Development and validation of a reduced chemical kinetic model …and... · 2015-08-09 · fundamental combustion experimental data. However, because the computational time of combustion

Fuel 160 (2015) 165–177

Contents lists available at ScienceDirect

Fuel

journal homepage: www.elsevier .com/locate / fuel

Development and validation of a reduced chemical kinetic modelfor dimethyl ether combustion

http://dx.doi.org/10.1016/j.fuel.2015.07.0660016-2361/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: State Key Laboratory of Multiphase Flows in PowerEngineering, Xi’an Jiaotong University, Xi’an 710049, People’s Republic of China.

E-mail address: [email protected] (L. Pan).

Lun Pan a,b,⇑, Sage Kokjohn b, Zuohua Huang a

a State Key Laboratory of Multiphase Flows in Power Engineering, Xi’an Jiaotong University, Xi’an 710049, People’s Republic of Chinab Engine Research Center, University of Wisconsin–Madison, 1500 Engineering Drive, Madison, WI 53705, USA

h i g h l i g h t s

� A reduced DME mechanism is proposed for ignition delay and combustion predictions.� The reduced mechanism was validated against experimental ignition delay times.� The reduced mechanism was validated against the parent mechanism.� The reduced mechanism was validated against engine and emissions data.

a r t i c l e i n f o

Article history:Received 13 May 2015Received in revised form 14 July 2015Accepted 20 July 2015Available online 1 August 2015

Keywords:Dimethyl etherReduced reaction mechanismCFDCombustion

a b s t r a c t

A new, reduced reaction mechanism for DME combustion in internal combustion engines is proposed.The new mechanism is based on a detailed DME mechanism suitable for low to high temperature ranges.The reduced model consists of 29 species and 66 reactions and contains a detailed H2/CO mechanism andreduced C1–C2 chemistry. The reduced mechanism inherits the major reaction pathways of the detailedmechanism, which ensures its predictive capability when coupled into CFD simulations. The performanceof the reduced mechanism was compared to simulation results of the detailed mechanism, ignition delaytimes from a shock tube and a rapid compression machine, and DME engine combustion data. Overall, thereduced mechanism shows a good balance between accuracy and computational efficiency necessary foruse in combustion system design applications.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Increasingly stringent pollutant regulations and increasingenergy demand have driven the need for development ofhigh-efficiency combustion technologies and alternative fuels forcleaner, more efficient, environmentally sustainable combustion.Recently, highly premixed, low temperature combustion (LTC)technologies including homogeneous charge compression ignition(HCCI), premixed charge compression ignition (PCCI) and reactivitycontrolled compression ignition (RCCI) have attracted great atten-tion, because they can provide higher thermal efficiency, lowernitrogen oxide (NOx) and soot emissions than conventional dieselengines [1]. The combustion event, performance, and emissionscharacteristics of these novel combustion technologies are typi-cally controlled by the fuels auto-ignition characteristics, high-lighting the importance of the fuel’s chemical properties [2].

Thus, accurate chemical kinetic mechanisms are of great impor-tance for simulation of advanced engine combustion concepts [3].

Dimethyl ether (DME) is a promising alternative fuel for com-pression ignition engines since it can provide low PM or soot emis-sions and be synthesized from emerging renewable energysources, such as biomass and existing fossil fuel sources [4].Furthermore, DME offers favorable combustion characteristicsincluding easily auto-ignited due to a high cetane number andsoot-free combustion due to the easy vaporization and no car-bon–carbon bond [5]. Accordingly, many DME spray [6–12] andengine investigations [5,13–19] have been performed in recentyears. Suh and Lee [6] investigated the macroscopic and atomiza-tion characteristics of DME and diesel fuel using a common-railinjection system in a compression ignition engine. They found thatthe DME spray is wider, shorter, and evaporates more rapidly thandiesel spray under identical injection conditions. Recently, Konnoet al. [20] studied DME spray characteristics at injection pressuresup to 140 MPa in a constant volume vessel under engine-like tem-perature/pressure conditions. Their results confirmed the results ofSuh and Lee [6], showing that the evaporation of DME is much

http://crossmark.crossref.org/dialog/?doi=10.1016/j.fuel.2015.07.066&domain=pdf

http://dx.doi.org/10.1016/j.fuel.2015.07.066

mailto:[email protected]


http://www.sciencedirect.com/science/journal/00162361

http://www.elsevier.com/locate/fuel

Nomenclature

DME dimethyl etherCFD computational fluid dynamicLTC low temperature combustionHCCI homogeneous charge compression ignitionPCCI premixed charge compression ignitionRCCI reactivity controlled compression ignitionNOx nitrogen oxideHC hydrocarbonCO carbon monoxideJSR jet stirred reactors

IC engine internal combustion engineNTC negative temperature coefficientRCM rapid compression machineST shock tubeIMEP indicated mean effective pressureSOI start of injectionATDC after top dead centerTDC top dead centerERC Engine Research Center

166 L. Pan et al. / Fuel 160 (2015) 165–177

faster than that of diesel fuel, resulting in a more evenly dis-tributed spray. Youn et al. [21] compared the combustion andemissions characteristics of DME to conventional diesel fuels in afour-cylinder compression ignition engine under various operatingconditions. They concluded that, when operated at the sameengine load, the DME fueled engine has a higher peak combustionpressure and shorter ignition delay than the diesel fueled engine.They also found that the DME fueled engine provides lower soot,hydrocarbon (HC) and carbon monoxide (CO) emissions andslightly higher NOx emission than the diesel fueled engine.Recently, Park et al. [18] reviewed the physical and chemical prop-erties, spray atomization characteristics, combustion, and exhaust

Fig. 1. Reaction pathway diagram for stoichiometric DM

emission characteristics of DME and concluded that DME is apromising alternative fuel for internal combustion engines and itcan be used in place of conventional diesel fuel in compressionignition engines.

Beyond spray and engine studies, DME has been the subject ofnumerous fundamental experimental studies, including ignitiondelay times [22–30], premixed laminar flames [31–36], pyrolysischemistry and oxidation chemistry [36–39]. These experimentalstudies are of great value for validation and construction of chem-ical kinetic mechanisms describing DME oxidation. Dagaut et al.[39] proposed a semi-detailed kinetic model of DME containing43 species and 286 reactions. It was able to reproduce species

E/air mixture at p = 50 atm using NUIG Mech_56.54.

Table 1Main high and low temperature reactions of the reduced DME mechanism.

Number Reaction

Main reactions in high temperature rangeR1 CH3OCH3(+M), CH3 + CH3O(+M)R2 CH3OCH3 + OH, CH3OCH2 + H2OR3 CH3OCH3 + H, CH3OCH2 + H2

R4 CH3OCH3 + CH3, CH3OCH2 + CH4

R5 CH3OCH3 + HO2, CH3OCH2 + H2O2

R6 CH3OCH3 + CH3O2, CH3OCH2 + CH3O2HR7 CH3OCH2, CH3 + CH2O

Main reactions in low temperature rangeR8 CH3OCH3 + O2, CH3OCH2 + HO2

R9 CH3OCH2 + O2, CH3OCH2O2

R10 CH3OCH2O2, CH2OCH2O2HR11 2CH3OCH2O2) O2 + 2CH3OCH2OR12 CH2OCH2O2H, 2CH2O + OHR13 CH3O + CH2O, CH3OCH2OR14 CH2OCH2O2H + O2, O2CH2OCH2O2HR15 O2CH2OCH2O2H, HO2CH2OCHO + OHR16 HO2CH2OCHO, OCH2OCHO + OHR17 OCH2OCHO, HOCH2OCOR18 HOCH2OCO, CH2OH + CO2

L. Pan et al. / Fuel 160 (2015) 165–177 167

profiles of DME in a Jet-Stirred Reactor (JSR). Late, anothersemi-detailed kinetic model of DME, consisting of 78 chemical spe-cies and 336 chemical reactions, was proposed by Curran et al.[40]. This model was validated against the ignition delay timesby Pfahl et al. [27] and species profiles by Daguat et al. [39].Daguat et al. [41] incorporated low-temperature pathways intohigh-temperature pathways of the DME mechanism developedby Curran et al. [40]. The updated model consists of 55 speciesand 331 reactions. The updated mechanism was validated againstboth low- and high-temperature oxidation of DME in a JSR and theignition delay times of DME in shock tubes. Fisher et al. [42] andCurran et al. [43] further refined the DME mechanism developedby Curran et al. [40] and performed validation using pyrolysisresults in a flow reactor at both low-temperature andhigh-temperature conditions. Recently, a comprehensive DMEpyrolysis and oxidation model was proposed by Zhao et al. [44].This model was validated with a large array of experimental data.A new detailed chemical kinetic model (NUIG Mech_56.54) whichtreats most of the low-temperature reactions as a pressure depen-dence was proposed by Burke et al. [30] proposed to represent

100

101

102

103

104

105

DME/Air mixtures

p = 13 barp = 23 barp = 40 bar

Symbols: Experimental dataThick Lines: Current reduced modelThin Lines: NUIG Mech_56.54

Igni

tion

Del

ay T

ime

(us)

1000/T (K-1)

(a) (

0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5

Fig. 2. Comparison of ignition delay predictions between the detailed DME mechanismpressures (a) and equivalence ratios (b).

their new measured data and data available in the literature. Thisdetailed model shows excellent performance over a range of condi-tions. Further validation of this model can be found in Pan et al.[45].

Currently, computational fluid dynamic (CFD) simulations showincreasing importance in increasing the understanding of combus-tion processes and development of IC engine technologies.Therefore, it is essential to develop reliable chemical reactionmechanisms that can represent combustion and in-cylinder pro-cesses and minimize computational expense. As reviewed above,there are numerous detailed mechanisms of DME that can capturefundamental combustion experimental data. However, because thecomputational time of combustion chemistry scales proportionallyto the square of the total number of species in the reaction mech-anism, it is too costly to use these detailed chemical kinetic mech-anisms for engine design and optimization using detailed CFDmodeling. Accordingly, efforts have been made in development ofsimplified DME models. Yamada et al. [46] developed a simplifiedreaction model of DME oxidation with 23 reactions and 23 speciesby extracting essential elementary reactions from the detailedmechanism by Curran et al. [43]. Meanwhile, Kim et al. [47] devel-oped a reduced chemical kinetic mechanism consisting of 45 reac-tions and 28 species by using the overall reaction scheme of DMEbased on the detailed mechanism by Curran et al. [43]. Later, Liang[48] proposed another reduced model of DME based on thedetailed kinetics model of Curran et al. [43] by analyzing reactionpathways and using sensitivity analysis. The proposed mechanismgives good agreement with the detailed model of ignition timing,cylinder gas temperature, and cylinder pressure. Recently, Chinet al. [49] developed a 28-species reduced chemical kinetic mech-anism for DME combustion based on the detailed mechanism byZhao et al. [44].

The previously discussed reduced/skeletal DME mechanismsare mostly based on the detailed mechanisms by Curran et al.[43] and Zhao et al. [44]. Recently, Burke et al. [30] proposed animproved detailed mechanism by updating the DMEsub-mechanism based on the detailed mechanism of DME fromNUIG Aramco Mech 1.3 [50]. The primary purpose of this studyis to develop a reduced kinetic mechanism for DME based on thedetailed mechanism of Burke et al. [30] that has shown superiorresults compared to previous mechanisms. The current mechanismis validated against ignition delay times and engine cases. To

100

101

102

103

104

105

Igni

tion

Del

ay T

ime

(us)

b)

1000/T (K-1)

0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5

φ= 0.5φ= 1.0φ= 1.5

Symbols: Experimental dataSolid Lines: Current reduced modelDash Lines: NUIG Mech_56.54

DME/Air mixtures

, reduced DME mechanism, and experimental results from a shock tube for various

(d) φ = 2.0

p = 11 atm p = 25 atm


(b) φ = 0.5

p = 11 atm p = 25 atm


(c) φ = 1.0

p = 11 atm p = 25 atm


0.6 0.8 1.0 1.2 1.4 1.610-2

10-1

100

101

102

103

p = 11 atm p = 30 atm

Symbols: Experimental dataThick Lines: Current reduced modelThin Lines: NUIG Mech_56.54

Igni

tion

Del

ay ti

me

(ms)

10-2

10-1

100

101

102

Igni

tion

Del

ay ti

me

(ms)

10-2

10-1

100

101

102

Igni

tion

Del

ay ti

me

(ms)

10-2

10-1

100

101

Igni

tion

Del

ay ti

me

(ms)

1000/T (K-1)

0.6 0.8 1.0 1.2 1.4 1.6

1000/T (K-1)

0.6 0.8 1.0 1.2 1.4 1.6

1000/T (K-1)

0.6 0.8 1.0 1.2 1.4 1.6

1000/T (K-1)

(a) φ = 0.3

Fig. 3. Comparison of ignition delay predictions between the detailed mechanism, reduced DME mechanism, and experimental results from an RCM for various pressures andequivalence ratios.

10 0

10 1

10 2

10 3

10 4

10 5

10 6

1 atm10 atm30 atm50 atm100 atm

Igni

tion

Del

ay T

ime

(us)

1000/T (K-1)

Symbols detailed NUIG Mech_56.54Lines: reduced NUIG Mech_56.54

(a)

10 0

10 1

10 2

10 3

10 4

10 5

10 6

Igni

tion

Del

ay T

ime

(us)

(b)

0.6 0.8 1.0 1.2 1.4 1.6 1.8

1000/T (K-1)

0.6 0.8 1.0 1.2 1.4 1.6 1.8

Symbols detailed NUIG Mech_56.54Lines: reduced NUIG Mech_56.54

φ = 0.2φ = 0.5φ = 1.0φ = 2.0φ = 6.0

Fig. 4. Comparison of predicted ignition delay times of pure DME between the detailed and reduced DME mechanisms under different pressures (a) and equivalenceratios (b).

168 L. Pan et al. / Fuel 160 (2015) 165–177

100 101 102 103 1040

1000

2000

3000

4000

100 101 102 103 104

0

1000

2000

3000

4000

DME/Air mixtureT = 700 K, p = 50 atm,φ= 1.0

Detailed NUIG Mech_56.54 Reduced Model

Tem

pera

ture

(K)

Time (us)



Time (us)

100 101 102 103 104 100 101 102 103 104

Time (us) Time (us)



Tem

pera

ture

(K)

0

1000

2000

3000

4000

0

1000

2000

3000

4000

Tem

pera

ture

(K)

Tem

pera

ture

(K)



(a) Temperature Profiles

Fig. 5. Comparison of predicted temperature (a) and pressure (b) profiles of pure DME between the detailed and reduced DME mechanisms at T = 700, 900, 1100, and 1300 K.

L. Pan et al. / Fuel 160 (2015) 165–177 169

ensure the engine predictions are not skewed by the spray modelpredictions, the spray model was validated against experimentaldata from the literature.

2. Reduced chemical kinetics model

It is well acknowledged that a reduced model should ideallymaintain important features of its parent mechanism and thus,the quality of the parent model must have strong influence onthe performance of the reduced model. Therefore, it is necessaryto select a state-of-the-art mechanism as a base mechanism beforeconducting reduction. As mentioned in the introduction, the previ-ously discussed reduced/skeletal DME mechanisms are mostlybased on the detailed mechanisms by Curran et al. [43] and Zhaoet al. [44]. However, as pointed out in our previous work[26,45,51], Zhao model contains the relatively poor H2/O2

sub-mechanism [26,45] and Curran model contains the relativelypoor DME sub-mechanism [51]. Additionally, four generallyaccepted mechanisms for DME oxidation have been checked inour previous work [51] and it is demonstrated that NUIG Mech56.54 developed by the Combustion Chemistry Center of theNational University of Ireland shows excellent performance in pre-dicting ignition delay time under both high and low temperatureranges. More importantly, NUIG Mech 56.54 contains the H2/COsub-mechanism of Kéromnès et al. [52], the C1–C2 basesub-mechanism of Metcalfe et al. [50] and the recently published

propene mechanism of Burke et al. [53,54]. This model has beenvalidated against available literature data of DME from JSR, flowreactor, RCM, shock tube, shock-tube speciation, flame speed, andflame speciation. Therefore, NUIG Mech 56.54 was chosen as basemodel in this study. The detailed mechanism consists of 113 spe-cies and 710 reactions and includes detailed high-temperatureand low-temperature reaction pathways for the oxidation ofDME. Accurate prediction of the chemistry of H2 and CO are essen-tial to describe the laminar flame speed of hydrocarbon fuels [55]and heat release processes of engine combustion [56]. Thus, in thisstudy, the detailed chemical kinetics sub-mechanism of H2 and COare directly used (i.e., without reduction). A simplified mechanismfor C1–C2 covering low- to high-temperature paths was con-structed using reaction path, sensitivity, and rate of productionanalysis. The reduced mechanism of DME is provided in theSupplementary material.

Reaction pathway analysis, shown in Fig. 1, is used to identifythe important reaction pathways controlling DME oxidation. Allreaction pathway analysis were taken at the timing of 20% fuelconsumption where the fuel chemistries still dominate the igni-tion. After each step of the reduction process, sensitivity and rateof production analysis were performed to evaluate the impact ofthe reduction processes. Sensitivity coefficients (S) were deter-mined using the following equation:

Si ¼sð2kiÞ � sð0:5kiÞ

1:5sðkiÞð1Þ

100 101 102 103 1040

50

100

150

200

250

300

100 101 102 103 1040

50

100

150

200

250

300

0

50

100

150

200

250

300

0

50

100

150

200

250

300



Pre

ssur

e (a

tm)

Time (us)



Pre

ssur

e (a

tm)

Time (us)

100 101 102 103 104 100 101 102 103 104

Time (us) Time (us)



Pre

ssur

e (a

tm)



Pre

ssur

e (a

tm)

(b) Temperature Profiles

Fig. 5 (continued)

170 L. Pan et al. / Fuel 160 (2015) 165–177

Based on this methodology, C1–C2 chemistry for the DMEsub-mechanism was reduced to 10 species and 19 reactions, aslisted in Table 1. Specifically, at high temperature conditions, forinstant T = 1400 K, the oxidation of DME is initiated by the molec-ular pyrolysis reaction (R1: CH3OCH3(+M), CH3+CH3O(+M)). Asshown in Fig. 1, DME is primarily consumed through theH-abstraction reactions (87.2%) by H (R2: CH3OCH3 + H,CH3OCH2+H2), OH (R3: CH3OCH3 + OH, CH3OCH2 + H2O) andCH3 (R4: CH3OCH3 + CH3, CH3OCH2 + CH4) to form methoxy-methyl radicals (CH3OCH2). Only a small amount of DME molecules(11.8%) directly undergo the molecular pyrolysis reactions to formCH3 and CH3O. Subsequently, the formed CH3OCH2 radicalsundergo b-scission to form CH3 and CH2O. Finally, the formedCH3, CH2O and CH3O enter the small species oxidation pathways.

At the low temperature range, for instant T = 600 K, the oxida-tion of DME is initiated by the reaction with O2 to produceCH3OCH2 radicals and HO2 (R8: CH3OCH3 + O2, CH3OCH2 + HO2).As shown in Fig. 1, DME is almost entirely consumed through theH-abstraction reactions by H (R2: CH3OCH3 + H, CH3OCH2 + H2),OH (R3: CH3OCH3 + OH, CH3OCH2 + H2O) and CH3 (R4:CH3OCH3 + CH3, CH3OCH2 + CH4) to produce methoxymethylradicals. Then, unlike the high temperature range, the formedmethoxymethyl radicals react with O2 to produce peroxy methox-ymethyl radicals (CH3OCH2O2) via reaction R9: CH3OCH2 + O2,CH3OCH2O2. The formed CH3OCH2O2 radical can then isomerize tothe hydroperoxy-methoxymethyl (CH2OCH2O2H) radicals via

reaction R10: CH3OCH2O2, CH2OCH2O2H. A small amount ofhydroperoxy-methoxymethyl (CH2OCH2O2H) radicals directlyundergo reaction R11: 2CH3OCH2O2) O2 + 2CH3OCH2O. TheCH2OCH2O2H radical either reacts by b-scission R12:CH2OCH2O2H, 2CH2O + OH or undergoes a second oxygenaddition R14: CH2OCH2O2H + O2, O2CH2OCH2O2H, leadingto a low-temperature radical branching pathway involving anumber of isomerization and decomposition steps, R15:O2CH2OCH2O2H, HO2CH2OCHO + OH, R16: HO2CH2OCHO,OCH2OCHO+OH, R17: OCH2OCHO, HOCH2OCO and R18:HOCH2OCO, CH2OH + CO2. The formed CH3, CH2O, CH3O andCH2OH then enter the small species reaction pathways. Thethermodynamic properties of species in the current reactionmechanism are taken from those in the parent mechanism.

3. Mechanism validation

To evaluate the reduced DME mechanism, a wide range of com-parisons between the reduced mechanism and detailed NUIGMECH_56.54 mechanism and experimental data from the litera-ture are performed. As mentioned above, it is the purpose of thepresent reduced mechanism to be applicable to compression igni-tion engines. The following combustion conditions were chosen totest the effectiveness of the reduced mechanism over a wide rangeof conditions found in these engines.

10-3

10-2

10-1

100

101

102

103

104

105

106

100 101 102 103 10410-3

10-2

10-1

100

101

102

103

104

105

106

100 101 102 103 104



DM

E m

ole

fract

ion

/ppm DME/Air mixture

T = 900 K, p = 50 atm,φ= 1.0 Detailed NUIG Mech_56.54 Reduced Model



DM

E m

ole

fract

ion

/ppm

10-3

10-2

10-1

100

101

102

103

104

105

106

10-3

10-2

10-1

100

101

102

103

104

105

106

DM

E m

ole

fract

ion

/ppm

DM

E m

ole

fract

ion

/ppm

Time (us)



Time (us)

100 101 102 103 104 100 101 102 103 104

Time (us) Time (us)

(a) DME

Fig. 6. Comparison of predicted species profiles of pure DME between the detailed and reduced DME mechanisms at T = 700, 900, 1100, and 1300 K.

L. Pan et al. / Fuel 160 (2015) 165–177 171

3.1. Ignition delay times

The auto-ignition of fuel–air is extremely important in highlypremixed combustion technologies including HCCI, PCCI, andRCCI where the overall combustion process is controlled by themixtures auto-ignition distribution. In this section, the calculatedignition delay time using the reduced model is compared to thecalculated ignition delay time using the detailed model and shocktube (ST) and rapid compression machine (RCM) data from theliterature.

Fig. 2 shows the comparison between the reduced and detailedmodel predictions and ST experiments from Li et al. [28] and Pfahlet al. [27] at a range of pressures and equivalence ratios. The calcu-lations are conducted using the constant volume, adiabatic andzero-dimensional reactor in the CHEMKIN II package [57] coupledwith the SENKIN/VTIM approach [58] to consider a measured pres-sure gradient in the reflected shock region of (dp/dt)(1/p0) = +3%ms�1 [27]. It can be seen that both the detailed and presentreduced model are able to accurately capture the measuredignition delay at all experimental conditions. Moreover, both thedetailed and reduced models quantitatively capture the depen-dence on pressure and equivalence, and magnitude of the negativetemperature coefficient (NTC) behavior.

Fig. 3 shows comparisons of the calculated ignition delay timesusing the detailed and reduced models and the experimentalresults from a RCM at / = 0.3–2.0 and p = 11–30 atm from Burke

et al. [29]. The calculations are also conducted using CHEMKIN IIpackage [52] coupled with the SENKIN/VTIM approach [53] toapproximate the facility effects during experiments due to the longresidence times in the RCM. That is, the RCM is simulated as ahomogeneous batch reactor with a varying volume, which isextracted from pressure of non-reactive experiments and then con-verted pressure profiles to volume-time profiles using isentropicassumptions [29]. Again, both the detailed and reduced modelsreproduce the experimental data over the wide temperature rangeof the experimental conditions. The magnitudes of the NTC behav-iors are well captured by the two models. More distinctly, thesmall pressure dependence at the low- and high-temperaturerange and the strong pressure dependence in the NTC regime arecaptured by the models.

3.2. Validation against the base model

Additional validation of the reduced model was performed bycomparing the reduced model to the detailed model over a widerange of conditions. Fig. 4 shows the comparison of the calculatedignition delay times between the detailed and reduced mecha-nisms under various pressures and equivalence ratios at a widetemperature range. The calculated ignition delay times were per-formed with the zero-dimensional, constant volume, adiabaticreactor in the CHEMKIN II package [52]. The calculated ignitiondelay time is defined as the time interval between the beginning

10-12

10-7

10-2

103

10-9

10-4

101

106

100

101

102

103

104

10-6

10-1

104

100

101

102

103

104

10-3

10-1

101

103

105



OH

mol

e fra

ctio

n /p

pm



OH

mol

e fra

ctio

n /p

pm



OH

mol

e fra

ctio

n /p

pm

Time (us)


Detailed NUIG Mech_56.54 Reduced ModelO

H m

ole

fract

ion

/ppm

Time (us)

100

101

102

103

104

100

101

102

103

104

Time (us) Time (us)

(b) OH

Fig. 6 (continued)

172 L. Pan et al. / Fuel 160 (2015) 165–177

of the simulation and the peak of the OH mole fraction. Fig. 4shows that the calculated ignition delay times using the reducedmechanism are quite close to the calculations using the detailedNUIG Mech_56.54 mechanism under various pressures and equiv-alence ratios over a wide temperature range. The good agreementindicates that the main reaction pathways of DME oxidation arewell captured by the reduced mechanism under a wide range ofconditions.

In addition to the ignition delay time, it is important that thereduced mechanism capture the detailed temperature, pressure,and species evolution profiles of the detailed mechanism. Thesefeatures are important to ensure accurate prediction of the heatrelease rate and emission characteristics of internal combustionengines [54]. Moreover, comparisons of species profiles provide arigorous examination of our reduced chemical kinetic mechanismof DME. Fig. 5 shows the comparison of the temperature and pres-sure profiles between the detailed and reduced model atp = 50 atm, / = 1.0, and T = 700, 900, 1100, and 1300 K. The overallagreement between the detailed and reduced model are quite goodat a wide temperature range. The NTC behavior of DME, observedexperimentally between 800 K and 1000 K [27], is reproduced byboth the detailed and reduced model. Fig. 6 shows the comparisonof DME, OH, CH3 and CH2O evolution profiles between the detailedand reduced model at p = 50 atm, / = 1.0 and T = 700, 900, 1100,and 1300 K. Although minor differences exist, the profiles of

DME, OH, CH3 and CH2O predicted by the detailed and reducedmodel are quite close. Considering that a logarithmic scale wasbeen used in both X-axis and Y-axis, the observed differencesbetween the evolution profiles of CH3 and CH2O (at the singleppm level) are acceptable.

3.3. Engine combustion

To evaluate the predictive ability of the reduced mechanism atpractical conditions, experimental data from engine experimentsfrom the literature were used for comparison and validation ofthe reduced mechanism. In this section, simulations were con-ducted using the KIVA-3v release 2 code [59] with the improve-ments to many physical sub-models developed at the EngineResearch Center (ERC). Details of important sub-models are givenin Table 2. A detailed description of the KIVA-3v release 2 codeused in this study can be found in Kokjohn [1].

Prior to comparing the model to engine and emissions data, it isimportant to ensure that the spray model used in this study isrobust and capable of capturing the DME spray characteristics.Measurements of DME spray under different pressures in a con-stant volume combustion system conducted by Konno et al. [19]were used for validation of the DME spray model. Experimentalconditions are given in Table 3.

10-3

10-2

10-1

100

101

102

103

104

105

106

100 101 102 103 104 100 101 102 103 104



CH

2O m

ole

fract

ion

/ppm

10-3

10-2

10-1

100

101

102

103

104

105

106

CH

2O m

ole

fract

ion

/ppm

10-3

10-2

10-1

100

101

102

103

104

105

106

CH

2O m

ole

fract

ion

/ppm

10-3

10-2

10-1

100

101

102

103

104

105

106

CH

2O m

ole

fract

ion

/ppm





Time (us)



Time (us)

100 101 102 103 104 100 101 102 103 104

Time (us) Time (us)

(c) CH2O

Fig. 6 (continued)

L. Pan et al. / Fuel 160 (2015) 165–177 173

A 2-D mesh with a cell size of 1 mm was used to represent aslice of the constant volume vessel. Fig. 7 shows comparisons ofthe measured and calculated DME spray penetration histories forinjection pressures of 20, 60, and 100 MPa at the 2.5 MPa and444 K chamber condition and 0.25 mm nozzle diameter. As theinjection pressure is increased from 20 MPa to 60 MPa, the spraydevelopment accelerates rapidly. With further increased pressure,the effect of pressure is less evident. It is also seen that the calcu-lations predict well the changing dependency of spray penetrationon injection pressure. Specifically, calculations quantitativelymatch the experimental data at 20 and 60 MPa, but slightlyunder-predict the penetration at the 100 MPa condition.Although some differences exist, in general, the spray model is ableto capture spray penetrations of DME at high pressures, demon-strating the applicability of the spray model, spray constant andphysical properties.

Kim et al. [18] performed DME compression ignition engineexperiments in a naturally aspirated, single cylinder,common-rail direct-injection diesel engine with a displacementvolume of 373 cm3. The engine was operated at 4 bar indicatedmean effective pressure (IMEP) and 1500 rev/min. The injectionpressure was constant at 50 MPa and the start of injection (SOI)timing was swept from �20� after top dead center (ATDC) to topdead center (TDC). The detailed engine specifications and operatingconditions are provided in Tables 4 and 5, respectively. Fig. 8shows the computational grid for the engine combustion

simulation. The computational domain is a 45� sector representinga single hole of the six hole injector used in the experiments of Kimet al. [18]. The mesh consists of 18, 590 cells at bottom dead centerwith a resolution of 1 � 1 � 1.5 mm at the piston bowl wall.

Fig. 9 shows comparisons of the measured and simulatedin-cylinder pressure and rate of heat release using both detailedand reduced mechanisms at SOI timings of �8� and �2� ATDC. Itcan be seen that the predicted pressure and heat release ratesusing both mechanisms agree well with the measurements undervarious SOI timings. The largest discrepancies between experi-ments and simulations are observed for the SOI timing = �2�ATDC, where the simulations under-predict cylinder pressure andover-predict the rate of heat release. Note that all model constantswere maintained fixed and no changes were made to the reactionmechanism or spray model to try to ‘‘tune’’ the simulations tomatch the experimental results.

Fig. 10 shows comparisons of the ignition delay times betweenthe experimental and simulated results over the SOI timing sweep.Starting from the most advanced injection timing and retardingresults in a decrease in ignition delay. The minimum ignition delayis found at SOI timings between �10� and �5� ATDC. Furtherretarding the SOI timing results in an increase in the ignition delaydue to the expansion cooling. The ignition delay trends and magni-tudes are well captured by both the detailed and reduced models.The good agreement between simulations and experiments over awide range of conditions demonstrates that the physical models in

Table 2ERC KIVA sub-models.

Model Description

Combustion Direct integration [61]Turbulence RNG k-e [62]NOx Reduced NO mechanism [63]Spray model LDEF with Gasjet Model [64]Drop breakup KH-RT [65]Drop collision O’Rourke [66] w/ROI Model [64]

Table 3Constant volume spray chamber conditions [20].

Description Specification

Injection pressure (Mpa) 20, 60, 100Nozzle diameter (mm) 0.25Nozzle length (mm) 0.7Injection duration (ms) 2Ambient pressure (Mpa) 2.5Ambient temperature (K) 444

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.40

10

20

30

40

50

60

70

80

20 Mpa 60 Mpa) 100 MpaS

pray

tip

pene

tratio

n (m

m)

Time after the start of injection (msec)

Symbols: experimental resultsLines: simulation results

Fig. 7. Comparison of spray tip (vapor) penetration between the experimentalresults and calculations from KIVA.

10-11

10-6

10-1

104

10-6

10-1

104

100 101 102 103 10410-7

10-2

103

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 102 103 10410-9

10-4

101

106



CH

3 m

ole

fract

ion

/ppm

CH

3 m

ole

fract

ion

/ppm

CH

3 m

ole

fract

ion

/ppm

CH

3 m

ole

fract

ion

/ppm





Time (us)

100 101 102 103 104

Time (us)100 101 102 103 104

Time (us)



Time (us)

(d) CH3

Fig. 6 (continued)

174 L. Pan et al. / Fuel 160 (2015) 165–177

Table 4Specifications of the test engine [12].


Stroke (mm) 84.5Bore (mm) 75Displacement volume (cc) 373.3Compression ratio 17.8Valves DOHC 4Injector hole number 6Fuel injection system Bosch common-railNozzle hole diameter (mm) 0.128L/D ratio of nozzle 7.8Spray angle (�C) 156

Table 5Experimental conditions of the engine test from Kim et al. [19].


Engine speed (rpm) 1500Injection pressure (MPa) 50Coolant temperature (�C) 70Oil temperature (�C) 70Mass of fuel (mg/cycle) 11.9Pulse duration (ls) 784Air intake system Naturally aspirated

Fig. 8. Computational grid at TDC for the simulation of engine measurements fromKim et al. [19].

0

1

2

3

4

5

6 Experimentdal data(Kim et al.) Detailed Model Reduced Model

Cyl

inde

r Pre

ssur

e (M

pa)

0

40

80

120

160

200

240

Rat

e of

Hea

t Rel

ease

(J/d

eg.)

Rate of Heat Release

Cylinder pressure

(a) SOE=2º BTDC

-5-20 -15 -10 0 5 10 15 20

Crank angle (deg. ATDC)

Fig. 9. Comparison of experimental cylinder pressure and rate of h

-20 -15 -10 -5 00

2

4

6

8

10

Experimentdal data(Kim et al.) Detailed Model Reduced Model

Igni

tion

Del

ay (d

eg.)

Start of injection pulse (deg. ATDC)

Fig. 10. Comparison of measured and predicted ignition delay times over an SOItiming sweep.

L. Pan et al. / Fuel 160 (2015) 165–177 175

the KIVA-3v release 2 code can well predict the dominant spray,mixing, and combustion processes.

The results above showed that the present mechanisms are cap-able of predicting the bulk combustion characteristics (e.g., cylin-der pressure). The validity of the mechanisms are furtherassessed by comparison to engine out emissions. Fig. 11 showsthe measured and predicted NOx and CO emissions over the SOItiming sweep. NOx emissions are calculated using a reduced NOmechanism [60] consisting of 4 additional species and 12 reactions,as listed in Table 6. Though an extremely reduced NOx mechanismwas used in the present study, the predicted NOx emissions agreewell with the experimental results over the range of SOI timingsconsidered. CO emissions are over predicted by both models atall SOI timings. Although some differences in the CO predictionsexist, the CO levels are relatively low for this conventional dieselcombustion mode. The measured turn up in CO emissions at thelatest SOI timing is likely due to an increase in cyclic instability,which is not modeled in the present study. In general, the presentreduced DME model is able to provide similar performance to thedetailed DME model with 94 fewer species and 644 fewer reac-tions, making the model useful for future optimization studies ofDME fueled engines.

-5-20 -15 -10 0 5 10 15 20

0

2

4

6

8 Experimentdal data(Kim et al.) Detailed Model Reduced Model

Cyl

inde

r Pre

ssur

e (M

pa)

Crank angle (deg. ATDC)

pressure

0

50

100

150

200

Rat

e of

Hea

t Rel

ease

(J/d

eg.)

Rate of Heat Release

(b) SOE=8º BTDC

eat release with the detailed and reduced DME mechanisms.

-5 00

250

500

750

1000

1250

1500


NO

x (p

pm v

ol.)

Start of injection pulse (deg. ATDC)-20 -15 -10 -5 0

Start of injection pulse (deg. ATDC)-20 -15 -10

0.00

0.25

0.50

0.75

1.00


CO

(% v

ol.)

Fig. 11. Comparison of measured and predicted NOx and CO emissions over an SOI timing sweep.

Table 6Reduced NOx mechanism used in this study [63].

Reactions

N + NO, N2 + ON + O2, NO + ON + OH, NO + HN2O + O, N2 + O2

N2O + O, 2NON2O + H, N2 + OHN2O + OH, N2 + HO2

N2O(+M), N2 + O(+M)HO2 + NO, NO2 + OHNO + O + M, NO2 + MNO2 + O, NO + O2

NO2 + H, NO + OH

176 L. Pan et al. / Fuel 160 (2015) 165–177

4. Conclusions

A new, reduced chemical kinetic model of DME consisting of 29species and 66 reactions was proposed in this paper to simulate thecombustion and emission formation processes of DME. The pro-posed model has been validated against the base mechanism andexperimental data over a wide range of operating conditions. Themajor conclusions of the study can be summarized as follows:

1. The proposed, reduced DME mechanism was shown to repro-duce the measured ignition delay times over a range of condi-tions applicable to compression ignition engine operation. Thereduced mechanism yields similar performance to that of thedetailed mechanism with 94 fewer species and 644 fewerreactions.

2. The species profiles of the reduced mechanism were shown tobe in good agreement with the detailed model over a range oftemperatures, pressures, and equivalence ratios.

3. The spray model in the ERC KIVA-3v release 2 codes coupledwith liquid physical properties of DME was used to validatedthe experimental data of vapor penetration from the literatureand good agreement between the experimental data and simu-lation results demonstrates applicability of the spray model,spray constants, and physical properties.

4. The comparison between the engine experiments and simula-tion results show that the reduced and detailed mechanismscan capture the cylinder pressure, rate of heat release, ignitiondelay time, and NOx emissions accurately. Further improve-ment in the prediction of CO is recommended.

Acknowledgements

The support provided by China Scholarship Council (CSC,201406280076) during a visit of Lun Pan to University ofWisconsin, Madison is acknowledged. Authors also appreciate thesupport from the Engine Research Center (ERC) at University ofWisconsin, Madison and Combustion and Spray Laboratory (CSL)from Xi’an Jiaotong University.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.fuel.2015.07.066.

References

[1] Kokjohn SL. Reactivity Controlled Compression Ignition (RCCI) Combustion,PhD dissertation. University of Wisconsin-Madison, Madison, WI; 2012.

[2] Persson H, Andersson Ö, Egnell R. Fuel effects on flame lift-off under dieselconditions. Combust Flame 2011;158:91–7.

[3] Wang H, Yao M, Yue Z, Jia M, Reitz RD. A reduced toluene reference fuelchemical kinetic mechanism for combustion and polycyclic-aromatichydrocarbon predictions. Combust Flame 2015;162:2390–404.

[4] Semelsberger TA, Borup RL, Greene HL. Dimethyl ether (DME) as an alternativefuel. J Power Sources 2006;156:497–511.

[5] Park SH, Lee CS. Combustion performance and emission reductioncharacteristics of automotive DME engine system. Prog Energy Combust Sci2013;39:147–68.

[6] Suh HK, Lee CS. Experimental and analytical study on the spray characteristicsof dimethyl ether (DME) and diesel fuels within a common-rail injectionsystem in a diesel engine. Fuel 2008;87:925–32.

[7] Yu J, Bae C. Dimethyl ether (DME) spray characteristics in a common-rail fuelinjection system. Proc Inst Mech Eng Part D J Automob Eng 2003;217:1135–44.

[8] Yu J, Lee J, Bae C. Dimethyl ether (DME) spray characteristics compared todiesel in a common-rail fuel injection system. SAE Technical Paper 2002-01-2898, 2002.

[9] Sidu X, Mingfa Y, Junfeng X. An experimental investigation on the spraycharacteristics of dimethyl ether (DME). SAE Technical Paper 2001-01-0142,2001.

[10] Park SH, Kim HJ, Lee CS. Macroscopic spray characteristics and breakupperformance of dimethyl ether (DME) fuel at high fuel temperatures andambient conditions. Fuel 2010;89:3001–11.

[11] Jun L, Sato Y, Noda A. An experimental study on DME spray characteristics andevaporation processes in a high pressure chamber. SAE Technical Paper 2001-01-3635, 2001.

[12] Kim HJ, Suh HK, Lee CS. Numerical and experimental study on the comparisonbetween diesel and dimethyl ether (DME) spray behaviors according tocombustion chamber shape. Energy Fuels 2008;22:2851–60.

[13] Longbao Z, Hewu W, Deming J, Zuohua H. Study of performance andcombustion characteristics of a DME-fueled light-duty direct-injection dieselengine. SAE Technical Paper 1999-01-3669, 1999.

[14] Huang Z, Wang H, Chen H, Zhou L, Jiang D. Study of combustion characteristicsof a compression ignition engine fuelled with dimethyl ether. Proc Inst MechEng Part D J Automob Eng 1999;213:647–52.


http://refhub.elsevier.com/S0016-2361(15)00757-7/h0010
























L. Pan et al. / Fuel 160 (2015) 165–177 177

[15] Tsuchiya T, Sato Y. Development of DME engine for heavy-duty truck. SAETechnical Paper 2006-01-0052, 2006.

[16] Ying W, Li H, Jie Z, Longbao Z. Study of HCCI-DI combustion and emissions in aDME engine. Fuel 2009;88:2255–61.

[17] Kim HJ, Lee KS, Lee CS. A study on the reduction of exhaust emissions throughHCCI combustion by using a narrow spray angle and advanced injection timingin a DME engine. Fuel Process Technol 2011;92:1756–63.

[18] Park SH, Kim HJ, Lee CS. Effects of dimethyl-ether (DME) spray behavior in thecylinder on the combustion and exhaust emissions characteristics of a highspeed diesel engine. Fuel Process Technol 2010;91:504–13.

[19] Kim MY, Yoon SH, Ryu BW, Lee CS. Combustion and emission characteristics ofDME as an alternative fuel for compression ignition engines with a highpressure injection system. Fuel 2008;87:2779–86.

[20] Konno M, Chiba K, Okamoto T. Experimental and numerical analysis of highpressure DME spray. SAE Technical Paper 2010-01-0880, 2010.

[21] Youn IM, Park SH, Roh HG, Lee CS. Investigation on the fuel spray and emissionreduction characteristics for dimethyl ether (DME) fueled multi-cylinderdiesel engine with common-rail injection system. Fuel Process Technol2011;92:1280–7.

[22] Cook RD, Davidson DF, Hanson RK. Shock tube measurements of ignition delaytimes and OH time-histories in dimethyl ether oxidation. Proc Combust Inst2009;32:189–96.

[23] Dagaut P, Daly C, Simmie JM, Cathonnet M. The oxidation and ignition ofdimethylether from low to high temperature (500–1600 K): experiments andkinetic modeling. Sym (Int) Combust 1998;27:361–9.

[24] Tang C, Wei L, Zhang J, Man X, Huang Z. Shock tube measurements and kineticinvestigation on the ignition delay times of methane/dimethyl ether mixtures.Energy Fuels 2012;26:6720–8.

[25] Pan L, Hu E, Deng F, Zhang Z, Huang Z. Effect of pressure and equivalence ratioon the ignition characteristics of dimethyl ether-hydrogen mixtures. Int JHydrogen Energy 2014;39:19212–23.

[26] Pan L, Hu E, Zhang J, Zhang Z, Huang Z. Experimental and kinetic study onignition delay times of DME/H2/O2/Ar mixtures. Combust Flame2014;161:735–47.

[27] Pfahl U, Fieweger K, Adomeit G. Self-ignition of diesel-relevant hydrocarbon-air mixtures under engine conditions. Symp (Int) Combust 1996;26:781–9.

[28] Li Z, Wang W, Huang Z, Oehlschlaeger MA. Dimethyl ether autoignition atengine-relevant conditions. Energy Fuels 2013;27:2811–7.

[29] Mittal G, Chaos M, Sung C-J, Dryer FL. Dimethyl ether autoignition in a rapidcompression machine: experiments and chemical kinetic modeling. FuelProcess Technol 2008;89:1244–54.

[30] Burke U, Somers KP, O’Toole P, Zinner CM, Marquet N, Bourque G, et al. Anignition delay and kinetic modeling study of methane, dimethyl ether, andtheir mixtures at high pressures. Combust Flame 2015;162:315–30.

[31] Fuest F, Magnotti G, Barlow R, Sutton J. Scalar structure of turbulent partially-premixed dimethyl ether/air jet flames. Proc Combust Inst 2015;35:1235–42.

[32] Qin X, Ju Y. Measurements of burning velocities of dimethyl ether and airpremixed flames at elevated pressures. Proc Combust Inst 2005;30:233–40.

[33] Wang Y, Holley A, Ji C, Egolfopoulos F, Tsotsis T, Curran H. Propagation andextinction of premixed dimethyl-ether/air flames. Proc Combust Inst2009;32:1035–42.

[34] De Vries J, Lowry WB, Serinyel Z, Curran HJ, Petersen EL. Laminar flame speedmeasurements of dimethyl ether in air at pressures up to 10 atm. Fuel2011;90:331–8.

[35] Huang Z, Wang Q, Yu J, Zhang Y, Zeng K, Miao H, et al. Measurement of laminarburning velocity of dimethyl ether–air premixed mixtures. Fuel2007;86:2360–6.

[36] Kaiser E, Wallington T, Hurley M, Platz J, Curran H, Pitz W, et al. Experimentaland modeling study of premixed atmospheric-pressure dimethyl ether–airflames. J Phy Chem A 2000;104:8194–206.

[37] Pyun SH, Ren W, Lam K-Y, Davidson DF, Hanson RK. Shock tube measurementsof methane, ethylene and carbon monoxide time-histories in DME pyrolysis.Combust Flame 2013;160:747–54.

[38] Hidaka Y, Sato K, Yamane M. High-temperature pyrolysis of dimethyl ether inshock waves. Combust Flame 2000;123:1–22.

[39] Dagaut P, Boettner J-C, Cathonnet M. Chemical kinetic study of dimethyletheroxidation in a jet stirred reactor from 1 to 10 atm: experiments and kineticmodeling. Symp (Int) Combust 1996;26:627–32.

[40] Curran H, Pitz W, Westbrook C, Dagaut P, Boettner JC, Cathonnet M. A widerange modeling study of dimethyl ether oxidation. Int J Chem Kinet1998;30:229–41.

[41] Dagaut P, Daly C, Simmie JM, Cathonnet M, et al. The oxidation and ignition ofdimethylether from low to high temperature (500–1600 K): Experiments andkinetic modeling. Symp (Int) Combust 1998;27:361–9.

[42] Fischer S, Dryer F, Curran H. The reaction kinetics of dimethyl ether. I: High-temperature pyrolysis and oxidation in flow reactors. Int J Chem Kinet2000:713–40.

[43] Curran H, Fischer S, Dryer F. The reaction kinetics of dimethyl ether. II: Low-temperature oxidation in flow reactors. Int J Chem Kinet 2000;32:741–59.

[44] Zhao Z, Chaos M, Kazakov A, Dryer FL. Thermal decomposition reaction and acomprehensive kinetic model of dimethyl ether. Int J Chem Kinet2008;40:1–18.

[45] Pan L, Hu E, Meng X, Zhang Z, Huang Z. Kinetic modeling study of hydrogenaddition effects on ignition characteristics of dimethyl ether at engine-relevant conditions. Int J Hydrogen Energy 2015;40:5221–35.

[46] Yamada H, Sakanashi H, Choi N, Tezaki A. Simplified oxidation mechanism ofDME applicable for compression ignition. SAE Technical Paper 2003-01-1819,2003.

[47] Kim H, Cho S, Min K. Reduced chemical kinetic model of DME for HCCIcombustion. SAE Technical Paper 2003-01-1822, 2003.

[48] Liang X. Numerical study on HCCI combustion of DME/MEOH dual fuels,Master’s Dissertation. Tianjin University, Tianjin, China; 2005.

[49] Chin GT, Chen J-Y, Rapp VH, Dibble R. Development and validation of a reducedDME mechanism applicable to various combustion modes in internalcombustion engines. J Combust 2011;2011.

[50] Metcalfe WK, Burke SM, Ahmed SS, Curran HJ. A hierarchical and comparativekinetic modeling study of C1–C2 hydrocarbon and oxygenated fuels. Int JChem Kinet 2013;45:638–75.

[51] Pan L, Hu E, Tian Z, Yang F, Huang Z. Experimental and kinetic study on ignitiondelay times of dimethyl ether at high temperatures. Energy Fuels2015;29:3495–506.

[52] Kéromnès A, Metcalfe WK, Heufer KA, Donohoe N, Das AK, Sung C-J, et al. Anexperimental and detailed chemical kinetic modeling study of hydrogen andsyngas mixture oxidation at elevated pressures. Combust Flame2013;160:995–1011.

[53] Burke SM, Burke U, Mc Donagh R, Mathieu O, Osorio I, Keesee C, et al. CombustFlame 2015;162:296–314.

[54] Burke SM, Metcalfe W, Herbinet O, Battin-Leclerc F, Haas FM, Santner J, et al.An experimental and modeling study of propene oxidation. Part 1: speciationmeasurements in jet-stirred and flow reactors. Combust Flame2014;161:2765–84.

[55] Ranzi E, Frassoldati A, Grana R, Cuoci A, Faravelli T, Kelley A, et al. Hierarchicaland comparative kinetic modeling of laminar flame speeds of hydrocarbon andoxygenated fuels. Prog Energy Combust Sci 2012;38:468–501.

[56] Irvin G. Combustion. New York: Academic Press; 1987.[57] Kee RJ, Rupley FM, Miller JA. Chemkin-II: a Fortran chemical kinetics package

for the analysis of gas-phase chemical kinetics. Livermore, CA (USA): SandiaNational Labs; 1989.

[58] Chaos M, Dryer FL. Chemical-kinetic modeling of ignition delay:considerations in interpreting shock tube data. Int J Chem Kinet2010;42:143–50.

[59] Chang Y, Jia M, Liu Y, Li Y, Xie M. Development of a new skeletal mechanismfor n-decane oxidation under engine-relevant conditions based on adecoupling methodology. Combust Flame 2013;160:1315–32.

[60] Amsden AA. KIVA-3V, release 2, improvements to KIVA-3V, Los AlamosNational Laboratory, Los Alamos, NM, Report No. LA-UR-99-915; 1999.

[61] Sun Y. Diesel combustion optimization and emissions reduction usingadaptive injection strategies (AIS) with improved numerical models, PhDdissertation. University of Wisconsin-Madison, Madison, WI; 2007.

[62] Ra Y, Reitz RD. A combustion model for IC engine combustion simulations withmulti-component fuels. Combust Flame 2011;158:69–90.

[63] Han Z, Reitz RD. Turbulence modeling of INTERNAL combustion engines usingRNG k-e models. Combust Sci Technol 1995;106:267–95.

[64] Abani N, Munnannur A, Reitz RD. Reduction of numerical parameterdependencies in diesel spray models. J Eng Gas Turb Power 2008;130:1–9.

[65] Beale JC, Reitz RD. Modeling spray atomization with the Kelvin-Helmholtz/Rayleigh-Taylor hybrid model. Atomization Spray 1999;9:623–50.

[66] O’Rourke PJ, Amsden AA. A Spray/wall interaction submodel for the KIVA 3wall film model. SAE Technical Paper 2000-01-0271, 2000.






























































































































Documents

Development and validation of a reduced chemical kinetic model …and... · 2015-08-09 · fundamental combustion experimental data. However, because the computational time of combustion