40 bar and ϕ’ = 0.87. Symbols: the flame temperature and location where 95% of total heat release is occurred. Figure 7. Comparison of species mole fraction between MeOH-Air and SG50-Air cases at ϕ’ = 0.87. Figure 8. Comparison of species mole fraction between SG50-Air and SG50-EGR cases at ϕ’ = 0.87. Figure 9. Comparison of species mole fraction between SG50-EGR and R-EGR cases at ϕ’ = 0.87.


32

Page 33 of 48

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 10. Comparison of H2 and CO mole fraction between five cases at ϕ’ = 0.87. Figure 11. Specific heat ratios as a function of ϕ’ of five mixtures after the compression. The initial gas temperature is 343 K and compression ratio of 17.74. Figure 12. Influence of gas properties on the ignition delay in two cases. Case 1: similar initial pressure of 1 bar (symbols) vs. case 2: similar energy content (lines). Figure 13. The end gas temperature vs. end gas pressure at different ϕ’ and dilution methods for the constant inlet energy scenario. Figure 14. The end gas conditions when knock occurs (τID = 1.125 ms) for the constant inlet energy scenario. Figure 15. The τID/τF ratio versus ϕ’ when knock occurs (τID = 1.125 ms) for the constant inlet energy scenario.


33

Energy & Fuels

condenser

a)

Hydrogen rich gas (wet)

Fuel tank (Methanol)

Intake

Water

reformer catalyst

ENGINE

Exhaust heat transfer

b)

Reformed EGR

reformer catalyst

Fuel tank (Methanol)

Intake

ENGINE

CO2, H2O, N2, CO, NOx…

Exhaust Gas Recirculation (EGR)

EGR

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 48

Exhaust

Fig 1.


34

Page 35 of 48

1E+5 Syngas_Sim 16 bar Syngas_Exp 16 bar Syngas_Sim 32 bar

Ignition delay (µs)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1E+4

Syngas_Exp 32 bar MeOH_Sim 30 atm MeOH_Exp 30 atm MeOH_Sim 50 atm

1E+3

MeOH_Exp 50 atm

pre-ignition

1E+2

1E+1 0.8

0.9

1

1.1

1000/T (K-1)

Fig. 2


35

Energy & Fuels

90 p = 40 bar Tu = 850 K MeOH-Air MeOH-EGR R-EGR SG50-Air SG50-EGR iC8H18-Air EtOH-Air

70

τID (ms)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 48

50

30

10 0.6

0.7

0.8

0.9

1

ϕ' (-)

Fig 3.


36

Page 37 of 48

1050 p = 40 bar 1025

knock 1000 τID = 1.125 ms 975

Tu (K)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

800 950

500

MeOH-Air MeOH-EGR R-EGR SG50-Air SG50-EGR

400

u L = 15 cm/s

stable combustion 700 600

300

unstable combustion

200 0.6

0.7

0.8

0.9

1

ϕ' (-)

Fig 4.


37

Energy & Fuels

400 MeOH-Air MeOH-EGR R-EGR SG50-Air SG50-EGR

300

τID / τF (-)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 48

p = 40 bar

200

100 τID = 1.125 ms 0 0.6

0.7

0.8

0.9

1

ϕ' (-)

Fig 5.


38

Page 39 of 48

2500 p = 40 bar, ϕ' = 0.87, uL = 15 cm/s

1500

1000

500

Tb


2000

Temperature (K)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

2000 1950

Tu

1900 0.002

0 -0.002

0

0.002

0.003

0.004

0.004

0.006

Axial distance (cm)

Fig 6.


39

Energy & Fuels

0.25

H2O

ϕ' = 0.87

Species mole fraction (-)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 48

0.2

0.15

O2

MeOH-Air SG50-Air

CH3OH

CO2

0.1

0.05

0 -0.002

0

0.002

0.004

0.006

Axial distance (cm)

Fig 7.


40

Page 41 of 48

0.25 ϕ' = 0.87


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

0.2

H2O O2 SG50-Air

0.15

SG50-EGR

CO2 0.1

CH3OH

0.05

0 -0.002

0

0.002

0.004

0.006

Axial distance (cm)

Fig 8.


41

Energy & Fuels

0.25 ϕ' = 0.87


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 42 of 48

H2O

0.2

0.15

SG50-EGR

O2

R-EGR

CO2 0.1

CH3OH

0.05

0 -0.002

0

0.002

0.004

0.006

Axial distance (cm)

Fig 9.


42

Page 43 of 48

0.06 ϕ' = 0.87


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

0.05

CO

0.04 MeOH-Air

0.03

MeOH-EGR R-EGR

0.02

SG50-Air SG50-EGR

0.01 0 -0.006

H2 -0.004

-0.002

0

0.002

0.004

0.006

Axial distance (cm)

Fig 10.


43

Energy & Fuels

1.33 MeOH-Air MeOH-EGR 1.325

R-EGR SG50-Air

γ (-)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 44 of 48

SG50-EGR 1.32

1.315

1.31 0.6

0.7

0.8

0.9

1

ϕ' (-)

Fig. 11


44

Page 45 of 48

70 1 MeOH-Air 60

2 MeOH-EGR 3 R-EGR

50

4 SG50-Air

τ ID (ms)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

40

5 SG50-EGR

30 20 10 0 0.6

0.7

0.8

0.9

1

ϕ' (-)

Fig. 12


45

Energy & Fuels

1100

Constant inlet energy scenario End gas temperature (K)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 46 of 48

1050

ϕ' decrease

1000 950 900

φ' = 1.0 φ' = 0.8 - Air φ' = 0.8 - EGR φ' = 0.6 - Air φ' = 0.6 - EGR

850 800 750 50

70

90

110

130

End gas pressure (bar)

Fig. 13


46

Page 47 of 48

150

Constant inlet energy scenario τID = 1.125 ms 130

Pressure (bar)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

110

MeOH-Air

90

MeOH-EGR R-EGR

70

SG50-Air SG50-EGR

ϕ' = 1.0 50 940

950

960

970

980

990

1000

1010

Temperature (K)

Fig. 14


47

Energy & Fuels

140

Constant inlet energy scenario τID = 1.125 ms

120


100

τID / τF (-)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 48 of 48

80 60 40 20 0 0.6

0.7

0.8

0.9

1

ϕ' (-)

Fig. 15


48

Computational Study of the Laminar Reaction Front Properties of

Recommend Documents