The role of the methyl ester moiety in biodiesel combustion: A kinetic modeling comparison of methyl butanoate and n-butane

Fuel 92 (2012) 16–26

Contents lists available at ScienceDirect

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

The role of the methyl ester moiety in biodiesel combustion: A kinetic modeling comparison of methyl butanoate and n-butane Kuang C. Lin, Jason Y.W. Lai, Angela Violi ⇑ Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI 48109, USA

a r t i c l e

i n f o

Article history: Received 19 August 2010 Accepted 17 May 2011 Available online 28 June 2011 Keywords: Methyl ester Biodiesel combustion Kinetic modeling Combustion characteristics Methyl butanoate

a b s t r a c t Growth of the biodiesel industry has motivated increased study of the combustion characteristics of its constituent molecules and building combustion modeling capability. Understanding how these characteristics differ between bio-derived and conventional diesel fuels can help in evaluating biodiesel performance. A kinetic modeling comparison of methyl butanoate and n-butane, its corresponding alkane, contrasted the combustion of methyl esters and normal alkanes, towards understanding the effect of the methyl ester moiety. Utilizing a combined n-heptane and methyl butanoate kinetic mechanism in shock tube simulations, the results predicted no region of negative temperature coefficient (NTC) behavior for methyl butanoate, compared to a well defined NTC region for n-butane. We observed that oxidation pathways associated with the methyl ester moiety inhibited NTC behavior, through increased production of hydroperoxy radicals (HO2) instead of hydroxyl radicals (OH). In addition, we compared the evolution of carbon monoxide, carbon dioxide, ethylene and acetylene. The early formation of CO and CO2, directly from methyl butanoate, revealed unique reaction pathways that also influenced a reduction in soot precursor formation. Overall, these results will help to understand how combustion processes change with the inclusion of oxygenated fuels, which will inform the study and design of combustion technologies. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Fuel security and anthropogenic carbon dioxide emissions, which are associated with climate change [1], have arisen primarily from fossil fuel use for energy generation. The relentless growth in demand for energy has exacerbated these issues and has thus driven a search for sustainable alternatives to fossil fuels. One attractive option is biodiesel, which is an oxygenated, diesel-like fuel consisting of fatty acid alkyl esters (most commonly fatty acid methyl esters, or FAME’s) that are derived from oils, fats [2] or algae [3]. Biodiesel has the advantage of being generally compatible with current combustion technologies and fuel infrastructure. However, the oxygenated chemical structure that is characteristic of biodiesel fuels alters their combustion and leads to differences in macro scale performance factors, such as reactivity and pollutant formation, when compared to conventional diesel fuels. Consequently, the development of novel bio-derived fuels and combustion technologies will benefit from an understanding of the relationship between chemical structure and combustion characteristics, as well as from the development of predictive capability

⇑ Corresponding author. Address: 2250 G.G. Brown Laboratory, 2350 Hayward Street, Ann Arbor, MI 48109-2125, USA. Tel.: +1 734 615 6448; fax: +1 734 647 E-mail address: [email protected] (A. Violi). 0016-2361/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.fuel.2011.05.014

for combustion through kinetic mechanisms. This study addresses both of these goals by contrasting the oxidation of methyl butanoate (MB) with that of n-butane, utilizing chemical kinetic modeling. As methyl butanoate and n-butane differ only by the methyl ester moiety in MB, the results of this study will isolate the effect of this moiety on combustion parameters, namely reactivity, as well as the formation of carbon dioxide, carbon monoxide, acetylene (C2H2), an important soot precursor [4], and the related species ethylene (C2H4). 1.1. Methyl butanoate Methyl butanoate (C5H10O2, MB), shown in Fig. 1, is a relatively simple methyl ester that possesses the primary features of biodiesel, namely the methyl ester moiety, as well as the alkyl chain. Due to size, MB was chosen as an initial surrogate for biodiesel, as the alkyl chain of MB was ostensibly long enough to reproduce the behavior of real biodiesel fuel. Other researchers have recently described kinetic mechanisms for methyl butanoate, as well as other methyl esters and biofuels, in detail [5,6]. Briefly, Fisher et al. at the Lawrence Livermore National Laboratory (LLNL) created the first methyl butanoate mechanism, containing 279 species and 1259 reactions [7], which formed a basis for future mechanism development. Metcalfe et al.

K.C. Lin et al. / Fuel 92 (2012) 16–26

O O Fig. 1. The structure of methyl butanoate.

[8] and Dooley et al. [9] each developed this mechanism further and verified it with a range of experiments, including high temperature shock tube and rapid compression machine studies. Gaïl et al. [10] also modified and validated the LLNL mechanism against opposed diffusion flame and variable pressure flow reactors. Ab Initio techniques have also been employed to improve the accuracy of important rates in the kinetic mechanism. Huynh and Violi [11] created a submechanism for the thermal decomposition of methyl butanoate. Huynh et al. [12] subsequently computed rates for reactions involving the methoxy formyl radical (CH3OCO), which is important in the formation of CO2 during oxidation. These improvements to the mechanism were significant when verified against experimental shock tube data for MB pyrolysis obtained by Farooq et al. [13]. One of the main results of prior methyl butanoate studies was that MB did not exhibit negative temperature coefficient (NTC) behavior, which is observed as a decrease in reactivity over an intermediate temperature range, between approximately 700 K and 1000 K. This type of behavior has been observed for larger methyl esters such as methyl palmitate [14]; consequently, methyl butanoate is unsuitable to predict biodiesel reactivity, which substantially reduces its viability as a biodiesel surrogate. Larger methyl esters, such as methyl decanoate, have subsequently been shown to be more suitable as biodiesel surrogates due to their ability to reproduce NTC behavior [15]. However, methyl butanoate does have an important role in both mechanism development and understanding biodiesel combustion. Since larger kinetic mechanisms are typically built upon collections of submechanisms for smaller molecules, an accurate model for MB will become part of a mechanism for a larger, more appropriate methyl ester. As a result, the development of a wellverified model for methyl butanoate is still an important task. Additionally, due to its relatively small size, methyl butanoate allows us to efficiently study methyl ester combustion in contrast with normal alkanes. We can therefore seek to understand how the presence of the methyl ester moiety affects the combustion of MB and utilize this knowledge to inform the study of other methyl esters. 1.2. Combustion features of methyl butanoate Several important combustion features of methyl esters have been studied, specifically in the context of methyl butanoate combustion [5]. One of the most important combustion features that has been studied is autoignition behavior, especially the absence of a negative temperature coefficient (NTC) region in methyl butanoate combustion. The NTC region exhibits a decrease in reactivity over a given intermediate temperature range, typically around 700–1000 K. Alkylperoxy radical isomerizations are generally associated with this region, as subsequent reaction pathways lead to either chain propagation or branching [16]. As temperature increases within the NTC region, the important reaction pathways change from chain branching to chain propagation, and reactivity decreases. Fisher et al. [7] predicted a region of NTC behavior for methyl butanoate using their model for homogeneous, isothermal, constant volume combustion. Subsequent studies, however, did not predict similar behavior in other environments. Gaïl et al. [10]

17

did not observe an NTC region in either simulations or experiments for variable pressure flow reactor (VPFR) combustion. Similarly, Dooley et al. [9] did not find NTC for either of flow reactor or rapid compression machine (RCM) experiments or modeling. In contrast, n-butane does exhibit NTC behavior [17], which implies that the presence of the methyl ester has a significant effect on reactivity. Since the NTC region is ascribed to alkylperoxy radical isomerizations, the lack of NTC behavior in MB oxidation suggests that the methyl ester alters these reaction pathways; however, as larger molecules do exhibit an NTC region, this effect diminishes with the increase of alkyl chain length. The unique pathways that influence this effect are of substantial interest, insofar as we wish to understand the role of the methyl ester moiety in combustion. A characteristic feature of methyl butanoate oxidation, in contrast with non-oxygenated fuels, is CO2 formation which occurs relatively early in combustion as a consequence of fuel-bound oxygen. Several researchers [12,18,13] have observed that the primary channels for this CO2 production involve the butanoic acid (BAOJ) and methoxy formyl (CH3OCO) radicals. These pathways are a direct result of the methyl ester moiety in methyl butanoate. Beyond simply noting the existence of this phenomenon, the formation of oxygenated species, such as CO2 and CO, have implications on the production of soot precursors such as acetylene. Studies involving both simulation and experiment have noted a decrease in soot formation when utilizing oxygenated fuels [19–21]. For example, an engine modeling study performed by Westbrook et al. [18] found that displacing a non-oxygenated diesel surrogate, n-heptane, with oxygenated molecules lowered the overall formation of soot precursors. The mechanism for this reduction is thought to be due to the strength of the carbon–oxygen bond that is higher as compared with carbon–carbon or carbon–hydrogen bonds, noted by Osmont et al. [22]. Ostensibly, the carbon–oxygen bond persists throughout the combustion process, which removes at least one carbon from a pool of carbon atoms that may form soot precursors. In the case of MB, Westbrook et al. [18] suggest that the fuel-bound oxygen atoms remain bonded to one carbon, forming mainly carbon dioxide. Arguably, the formation of CO2 effectively wastes an oxygen atom that could potentially prevent a different carbon atom from forming a soot precursor; however, the exact mechanism by which the methyl ester moiety reduces soot precursor formation has not been yet been fully analyzed. A more detailed analysis of this phenomenon will provide more insight into this mechanism. 1.3. Objectives The objective of this work is to understand how the methyl ester structure alters the evolution of important combustion phenomena, namely ignition characteristics and NTC behavior, fuel-originated oxygenated species formation and soot precursor production. To achieve this goal, we contrasted the unique pathways that are associated with methyl butanoate combustion with its corresponding normal alkane, n-butane. 1.4. Methodology In the present study, we used a kinetic modeling approach to investigate the differences between the oxidation of methyl butanoate and n-butane. Despite an inequality in number of carbons of these molecules, n-butane was chosen for comparison as it is the corresponding alkane to methyl butanoate. In this way, we isolated the effect of the methyl ester moiety, including that of the extra carbon atom of the methyl group. Simulations were performed using the CHEMKIN 4.1 software package [23], along with a combined mechanism derived from MB mechanism of the University of Michigan [12], and the Lawrence Livermore National

18

K.C. Lin et al. / Fuel 92 (2012) 16–26

Laboratory’s (LLNL) detailed n-heptane mechanism [24]. Both of these mechanisms are relatively mature and have been validated under a variety of experimental conditions. Both LLNL mechanisms, n-heptane and MB, upon which our current mechanism is based, have been utilized previously in a modeling study of soot reduction by Westbrook et al. [18]. While this mechanism is by no means completely accurate, we believe it has sufficient detail to provide qualitative insight in the reaction pathways involved in the phenomena we have chosen to study. Table 1 enumerates some relevant molecule abbreviations used in this mechanism and throughout this study.

2. Results 2.1. Ignition characteristics We studied shock tube oxidation of stoichiometric mixtures of each fuel and air for a range of temperatures from 750 K to 1300 K and pressures of 12.5 atm and 40 atm (values after reflected shock). These pressures were chosen in accordance with a study by Hoffman et al. [25] comparing several fuel surrogates in HCCI engine conditions. Additionally, we investigated the effect of changing the absolute amount of fuel and oxygen in the system while maintaining stoichiometric conditions. The simulations were performed for oxygen concentrations (mole fraction) of 6.5% and 19.5%, or 1% and 3% fuel respectively. We defined ignition delay as the time between the beginning of the simulation, where the shock wave reflects from the apparatus wall, to the maximum rate of increase in OH concentration, d½OH . Fig. 2 shows the computed dt ignition delay times for both fuels under these conditions. The effect of the absolute amount of fuel and oxygen was insignificant, so further discussion will be limited to the 19.5% O2 case. The primary difference in combustion, as expected, was the strong NTC behavior of n-butane that was not matched by methyl butanoate, which occurs between 740 K and 890 K at 12.5 atm, or 750 K and 920 K at 40 atm. At temperatures above these regions, the two fuels show similar autoignition characteristics. We subsequently conducted sensitivity analysis to identify important reactions governing autoignition. Our analysis technique involved increasing the A-factors of the forward and reverse rate constants of a reaction or class of reactions by a factor of 2 and calculating the resulting sensitivity coefficient. The sensitivity coefficient, r, was s0 s defined as r ¼ ids id , where sid is the ignition delay of the unperid turbed mechanism and s0id is that of the adjusted mechanism. A positive sensitivity coefficient signifies an increase in ignition delay, while a negative sensitivity coefficient indicates a decrease. Through this method, we qualitatively obtained the sensitivity of ignition delay to a number of reactions or reaction classes.

Table 1 Molecule abbreviations for methyl butanoate species used throughout this study. Molecule

Abbreviation

BAOJ C5H7O2 MB MB2J MB3J MB4J MBMJ MB2D MB3D ME2J MP3J MP2D MP2D3J

CH3CH2CH2C(@O)O CH2@CHCHC(@O)OCH3 CH3CH2CH2C(@O)OCH3 CH3CH2CHC(@O)OCH3 CH3CHCH2C(@O)OCH3 CH2CH2CH2C(@O)OCH3 CH3CH2CH2C(@O)OCH2 CH3CH@CHC(@O)OCH3 CH2@CHCH2C(@O)OCH3 CH2C(@O)OCH3 CHCH2C(@O)OCH3 CH@CH2C(@O)OCH3 CH@CHC(@O)OCH3

Fig. 2. Calculated ignition delay times for stoichiometric mixtures of MB/O2/Ar and C4H10/O2/Ar as a function of temperature. (a) P = 12.5 atm; (b) P = 40 atm. The following mixtures were used: 1% fuel, 6.5% O2 and 92.5% N2; 3% fuel, 19.5% O2 and 77.5% N2.

Since the results were similar between both studied pressures, we will only discuss the results at 12.5 atm. The results of this analysis are shown in Figs. 3 and 4, for the temperatures 1400 K and 780 K respectively and for a pressure of 12.5 atm. At 1400 K, each fuel’s ignition properties are sensitive to analogous types of reactions, reflected in the sensitivity results. Fig. 4 shows the differences in sensitivity results between both fuels at 780 K. A major feature of the results was the dependence of autoignition on reactions that involve HO2 or OH. The hydroperoxyl radical HO2 negatively affects the concentration of hydroxyl radical OH, which governs the reactivity of the system. Specifically, the recombination reaction HO2 + HO2 H2O2 + O2 is a chain termination reaction; moreover, H2O2 further consumes OH through the reaction H2O2 + OH H2O + HO2, which forms a radical consuming cycle. Consequently, reactions that produce HO2 significantly reduce reactivity. Extending this concept further, any reactions involved in pathways that form either HO2 or OH will influence the autoignition properties of their respective fuels. For example, at 1400 K (Fig. 3), hydrogen abstraction from MB increased ignition delay, as these reactions consumed radicals such as H, OH, or CH3. In particular, hydrogen abstraction from MB by H-atoms decreased

K.C. Lin et al. / Fuel 92 (2012) 16–26

(a)

19

(a)

CH 3 + CH 3 = C2H 6 CH 3+ HO2 = CH 4 + O2 HO2+ OH = H 2O + O2 CH 2O + H = HCO + H 2 CH 2O + OH = HCO + H 2O C2H 3 + O2 = CH 2CHO + O C2H 4 + X = C 2H 3 + XH CH 3 + HO2 = CH 3O + OH H + O2 = O + OH MB + H MB + X = MBMJ + XH MB + X = MB3J + XH MB + X = MB4J + XH MB + X = MB2J + XH MB + OH MB + CH 3 MB Decomposition

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

Sensitivity coefficient

(b)

(b)

CH 3 + CH 3 = C2H 6 CH 3+ HO2 = CH 4 + O2 HO2+ OH = H 2O + O2 C 2H 3 + O2 = CH 2CHO + O CH 3 + HO2 = CH 3O + OH C 2H 4 + X = C2H 3 + XH H + O2 = O + OH

C4H 10 + X = sC 4H 9 + XH C 4H 10 + H C4H 10 + OH C4H 10 + CH 3 C4H 10 + X = pC4H 9 + XH C 4H 10 Decomposition

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

Sensitivity coefficient

(c)

Fig. 3. Ignition delay sensitivity results at T = 1400 K, P = 12.5 atm, for mixtures of 3% fuel, 19.5% O2 and 77.5% N2. (a) Methyl butanoate; (b) n-butane. The symbol ‘X’ denotes the combined sensitivity to reactions by the radicals H, OH and CH3.

reactivity through competition with the branching reaction H + O2 O + OH. Conversely, as the importance of this branching reaction decreases with temperature, hydrogen abstractions increase reactivity at lower temperatures, shown for 780 K in Fig. 4. This result illustrates the indirect effect that reactions can have on reactivity, without involving OH or HO2 directly. The reaction pathways involved in the decomposition of MB and n-butane to either HO2 or OH will help to elucidate the effect of the methyl ester on the NTC region. Therefore, for shock tube oxidation of both fuels at 780 K, we performed rate of production (ROP) or flux analysis at times throughout the simulation, namely 2%, 33% and 50% fuel consumption, or 0.006 s, 0.084 s and 0.096 s respectively. The most important reactions in the consumption of each fuel are shown in Table 2. Table 3 lists the proportions of the corresponding fuel that form each alkylester or alkyl radical for combustion of methyl butanoate and n-butane. The evolution of these radicals ultimately influences the autoignition behavior of their corresponding fuels; in the low temperature region, these radicals react with O2 to form RO2 radicals, which isomerize and subsequently lead to the formation of either HO2 or OH and thus significantly affect ignition delay.

Fig. 4. Ignition delay sensitivity results at T = 780 K, P = 12.5 atm, for mixtures of 3% fuel, 19.5% O2 and 77.5% N2. (a) methyl butanoate; (b) n-butane; (c) sensitivity of ignition delay to hydrogen abstraction reactions. The symbols ‘X’ and ‘x’ denote the combined sensitivity to reactions at possible abstraction sites for MB and n-butane, respectively.

Figs. 5 and 6 illustrate the primary decomposition pathways for the fuel alkylester or alkyl radicals of both fuels, derived from flux analysis at 33% fuel consumption. The radical MB2J (Fig. 5a) reacted to form MB2OO, which subsequently isomerized through 5-, 6- and 7-membered transition states to form the hydroperoxy alkylester radicals MB2OOH3J, MB2OOH4J and MB2OOHMJ respectively. The most important channels in this case were the

20

K.C. Lin et al. / Fuel 92 (2012) 16–26

Table 2 The most important reactions involved in direct methyl butanoate and n-butane consumption in our autoignition study. MB

C4H10

R25. MB + HO2 MB3J + H2O2 R35. MB + CH3O2 MB2J + CH3O2H R36. MB + HO2 MB2J + H2O2 R371. MB + H MB4J + H2 R373. MB + OH MB4J + H2O R379. MB + H MB3J + H2 R381. MB + OH MB3J + H2O R385. MB + H MB2J + H2 R387. MB + OH MB2J + H2O R403. MB + H MBMJ + H2 R405. MB + OH MBMJ + H2O

R738. R739. R740. R741. R745.

C4H10 + H pC4H9 + H2 C4H10 + H sC4H9 + H2 C4H10 + OH pC4H9 + H2O C4H10 + OH sC4H9 + H2O C4H10 + HO2 sC4H9 + H2O2

an average of 16% of MB following the analogous pathway through MB4J. The methyl ester therefore reduces the amount of fuel that follows channels that enhance reactivity and also enables the reactivity inhibiting pathway through MB3J. This effect offers an explanation for the absence of NTC behavior in MB combustion. Ostensibly, the alkyl chain of MB is too short to overcome this effect, which is consistent with previous observations by Walton et al. [26]. As well, the assymetry of the methyl ester moiety can be expected to decrease in significance as the alkyl chain increases in length. This theory is therefore consistent with observed NTC behavior in the combustion of larger methyl esters [14]. 2.2. Formation of oxygenated species

5-membered and 6-membered reactions, which led to the formation of HO2 and OH respectively. This result corroborates our sensitivity analysis, where the reaction MB2OO MB2OOH3J, with a 5-membered transition state, increased ignition delay, while the reaction MB2OO MB2OOH4J, with a 6-membered transition state, decreased ignition delay. The radical MB3J formed MB3OO, and then primarily reacted through a 5-membered ring, led to the formation of HO2. MB4J mainly followed a 6-membered ring pathway, leading to the formation of OH. Similarly, the fuel alkyl radicals of n-butane, sC4H9 and pC4H9, can react through 5- or 6- membered ring pathways. The radical sC4H9 behaved similarly to the radical MB2J, where the 5- and 6membered pathways led to HO2 and OH respectively. The sensitivity analysis in Fig. 4 reflects the effect of these pathways, where the 5-membered reaction sC4H9O2 C4H8OOH2-3 inhibits reactivity, whereas the 6-membered reaction sC4H9O2 C4H8OOH2-4 enhances reactivity. The radical pC4H9 mainly reacted through pathways leading to the formation of OH, namely through the reaction pC4H9O2 C4H8OOH1-3, a 6-membered isomerization. These results demonstrate how the methyl ester moiety changes the low temperature oxidation of methyl butanoate in contrast to its corresponding normal alkane, n-butane. Table 4 shows that, over the time period that we studied, an average of 58% of n-butane and 40% of MB reacted through sC4H9 and MB2J respectively, radicals which behave similarly. Due to the methyl ester, however, MB3J behaves differently than MB2J; namely, in the case of n-butane, the positions of these carbons are symmetrically equivalent and the corresponding radicals are not treated differently. In the same way, the radical pC4H9 is formed from two symmetrically equivalent carbons on either end of n-butane; in methyl butanoate, one of these carbons is bonded as part of the methyl ester moiety. This assymetry also explains the result that considerably more of n-butane, an average of 29%, reacted through the reactivity enhancing pathway involving pC4H9, compared with

To investigate the production of oxygenated species from fuelbound oxygen, we investigated shock tube oxidation of stoichiometric mixtures of both fuels at temperatures of 1100 K and 1600 K and a pressure of 4 atm. These conditions were chosen to match the range of conditions studied by Dooley et al. [9]. Using the shock tube model allowed us to capture the time dependence of CO2 or CO formation from methyl butanoate oxidation and contrast it with that of n-butane. Clearly, n-butane contains no fuel-bound oxygen; therefore understanding the formation of oxygenated species from n-butane facilitated the identification of reaction channels unique to the methyl ester. Our chosen stoichiometric conditions allowed oxidation of either fuel that was unconstrained by the amount of environmental oxygen. Fig. 7 compares n-butane and methyl butanoate combustion, at both 1100 K and 1600 K; the plots show temperature profiles, as well as the concentration profiles of MB, C4H10, O2, CO and CO2 plotted as a function of normalized residence time t ¼ st . Methyl butanoate combustion id produces a substantial amount of CO2 prior to ignition (t = 1), in comparison to that of n-butane, which is typically ascribed to the effect of the methyl ester, or specifically fuel-bound oxygen. We sought to capture the effect of this fuel-bound oxygen and conducted ROP analysis to elucidate the pathways involved in both CO and CO2 production from both fuels. The flux analysis focused on reaction pathways at t = 0.8 and t = 0.2 for the temperatures 1100 K and 1600 K respectively. These times were chosen to be approximately at the intersection of the concentration curves of MB decomposition and CO2 production. In this way, we ensured that there would be sufficient amounts of fuel and oxygenated species such that we could find pathways connecting these species. Under these conditions, the first step in the formation of oxygenated species was the formation of fuel alkylester or alkyl radicals from each fuel. The main reactions involved in this process are listed in Table 5; as well, the proportions of each fuel forming each radical are listed in Table 6 for both 1100 K and 1600 K. As mentioned in our analysis of reactivity, methyl butanoate’s

Table 3 Proportions of consumed fuel by specific reactions in our autoignition study. Normalized rates of production were computed at 2%, 33% and 50% fuel consumption (0.006 s, 0.084 s and 0.096 s). T = 780 K; P = 12.5 atm; / = 1. Time (s)

0.006 0.084 0.096

Reaction consuming methyl butanoate (%) R25

R35

R36

R371

R373

R379

R381

R385

R387

R403

R405

– 6.4 –

14.8 8.5 –

15.8 30.5 18.8

– 7.1 13.3

14.8 5.9 5.3

– 7.5 13.0

18.2 6.1 10.9

– 6.4 –

15.3 6.1 5.4

– – 7.0

10.2 – –

R745

Reaction consuming n-butane (%)

0.003 0.0106 0.0126

R738

R739

R740

R741

– – 5.5

– 11.2 17.1

32.7 24.0 23.8

58.4 40.9 39.8

– 6.2 –

21

K.C. Lin et al. / Fuel 92 (2012) 16–26

(a)

(a)

HO

O C-C-C-C-O-C MB2J

sC4H9

O +HO2 C-C=C-C-O-C

MB2OOH3J

5 atom ring + O2 isomerization

HO

+OH

O O C-C-C-C-O-C

OO C-C-C-C-O-C

MB2OOH4J

hydroperoxyketone

O C-C-C-C-O-C

OH O O O C-C-C-C

+ O2

O O O-O-C-C-C-C-O-C

(b)

C-C-C-C

O O C-C-C-C-O-C

+ O2

β-scission

MB2D

5 atom ring isomerization

O O C-C-C-C

HO O O C-C-C-C-O-C

O

β-scission

6 atom ring isomerization

+OH

HO O O O C-C-C-C-O-C hydroperoxyketone

O O C-C-C-C-O-C +OH cyclic ether

+ O2

cyclic ether

O + C-C C=C +OH

C=C + C-C

OH O C-C-C-C

pC4H9O2

+ O2

pC4H8OOH1-3

OH O O O C-C-C-C 6 atom ring isomerization

C=C-C

+OH O C-C-C-C

HO O O +OH C-C-C-C

+OH +CH2O

cyclic ether

6 atom ring isomerization

5 atom ring isomerization

O C-C-C-C

hydroperoxyketone

β-scission

pC4H8OOH1-2

MB4OO

MB4J

O C-C-C-C +OH

OH O C-C-C-C sC4H8OOH2-4

OH

O O C-C-C-C-O-C

+ O2

C-C=C-C +HO2

hydroperoxyketone

+ O2 6 atom ring isomerization

MB3OOH2J

(c)

OH O β-scission C-C-C-C sC4H8OOH2-3

pC4H9

MB3OO

O +HO2 C-C=C-C-O-C

C=C-C-C +HO2

O C-C-C-C-O-OH +OH

O

MB3J

O C-C-C-C-O-C

+ O2

6 atom ring isomerization

HO

Alkylester hydroperoxy peroxy radical

(b)

6 atom ring isomerization

sC4H9O2

O O C-C-C-C-O-C

OH O β-scission C-C-C-C sC4H8OOH2-1

5 atom ring isomerization

cyclic ether

HO

6 atom ring isomerization

5 atom ring isomerization

OO C-C-C-C

O C-C-C-C-O-C O

MB2OO

+OH OH OO O C-C-C-C-O-C

+ O2

MB2D

MB2OOHMJ 7 atom ring isomerization 6 atom ring isomerization

O

C-C-C-C

O O β-scission C-C-C-C-O-C

OH O O C-C-C-C-O-C

OH O OO O C-C-C-C-O-C

C=C-C-C +HO2

Fig. 6. Decomposition pathways at 33% fuel consumption for primary fuel alkyl radicals (n-butane) at T = 780 K. P = 12.5 atm; / = 1. (a) sC4H9; (b) pC4H9.

MB4OOH2J

Alkyl hydroperoxy peroxy radical

O C=C-C-O-C

+OH

Table 4 Percentages of each fuel forming specific alkylester (methyl butanoate) or alkyl (n-butane) radicals. T = 780 K; P = 12.5 atm; / = 1. Time (s)

MP2D

+CH2O

Fig. 5. Decomposition pathways at 33% fuel consumption for primary fuel alkylester radicals (methyl butanoate) at T = 780 K. P = 12.5 atm; / = 1. (a) MB2J; (b) MB3J; (c) MB4J.

Alkylester radical (% MB) MB2J

MB3J

MB4J

MBMJ

0.006 0.084 0.096

46 52 24

18 20 24

15 13 19

10 – 7

Avg.

40

20

16

6

Alkyl radical (% C4H10)

asymmetric structure leads to four distinct alkylester radicals, compared with two in the case of n-butane. The effect of increasing temperature from 1100 K to 1600 K was to enhance the importance of unimolecular reactions for both fuels. Thus, at 1600 K, increased proportions of the fuel followed reactions R5, R6 and R7 from MB, and reaction R724 from n-butane.

2.2.1. Carbon monoxide The next step in the formation of oxygenated species involved the decomposition of these fuel radicals. The reaction pathways that connected alkyl ester radicals to carbon monoxide at 1100 K are shown in Fig. 8. The alkylester radicals MB4J, MB3J, and MB2J contributed the most to CO formation. In particular, the pathway beginning with MB4J illustrates a mechanism for direct CO production from MB. MB4J underwent a b-scission reaction to form the methyl ethanoate radical (ME2J). ME2J further decomposed to CH2CO and CH3O. Similarly, MB3J formed C5H7O2, which subsequently produced CH3O. Table 7 lists the reactions and their relative proportions that directly contributed to CO formation for

pC4H9

sC4H9

0.003 0.0106 0.0126

33 24 29

58 58 57

Avg.

29

58

both temperatures. The majority of CO is formed from reactions involving either HCO or CH3CO. Fig. 8 shows that CH3O primarily formed CH2O and then HCO; additionally, CH2CO formed CH3CO through hydrogen addition. Consequently, a significant amount of CO formation in these conditions is directly traceable fuel-bound oxygen. In contrast, at 1100 K, n-butane produce CO through the radical sC4H9, which decomposed to C3H6 and CH3. The methyl radical, CH3, reacted with HO2 to form CH3O and OH. CH3O then formed CH2O which led to HCO and finally CO. The oxygen involved in this process obviously originated from the environment, which is demonstrated by this pathways dependence on the reaction of CH3 with HO2. Similar trends apply at 1600 K, where a significant

22

K.C. Lin et al. / Fuel 92 (2012) 16–26

(a)

(b)

(c)

(d)

Fig. 7. Temperature and concentration profiles of selected species for shock tube oxidation at P = 4 atm, / = 1 (1% fuel, 6.5% O2, 92.5% N2): t ¼ st ; (a) MB at T = 1100 K, id sid = 0.0075 s; (b) n-butane at T = 1100 K, sid = 0.00945 s; (c) MB at T = 1600 K, sid = 2.8  105 s; (d) n-butane at T = 1600 K, sid = 2.3  105 s.

Table 5 The most important reactions involved in methyl butanoate and n-butane consumption in our oxygenated species study. Reactions of MB

Reactions of C4H10

R5. MB (+M) MP3J + CH3 (+M) R6. BAOJ + CH3 MB R7. MB (+M) ME2J + C2H5 (+M) R371. MB + H MB4J + H2 R379. MB + H MB3J + H2 R385. MB + H MB2J + H2 R403. MB + H MBMJ + H2

R724. R738. R739. R740. R741.

C4H10 C2H5 + C2H5 C4H10 + H pC4H9 + H2 C4H10 + H sC4H9 + H2 C4H10 + OH pC4H9 + H2O C4H10 + OH sC4H9 + H2O

Table 6 Proportions of consumed fuel by specific reactions in our oxygenated species study. Normalized rates of production were computed for various T, t. P = 4 atm; / = 1. T (K)

t

1100 1600

0.8 0.2

Reaction consuming methyl butanoate (%) R5 – 1

R6 – 22

R7 – 5

R371

R379

R385

R403

25 20

21 13

18 10

14 12

Reaction consuming n-butane (%)

1100 1600

0.8 0.2

R724

R738

R739

R740

R741

– 20

13 20

37 37

15 6

25 9

portion of CO formed from MB was via the radical ME2J, which linked fuel-bound oxygen to CO formation. CO formation from nbutane again was clearly dependent on environmental oxygen.

2.2.2. Carbon dioxide Table 8 enumerates important reactions in carbon dioxide formation both before and after ignition. Unique pathways involved in CO2 formation from fuel-bound oxygen in methyl butanoate were easily identified, as n-butane produced negigible amounts of CO2 at the time of our flux analysis, namely t = 0.8 at 1100 K and t = 0.2 at 1600 K. At ignition, both fuels produced CO2 via

common pathways, namely through CO, that are less readily linked to fuel-bound oxygen. Prior to ignition however, MB produced a substantial amount of CO2 that was traced to the oxygenated methyl ester moiety. At 1100 K, fuel-bound oxygen in MB produced CO2 primarily through the the thermal decomposition of the methoxy formyl radical, CH3OCO, to CO2 and CH3, and to a lesser extent the reaction of HCCO with oxygen. The decomposition of MB through fuel alkyl radicals produced CH3OCO, which connects this CO2 production with the the oxygen in the methyl ester moiety. Also at 1100 K, the pathway through HCCO represents another link to fuel-bound oxygen. ME2J, which we have discussed in the context of CO formation, decomposed to CH3O and CH2CO, which formed CO. CH2CO also formed HCCO via hydrogen abstraction by OH (R534) and H (R535). Similarly, C5H7O2 decomposed to CH3O, which formed CO, as well as CH2CHCHCO. CH2CHCHCO reacted with H (R100) and OH (R101) to form HCCO. Table 9 summarizes the proportions of HCCO that each of these reactions formed. At 1600 K, the higher temperature enabled the reaction pathway through the butanoic acid radical (BAOJ). 22% of MB formed BAOJ at this temperature, which subsequently formed 40% of CO2 at this temperature through thermal decomposition. This channel also connects CO2 production to the methyl ester. Previous studies [4,18] have suggested that methyl butanoate ‘wastes’ oxygen by directly forming carbon dioxide from both fuel-bound oxygen atoms, instead of splitting this oxygen between different carbon atoms. We evaluated this notion by examining CO2 formation directly from both fuel bound oxygen atoms. Fig. 9 illustrates the reaction pathways that directly link both oxygen atoms in MB to CO2 formation at both 1100 K and 1600 K. These pathways involve the radicals CH3OCO and BAOJ, which is consistent with one of our previous studies [27]. For each reaction in a given pathway, Fig. 9 denotes the percentage of each reactant forming each product. The mathematical product of these percentages gives the proportion of MB that formed CO2 through that channel. Summing these proportions gives the amount of methyl butanoate that formed carbon dioxide using both fuel oxygen atoms. The results of this study showed that 15% and 28% of MB

23

K.C. Lin et al. / Fuel 92 (2012) 16–26 -CH3O, R375, 93%

CH2CO

CO

-CH3, R377, 11%

-H, R466, 11% +O2, R497, 52%

HCO

CH3CO

+H, R376, 63%

-CO, R377, 21% -CH3, R609, 14%

nC3H7

+OH, R484, 40% +H, R485, 31% +CH3, R490, 7%

ME2J

-H, R610, 72%

nC3H6

-C2H4, R390, 19%

-H, R378, 63% -CH2CO, R375, 28%

CH3OCO

-CO2, R383, 36%

CH3

CH2O

+HO2, R476, 30%

MP2D3J

CH3O

CH=CHC(=O)OCH3

-CH2CHCHCO, R91, 36%

-MP2D, R388, 13%

-H, R389, 99%

-C2H4, R374, 82%

C5H7O2

MP2D CH2=CHC(=O)OCH3

+OH, R69, 73% +H, R63, 19%

MB2D

MB4J

-CH3OCO, R382, 56%

-CH3, R388, 99% -H, R55, 73%

MB2J

-C3H6, R382, 72%

-H, R56, 24%

MB3J Fig. 8. Reaction pathways from fuel alkylester radicals (methyl butanoate) to carbon monoxide. T = 1100 K; t = 0.8 s; P = 4 atm; / = 1.

Table 7 Proportions of carbon monoxide formed by specific reactions. Normalized rates of production were computed for various T, t. P = 4 atm; / = 1. Fuel

t

Reactions at T = 1100 K (%) R377

MB C4H10

0.8 0.8

11 6

R466 11 11

R497 52 55

R1840

Table 8 Proportions of carbon dioxide formed by specific reactions. Normalize rates of production were computed for various T, t. P = 4 atm; / = 1. Fuel

t

MB

0.8 1 1

Reactions at T = 1100 K (%)

R2024

9 24

5 –

C4H10

Reactions at T = 1600 K (%)

MB C4H10

0.2 0.2

R1893

R465

R518

R2316

R2318

63 – –

28 – –

– 98 67

– – 15

– – 6

– – 6

Reactions at T = 1600 K (%)

R377

R466

R491

R497

R2024

R2326

18 23

26 24

6 6

20 19

9 –

6 12

Reaction number definitions R377 R466 R491 R497 R1840 R2024 R2326

R383

CH3CO CH3 + CO HCO + M H + CO + M HCO + CH3 CH4 + CO HCO + O2 CO + HO2 CH2CHO + O2 ? CH2O + CO + OH nC3H7CO nC3H7 + CO CH2(s) + O2 ? CO + OH + H

formed CO2 in this manner at 1100 K and 1600 K, respectively. Further discussion of these results involves examining their consequences in the context of the formation of soot precursors.

2.3. Implications on acetylene and ethylene formation We chose to study the formation of acetylene (C2H2), an important species that is a component of the hydrogen abstraction acetylene addition (HACA) soot formation mechanism. Based on prior

MB

0.2

R54

R383

1893

40

38

20

Reaction number definitions R54 R383 R465 R518 R1893 R2316 R2318

BAOJ CO2 + nC3H7 CH3OCO CO2 + CH3 CO + OH CO2 + H CH2 + O2 CO2 + 2H HCCO + O2 CO2 + HCO HOCHO + OH H2O + CO2 + H HOCHO + H H2 + CO2 + H

Table 9 Reactions forming specific proportions of HCCO. T = 1100 K; t = 0.8; P = 4 atm; / = 1. Reaction R100 R101 R534 R535

HCCO formed (%) CH2CHCHCO + H C2H4 + HCCO CH2CHCHCO + OH HCCO + CH3CHO CH2CO + OH HCCO + H2O CH2CO + H HCCO + H2

17 29 20 28

studies and our analysis of oxygenated species, the formation of CO and CO2 in the breakdown of methyl butanoate should inhibit

24

K.C. Lin et al. / Fuel 92 (2012) 16–26

(a)

(b)

MB

R371 -25%

R385 -18%

R379 -21%

MB4J

MB3J

MB2J

R388 -34%

R55 -66%

R56 -22%

MB3D

R57 -2%

R88 -3%

R382 -72% R89 -6%

CH3OCO

R390 -100%

R383 -93%

R385 R5 -10% -1%

MB2J R55 -63%

MB2D MP2D

R389 -1%

R6 -22%

MB

R58 -22%

BAOJ

R379 -13%

R54 -74%

MB3J MP3J

R8 -2%

R388 R396 -37% -39%

R397 -100%

CO2 R382 -67%

R383 -81%

MP2D MP2D3J

CO2

MB2D

R89 -11%

CH3OCO

Fig. 9. Reaction pathways from methyl butanoate to carbon dioxide. P = 4 atm; / = 1. (a) T = 1100 K and t = 0.8 s; (b) T = 1600 K and t = 0.2 s.

the formation of soot precursors, so we sought to understand the relationships between the pathways that form these species from the breakdown of MB and n-butane. We investigated soot precursor formation under similar conditions as we did for oxygenated species. To iterate, our analysis focused on shock tube oxidation of both fuels at a pressure of 4 atm and temperatures of 1100 K and 1600 K. However, to enhance the amount of acetylene production in the system, we chose an equivalence ratio of 3, consistent with a prior soot formation study performed by Westbrook et al. [18]. Fig. 10 compares the oxidation of n-butane and methyl butanoate at 1100 K and 1600 K; the plots show the time history of the temperature of the system, as well as the concentration of fuel, oxygen, ethylene and acetylene. Ethylene is an important species because it yields acetylene through hydrogen abstraction reactions. At 1100 K, there was insignificant acetylene production until around the time of ignition. However, the concentration of ethylene grew significantly prior to ignition; subsequently, the ethylene mole fraction decreased on the same time scale as the increase in acetylene concentration. Rate of production analysis at t = 1 showed that ethylene formed a significant portion of acetylene by hydrogen abstraction through the

(a)

(c)

ethenyl radical (C2H3); specifically, C2H4 + H C2H3 + H2 (R504), followed by C2H3 + H C2H2 + H2 (R503). Thus, the reaction pathways forming ethylene from the decomposition both fuels are key in the context of soot precursor formation. Figs. 11 and 12 show these pathways for both fuels at 1100 K and 1600 K respectively. For both fuels, the decomposition of fuel alkylester or alkyl radicals, themselves created by hydrogen abstraction, contributed the most to ethylene formation. In the case of MB, these radicals were MB4J, MBMJ and MB2J, while in the case of n-butane, these radicals were pC4H9 and sC4H9. The effect of fuel-bound oxygen is apparent in the decomposition of these radicals, inasmuch as the pathways that lead to ethylene from MB are related to those that form oxygenated species. For example, MB followed a pathway through MB2J that led to ethylene, but which also produced the methoxy formyl radical (CH3OCO). We previously observed that CH3OCO produces CO2 from fuel bound oxygen and thus has a role in soot reduction. This channel is consistent with previously postulated and observed soot reduction channels [18]. Fig. 11 also shows that MB4J underwent b-scission to form ME2J and C2H4 and that analogously, pC4H9 formed C2H5 and C2H4. As we have discussed, ME2J led to CO

(b)

(d)

Fig. 10. Temperature and concentration profiles of selected species for shock tube oxidation at P = 4 atm, / = 3 (1% fuel, 6.5% O2, 92.5% N2): t ¼ st ; (a) MB at T = 1100 K, id sid = 0.00615 s; (b) n-butane at T = 1100 K, sid = 0.00645 s; (c) MB at T = 1600 K, sid = 8.6  105 s (d) n-butane at T = 1600 K, sid = 8.8  105 s.

25

K.C. Lin et al. / Fuel 92 (2012) 16–26

(a)

O C-C-C-C-O-C MB

+H (R403) 15/81

-CH2O (R406) 99/99

MB2J

MP2D -H (R396) 64/99 O C-C-C-O-C

MP3J

sC4H9

+H,(R739) +OH,(R741) 62/88

pC4H9 -ME2J (R374) 97/72

C3H6

O C-C-C-C

-C2H4 (R696) 99/90

-C2H5 (R696) 99/49 +H (R619) 9/6

nC3H7CO

C2H5 -H,(R469) +O2,(473) 82/39

C2H4

-CO (R2024) 99/88

-CH3 (R609) 97/7

-H (R610) 11/98

nC3H7

nC3H7 -CH3OCO (R397) 99/6

C4H10

+H,(R738) +OH,(R740) 27/86

-CH3 (R691) 96/78

MBMJ

O C-C-C-C-O-C

O C=C-C-O-C

(b)

MB4J

O C-C-C-C-O-C

+H (R385) 20/63

-CH3 (R385) 31/99

O C-C-C-C-O-C

+H (R371) 28/88

-CH3 (R609) 97/18

C2H4

Fig. 11. Reaction pathways forming ethylene from each fuel at T = 1100 K and 30% fuel consumption. (a) Methyl butanoate; (b) n-butane. The notation ‘‘A/B’’ denotes that A% of the reactant produces B% of the product in the specified reaction channel.

and CO2, but C2H5 produced C2H4 via hydrogen abstraction. Therefore, the net effect of the methyl ester is to divert the production of C2H5 to ME2J, leading to oxygenated products. In this way, this result illustrates another soot reducing mechanism of the methyl ester structure. Reaction pathways at 1600 K, shown in Fig. 12, illustrate additional pathways for soot production that reflect the enhancement of unimolecular decomposition reactions by temperature. Similarly to observations at 1100 K, a portion of MB followed a pathway through ME2J, indicating that some fuel carbon was diverted from soot precursor formation. As well, the methoxy formyl radical was also involved at this temperature. The butanoic acid radical (BAOJ) became important at this temperature and directly formed CO2 and nC3H7 via a bond breaking reaction (R54). This production of CO2 removes potential fuel carbon for soot precursor formation, which is another pathway that has been previously observed. A consequence of these pathways was the formation of the methyl radical, CH3. The methyl radical was produced from: thermal decomposition of CH3OCO to CO2 and CH3; and the decomposition of MB to BAOJ and CH3; and b-scission of nC3H7 to form C2H4 and CH3. The abundance of CH3 enhanced the recombination reaction 2CH3 = C2H6. Ethane (C2H6) underwent successive hydrogen abstraction abstractions to form ethylene. Thus, fuel carbons in alkylester radicals that are not bonded directly to oxygen may still manifest in soot emissions through this pathway. Our analysis elucidates a mechanism for soot precursor reduction in MB oxidation. Previous literature has suggested that MB wastes fuel-bound oxygen through direct CO2 formation, as ideally, each fuel-bound oxygen should remove one carbon atom from soot precursor formation. In our discussion of the results of Fig. 9, direct CO2 production certainly represented a substantial portion of fueloriginated oxygenated species, but the classification as wasteful is merely a matter of perspective. Moreover, we observed additional reaction channels, namely those that produce CO, which split fuelbound oxygen between two different carbons. Ultimately, MB represents a significant amount of soot precursor reduction when compared with n-butane, which is unequivocally positive.

3. Conclusions This kinetic modeling analysis has illustrated a number of phenomena that can be ascribed to the oxygenated group present in methyl butanoate and other methyl esters. The methyl ester moiety changes the pathways that occur during methyl butanoate oxidation compared with n-butane. Most notably, methyl butanoate does not exhibit a region of NTC behavior, in contrast with n-butane. Sensitivity and reaction pathway analysis in the low temperature regime showed that this phenomenon can be ascribed to the effect of the methyl ester, whose assymetrical structure enables 5-membered isomerization pathways that inhibit reactivity. As well, we observed that the fuel-bound oxygen contained in the methyl ester moiety leads to significant production of both carbon monoxide and carbon dioxide, which cannot occur in the combustion of n-butane. In addition to previously studied pathways that formed CO2, we also found an additional pathway through the methyl ethanoate radical that leads to the formation of carbon monoxide. These reaction channels, as expected, were related to those that lead to the formation of soot precursors, namely ethylene and acetylene. These pathways account for the reduction in soot precursor production associated with methyl butanoate oxidation when compared to n-butane. Production of oxygenated radicals, such as methyl ethanoate radicals (ME2J), displace the production of alkyl radicals when compared to n-butane. Since these oxygenated radicals preferentially form oxygenated species, a reduction in soot precursor formation logically follows. In the future, it will be useful to know how these predictions scale with molecule size, as the effects of the methyl ester moiety would be expected to decrease with increasing alkyl chain length. Other studies, for example, have predicted NTC behavior for larger methyl esters such as methyl decanoate [15]. Larger methyl esters and biodiesel have exhibited lower soot emissions, which suggests that the methyl ester continues to have a beneficial effect. Ultimately, contrasting biodiesel combustion with conventional hydrocarbon fuels will help to elucidate the changes in performance as biodiesel displaces conventional diesel in fuel

26

K.C. Lin et al. / Fuel 92 (2012) 16–26

(a)

References

+CH3 (R478) 86/100

C2H6

CH3 -CO2 (R383) 82/17

O C-O-C CH3OCO -C3H6 (R382) 64/76

-nC3H7 (R8) 2/11

MB3J +CH3,+H,+OH (R463,R471,R474) -ME2J (R7) 95/38 5/54

O C-C-C-C-O-C MB

MBMJ

MB4J -H (R469) 100/13

+H,(R739) +OH,(R741) 49/94

sC4H9

pC4H9 C3H6

-CO2 (R54) 73/56

C 2 H4

C2H6

C4H10

+H,(R738) +OH,(R740) 25/93

-CH3 (R691) 94/39

BAOJ -ME2J (R374) 97/36

+CH3 (R478) 52/97

CH3

-CH3 (R609) 92/41

O C-C-C-C-O-C

O C-C-C-C-O

C2 H5

-C3H6 (R691) 94/47

-CH2O (R406) 69/99

-CH3 (R6) 24/100

+H (R371) 23/89

nC3H7

nC3H7CO

+H (R403) 14/78

O C-C-C-C-O-C

(b)

-CO (R910) 99/38

O C-C-C-C

+H 15/77 (R379)

O C-C-C-C-O-C

-C2H4 (R609) 92/37

-BAOJ (R6) 24/30

+CH3,+H,OH (R463,R471,R474) 90/41

-C2H5 (R724) 18/29

-C2H4 (R696) 99/22

C 2 H5

-C2H5 (R696) +H 99/15 (R619) 13/13

-H,(R469) 98/67

C 2 H4 -H (R610) 5/75

nC3H7

-CH3 (R609) 100/6

Fig. 12. Reaction pathways forming ethylene from each fuel at T = 1600 K and 30% fuel consumption. (a) Methyl butanoate; (b) n-butane. The notation ‘‘A/B’’ denotes that A% of the reactant produces B% of the product in the specified reaction channel.

blends. Understanding these changes will ideally facilitate the design of combustion technologies. Acknowledgments The authors would like to acknowledge the support of the Rackham Graduate School at the University of Michigan. Jason would like to express gratitude to the Natural Sciences and Engineering Council of Canada for his graduate fellowship. This research is also partially funded by the U.S. Air Force Office of Scientific Research – Grant F022058.

[1] Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt K, et al. IPCC, climate change 2007: the physical science basis. Contribution of working group I to the fourth assessment report of the intergovernmental panel on climate change. Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press; 2007. 996 pp. [2] Demirbas A. Importance of biodiesel as transportation fuel. Energy Policy 2007;35(9):4661–70. [3] Pienkos PT, Darzins A, promise The. The promise and challenges of microalgalderived biofuels. Biofuels Bioprod Bioref 2009;3(4):431–40. [4] Mueller CJ, Pickett LM, Siebers DL, Pitz WJ, Westbrook CK, Martin GC. Effects of oxygenates on soot processes in DI diesel engines: experiments and numerical simulations. Tech. Rep., Society of Automotive Engineers Inc.; 2003. [5] Lai JYW, Lin KC, Violi A. Biodiesel combustion: advances in chemical kinetic modeling. Prog Energy Combust Sci 2011;37(1):1–14. [6] Kohse-Höinghaus K, Osswald P, Cool TA, Kasper T, Hansen N, Qi F, et al. Biofuel combustion chemistry: from ethanol to biodiesel. Angew Chem Int Ed 2010;49(21):3572–97. [7] Fisher EM, Pitz WJ, Curran HJ, Westbrook CK. Detailed chemical kinetic mechanisms for combustion of oxygenated fuels. Symp (Int) Combust 2000;28(2):1579–86. [8] Metcalfe WK, Dooley S, Curran HJ, Simmie JM, El-Nahas A, Navarro MV. Experimental and modeling study of C5H10O2 ethyl and methyl esters. J Phys Chem A 2007;111(19):4001–14. [9] Dooley S, Curran HJ, Simmie JM. Autoignition measurements and a validated kinetic model for the biodiesel surrogate, methyl butanoate. Combust Flame 2008;153(1–2):2–32. [10] Gaïl S, Thomson MJ, Sarathy SM, Syed SA, Dagaut P, Diévart P, et al. A wideranging kinetic modeling study of methyl butanoate combustion. Proc Combust Inst 2007;31(1):305–11. [11] Huynh LK, Violi A. Thermal decomposition of methyl butanoate: Ab initio study of a biodiesel fuel surrogate. J Org Chem 2008;73(1):94–101. [12] Huynh LK, Lin KC, Violi A. Kinetic modeling of methyl butanoate in shock tube. J Phys Chem A 2008;112(51):13470–80. [13] Farooq A, Davidson DF, Hanson RK, Huynh LK, Violi A. An experimental and computational study of methyl ester decomposition pathways using shock tubes. Proc Combust Inst 2009;32(1):247–53. [14] Hakka MH, Glaude P, Herbinet O, Battin-Leclerc F. Experimental study of the oxidation of large surrogates for diesel and biodiesel fuels. Combust Flame 2009;156(11):2129–44. [15] Herbinet O, Pitz WJ, Westbrook CK. Detailed chemical kinetic oxidation mechanism for a biodiesel surrogate. Combust Flame 2008;154(3):507–28. [16] Curran HJ, Gaffuri P, Pitz WJ, Westbrook CK. A comprehensive modeling study of iso-octane oxidation. Combust Flame 2002;129(3):253–80. [17] Barckholtz TA. Modeling the negative temperature coefficient in the low temperature oxidation of light alkanes. Abstr Pap Am Chem Soc 2003;48(2):518. [18] Westbrook CK, Pitz WJ, Curran HJ. Chemical kinetic modeling study of the effects of oxygenated hydrocarbons on soot emissions from diesel engines. J Phys Chem A 2006;110(21):6912. [19] Beatrice C, Bertoli C, Giacomo N. New findings on combustion behavior of oxygenated synthetic diesel fuels. Combust Sci Technol 1998;137(1):31–50. [20] Yamane K, Kawasaki K, Sone K, Hara T, Prakoso T. Oxidation stability of biodiesel and its effects on diesel combustion and emission characteristics. Int J Engine Res 2007;8(3):307. [21] Haas MJ, Scott KM, Alleman TL, McCormick RL. Engine performance of biodiesel fuel prepared from soybean soapstock: a high quality renewable fuel produced from a waste feedstock. Energy Fuels 2001;15(5):1207–12. [22] Osmont A, Yahyaoui M, Catoire L, Gökalp I, Swihart MT. Thermochemistry of CO, (CO)O, and (CO)C bond breaking in fatty acid methyl esters. Combust Flame 2008;155(1–2):334–42. [23] Kee RJ, Rupley FM, Miller JA, Coltrin ME, Grcar JF, Meeks E, et al. Puduppakkam, CHEMKIN release 4.1. Reaction design, San Diego, CA; 2006. [24] Curran HJ, Gaffuri P, Pitz WJ, Westbrook CK. A comprehensive modeling study of n-heptane oxidation. Combust Flame 1998;114(1-2):149–77. [25] Hoffman SR, Abraham J. A comparative study of n-heptane, methyl decanoate, and dimethyl ether combustion characteristics under homogeneous-charge compression ignition engine conditions. Fuel 2009;88(6):1099. [26] Walton SM, Wooldridge MS, Westbrook CK. An experimental investigation of structural effects on the auto-ignition properties of two c5 esters. Proc Combust Inst 2009;32(1):255–62. [27] Lin KC, Lai JYW, Violi A. Insights into the differences in oxidation pathways of methyl butanoate and n-butane. In: Proceedings of the 6th U.S. national combustion meeting; 2009.