Diesel Surrogate

Energy 148 (2018) 424e447
Contents lists available at ScienceDirect
Energy
journal homepage: www.elsevier.com/locate/energy
A new methodology for diesel surrogate fuel formulation: Bridging

fuel fundamental properties and real engine combustion
characteristics
Yong Qian, Liang Yu, Zilong Li, Yahui Zhang, Leilei Xu, Qiyan Zhou, Dong Han, Xingcai Lu*
Key Lab. for Power Machinery and Engineering of M. O. E., Shanghai Jiao Tong University, 200240, Shanghai, PR China
a r t i c l e i n f o a b s t r a c t
Article history: Given the complexity and uncertainty of diesel compositions, it is of great challenges to study the
Received 11 April 2017 fundamental combustion processes and mechanism of diesel fuel, in particular the gaseous and partic-
Received in revised form ulate pollutant formation mechanisms. A reasonably designed diesel surrogate is proved to be an
29 January 2018
effective way to study the fundamental combustion mechanism of diesel fuel. However, most existing
Accepted 30 January 2018
diesel surrogates mainly adopt light hydrocarbon components, increasing the difficulty of accurately
Available online 3 February 2018
reflecting the physical and chemical properties of practical diesel fuel or the combustion and emissions
characteristics of diesel engines. Therefore, in this study, a methodology to construct diesel surrogates
Keywords:
Diesel surrogate
with C10 ~ C18 hydrocarbon components based on fuel properties and engine combustion and emissions
Combustion characteristics was proposed. First, the key physical and chemical fuel properties that affect fuel injec-
Emissions tion, atomization, ignition, combustion, engine efficiency and emissions were discussed in detail. Second,
Common rail diesel engine 13 candidate components were chosen to represent n-alkanes, iso-alkanes, cycloparaffins and aromatics
Fuel and engine interaction and blended with commercial diesel fuel in different proportions. Fuel injection, spray, ignition and
combustion phase, engine efficiency and emissions versus changed blending component and proportion
were systematically investigated. In particular, the effects of the cetane number and fuel volatility on
combustion and emissions were investigated under fixed injection timing and fixed ignition timing
conditions, respectively. Afterward, the key physical and chemical parameters of the surrogate fuels were
defined, and the constraint equations were constructed for different parameters. Considered the un-
certainties in the components and physicochemical properties of the actual diesel fuel, it is believed that
the construction of surrogate fuel has only reasonable solutions and there is no optimal solution. Finally,
three diesel surrogates were proposed from the feasible domains of constraint equations, including a 3-
Components surrogate (41.3% n-hexadecane, 36.8% 2,2,4,4,6,8,8-heptamethylnonane (HMN), 21.9% 1-
methylnaphthalene, by mol.), a 5-Components surrogate (21.6% n-hexadecane, 15.5% n-octadecane,
26.0% HMN, 20.7% 1-methylnaphthalene, 16.2% decalin, by mol.) and a 7-Components surrogate (21.5% n-
hexadecane, 15.4% n-octadecane, 25.8% HMN, 13.7% 1-methylnaphthalene, 8.1% decalin, 8.1% n-butyl-
benzene, 7.4% n-butylcyclohexane, by mol.). It is found that the surrogates proposed in this study can
accurately reproduce actual engine combustion characteristics. Also, the gaseous emissions of the 5-
Components surrogate are closest to those of the target diesel fuel in the tested engine operation
conditions.
© 2018 Elsevier Ltd. All rights reserved.
1. Introduction cylinder combustion process. It is well known that the physical

and chemical characteristics of the fuel directly impact fuel spray
In the continuous pursuit of internal combustion engines of high atomization, air-fuel mixture formation, mixture ignition and
efficiency and near-zero emissions, it is necessary to comprehen- combustion, emissions characteristics and engine efficiency [1].
sively understand the physical and chemical processes of the in- However, fossil fuels (including gasoline, diesel and kerosene) are
of complex composition, including hundreds or thousands of hy-
drocarbon components. In general, diesel is composed of C10~C24
* Corresponding author. hydrocarbons including 50e65% alkanes (mostly n-alkanes),
E-mail address: [email protected] (X. Lu). 20e30% cycloalkanes and 10e30% aromatics [2,3]. It is noteworthy
https://doi.org/10.1016/j.energy.2018.01.181
0360-5442/© 2018 Elsevier Ltd. All rights reserved.
Y. Qian et al. / Energy 148 (2018) 424e447 425
Abbreviations PM Particulate matter

RCM Rapid compression machine
ASTM American Society for Testing and Materials SMD Sauter mean diameter
ATDC After Top Dead Centre ST Surface tension
BP Boiling point T10 Temperature at 10% volume fraction recovery
BTDC Before Top Dead Centre T30 Temperature at 30% volume fraction recovery
CA Crank Angle T50 Temperature at 50% volume fraction recovery
CA10 Crank Angle at 10% Completion of Heat Release T70 Temperature at 70% volume fraction recovery
CA50 Crank Angle at 50% Completion of Heat Release T90 Temperature at 90% volume fraction recovery
CA90 Crank Angle at 90% Completion of Heat Release ULSD Ultra-low Sulphur diesel fuel
CN Cetane number DP Difference between injection pressure and ambient
CO Carbon monoxide pressure
Dist Distillation range Re Reynolds number
DIT Direct injection timing S Completely developed spray penetration
EGR Exhaust gas recirculation We Weber number
HC Hydrocarbon q Spray cone angle
HMN 2,2,4,4,6,8,8-Heptamethylnonane ∅ Equivalence ratio
ITE Indicated thermal efficiency ra Environmental air density
JSR Jet-stirred reactor rl Liquid density
LHV Lower heating value ua Environmental air viscosity
MCH Methyl cyclohexane ul Liquid viscosity
NTC Negative temperature coefficient h Efficiency
NOx Nitrogen oxide
that the specific ingredients of diesel, and the proportion of each and chemical properties of the real fuel. Edwards and Maurice [10]
component are not fixed, as many factors such as crude oil origin, considered that the methodology for surrogate fuel formulation
production season, refinery standards and processes may lead to could be divided into two categories: the physical surrogate, which
changes in fuel composition. Fig. 1 compares the compositions of is targets to simulate the physical characteristics of commercial
the diesel fuels used in the US #2 ULSD fuel [4] and China stage V fuels, and the chemical surrogate which usually contains the
diesel fuel. There is a significant difference between these two fuels chemical composition of the main components in commercial fuels.
in the proportion of alkanes, alkenes and aromatics. The change in Actually, each physical and chemical characteristic of the com-
fuel compositions is a huge barrier to the comprehensive under- mercial fuel may influence the entire combustion process and en-
standing of the engine emissions formation mechanism. In addi- gine performance [11]. Mueller et al. [12] selected nine kinds of
tion, the uncertainty of fuel composition will also make it difficult candidate components to construct two kinds of 8-component
to clarify the precise interaction between the fuel and the engine. diesel surrogate fuel, which can reproduce the distillation charac-
For the above reasons, it is common to use surrogate fuel to emulate teristics of the diesel fuel accurately. They further established 4-
commercial fuel in the investigation of combustion properties in component, 5-component, 8-component and 9-component surro-
fuel science [5e8]. Recently, to construct a diesel surrogate fuel, gate fuels and compared the physical and chemical properties of
international scholars have carried out a great deal of research [4,9]. the surrogate fuels and diesel [13]. Therefore, a reasonable surro-
The purpose of creating surrogate fuel is to simulate the physical gate fuel should be able to accurately simulate the features of the
(a) Components of US #2 ultra-low-sulfur diesel fuel [4] (b) Components of China diesel fuel (w%)
Fig. 1. Components of US #2 ULSD fuel and China stage V diesel fuel.

426 Y. Qian et al. / Energy 148 (2018) 424e447
objective fuel, including fuel characteristics (chemical components, above, n-alkanes represent a high proportion of diesel fuel, and
C/H ratio, density, molecular weight, volatility, cetane number/oc- some light n-alkanes have been widely researched as a diesel sur-
tane number, adiabatic flame temperature, combustion velocity, rogate fuel in previous studies [28,29]. As early as 1998, Curran et al.
etc.), engine combustion characteristics (ignition, heat release, [30] considered n-heptane one of the basic reference components
performance, emissions) [14]. The accuracy of surrogate fuel can be for diesel and constructed a detailed chemical reaction mechanism
verified by testing the intermediate products of the fuel reaction, of n-heptane. Kelly et al. [31] demonstrated that the laminar flame
the ignition characteristics, the flame characteristics (flame speeds of C5eC8 n-alkanes were basically similar. In further studies
extinction, laminar flame speed, turbulent flame structure) with on diesel surrogate fuel, the shortcomings of n-heptane and other
the standard combustion equipment (premixed flame, shock tube, short-chain n-alkanes as a diesel surrogate fuel in physical property
rapid compression machine, jet-stirred reactor, flow reactor, etc.). emulation, especially volatility, viscosity and other aspects, are
On the other hand, it can also be verified by testing the perfor- concerned. Researchers then began to study the combustion char-
mance parameter of the actual engine (combustion history, emis- acteristics of hydrocarbons with a carbon number higher than 10. Ji
sions, efficiency etc.). et al. [32] studied the laminar premixed flame propagation and
extinction of C5eC12 n-alkanes at atmospheric condition. They
1.1. Effects of physical properties of diesel surrogates on fuel spray found that the laminar flame speeds of all studied n-alkane/air
mixtures were similar to each other. Shen et al. [33] measured the
In a diesel engine, the combustion rate is mainly controlled by ignition delay of n-heptane, n-decane, n-dodecane and n-tetrade-
the fuel/air mixing rate. The physical features of the fuel directly cane and found that the ignition delay of the tested n-alkanes was
determine the spray atomization, fuel-air mixing process, in- quite similar to each other over a wide range of pressures and
cylinder fuel-air distribution, combustion velocity and duration, temperatures. Ranzi et al. [34] studied the pyrolysis, partial
efficiency and emissions [15e17]. Thus, the components of the oxidation, and combustion characteristics of heavy n-alkanes. It
surrogate fuel and their vaporization characteristics are widely was found that the chemical kinetics of different n-alkanes were
concerned [18,19]. similar in a wide temperature and equivalence ratio range.
Myong et al. [20,21] investigated the spray characteristics of n- Due to the presence of branched chains in the molecular
octane, n-dodecane, n-hexadecane and their 3-component mix- structure, the reaction kinetics of iso-alkanes are quite different
tures. They found that components with a higher boiling point from n-alkanes. Studies have shown that the proportion of iso-
dominated the liquid phase length during the vaporization process. alkanes in fuels has a significant effect on fuel ignition character-
Aye et al. [22] studied the effects of different proportions of n- istics [35,36]. Curran et al. [37] established a detailed reaction ki-
decane/1-methylnaphthalene mixtures on spray penetration. It netics model of iso-octane based on the experiment results of jet-
was found that the addition of n-decane, which had lower satu- stirred reactors (JSR), flow reactors, shock tubes, and engines.
rated vapor pressure, resulted in a decrease in the penetration Sarathy et al. [38] studied the chemical kinetics of the oxidation of
distance and an increase in the spray cone angle. Zhang et al. [23] 2-methylalkanes from C7 to C20. They found that iso-alkanes had
studied the spray and combustion process of multi-component longer ignition delay than n-alkanes given the same carbon chain
surrogate fuel and found that light components were gasified length. 2-methylalkene-like intermediate products had a signifi-
during the spray process, and the components with larger mole- cant impact on the reaction process. Ji et al. [39] studied the effects
cules were mostly distributed around the spray beam. Sibers et al. of the branched chain structures on laminar flame velocity with 2-
[24,25] compared the spray liquid penetration length of HMN, n- methylheptane, 3-methylheptane, 2,5-dimethylhexane and iso-
hexadecane and diesel fuel. They found that the liquid penetration octane as fuels. The results showed that the laminar flame propa-
length of diesel was longer than the other two fuels. The difference gation velocity decreases as the number of branches increases.
was obvious in low temperature conditions (700 K) and became In the process of diesel preparation, catalytic cracking, hydro-
less obvious in high temperatures. Iyengar et al. [26] studied the genation and other processes will make polycyclic alkanes and
effects of the fuel physical properties on the spray. They used n- polycyclic aromatics translate into monocycloalkane or bicy-
tetradecane and mixtures of n-tetradecane/n-decane/1- cloalkane. Sivaramakrishnan et al. [40] measured the reaction rate
methylnaphthalene as diesel surrogate fuels. The study indicated constants of cyclopentane, cyclohexane, methylcyclopentane and
that n-decane makes the 3-component mixed fuel evaporate earlier methylcyclohexane. The study found that methylcyclopentane and
than the n-tetradecane. Srivastava et al. [27] numerically studied methylcyclohexane had a higher reaction rate constant than
the turbulent spray of multi-component diesel surrogates. They cyclopentane and cyclohexane. Natelson et al. [41] demonstrated
compared the fuel injection distance between the binary surrogate that the negative temperature coefficient (NTC) phenomenon was
(n-decane/1-methylnaphthalene) and the 6-component surrogate present in n-butylcyclohexane oxidation by researching n-butyl-
(n-decane/n-dodecane/n-tetradecane/n-hexadecane/toluene). It cyclohexane low-temperature oxidation kinetics in a flow reactor.
was found that with the increased ambient air density or the Zhu et al. [42] studied the pyrolysis and oxidation mechanism of
increased ambient temperature, the liquid penetration distance decalin by shock tube. They first discovered the NTC phenomenon
decreases gradually. Under roughly the same conditions, the 6- of decalin in a temperature range below 920 K. After the H-atom
component surrogate had a higher penetration than the binary abstraction reaction, the cleavage of the CeC bond resulted in the
surrogate. conversion of the bicycloalkane to the monocycloalkane. Li et al.
[43] studied the intermediate products of decalin in non-premixed
1.2. Chemical reaction kinetics of diesel surrogates flame and found that soon after ring cracking and the isomerization
reaction, decalin transformed to C1~C4 small molecular
Compared with studies on the influence of surrogate compo- hydrocarbons.
nents on physical properties, a large amount of studies has focused Aromatic hydrocarbons are considered to be the main sources of
on the combustion characteristics (ignition delay, flame propaga- particulate matter during the combustion of diesel fuels due to
tion rate, emissions formation and so on) of the diesel surrogate their high degree of unsaturation [44,45]. The arenes in diesel are
components and the multi-component surrogate fuel. The com- complicated, including benzene, toluene, ethylbenzene, pro-
ponents in diesel fuel are mainly divided into four categories: n- pylbenzene, butylbenzene, 1-methylnaphthalene, etc. Davis et al.
alkanes, iso-alkanes, cycloalkanes and aromatics. As mentioned [46] compared the laminar flame speeds of benzene and toluene.
They found that the laminar flame speed of toluene was lower than components and the engine had not been detailed studied although
benzene. Darcy et al. [47,48] measured the ignition delay of n- some surrogate had been proposed with heavier components
propylbenzene by rapid compression machine and shock tube. [12,13]. Therefore, the ignition, combustion, performance and
They found that n-propylbenzene did not show typical low- emissions of the real engines, as well as the physical and chemical
temperature chemistry (namely NTC behavior). Nakamura et al. characteristics of the fuel should be taken into account simulta-
[49] measured the ignition delay over the full temperature range neously for reasonable surrogate fuel formulation.
for n-butylbenzene with rapid compression machine and high- Based on reasons above, the brand new methodology for diesel
pressure shock tube. They compared the combustion characteris- surrogate formulation was proposed in this paper. As illustrated in
tics between n-propylbenzene and n-butylbenzene and found that Fig. 2, the methodology includes six aspects. Firstly, the main
n-butylbenzene had higher reactivity than n-propylbenzene at components of diesel fuel and their concentration distribution
lower temperatures. They believe that this is mainly due to the were measured and analyzed first. Then, thirteen surrogate
wider variety of low temperature reactions available to the n-butyl candidate components (including four n-alkanes, three iso-alkanes,
compared to the n-propyl side-chain. Dagaut et al. [50] researched two cycloalkanes and four arenes) were chosen and blended into
the oxidation mechanism of tetralin with a jet-stirred reactor. By the diesel. The properties of mixed fuels were calculated or
measuring the concentration of stable intermediate products, a measured including density, viscosity, surface tension, cetane
semi-detailed kinetic reaction scheme (~10000 reactions and~400 number, distillation range, H/C ratio and other physical and
species) for the oxidation of decalin was proposed. Wang et al. [51] chemical properties. After that, for a comprehensive assessment of
researched the high-temperature autoignition of tetralin using a the effects of each component on engine performance and emis-
shock tube and compared it with toluene, decalin and cyclohexane. sions, experiments fueled with different mixtures were conducted
They found that tetralin was nearly wholly consumed via H on a modified single cylinder engine. For each fuel mixture, under
abstraction reactions to form tetralyl radicals. Mati et al. [52] the same engine loads, there are two groups of experiments,
studied the oxidation mechanism of 1-methylnaphthalene in JSR including the control strategy of the same injection timing and
and proposed a detailed kinetics mechanism. same ignition timing, to clarify the effects of ignition and volatility
Based on the research on the ignition and combustion reaction on engine operation, respectively. The effects of different compo-
mechanisms of diesel surrogate components, various diesel surro- nents can be analyzed by comparing the characteristics of their
gates have been studied. In previous studies, many researchers mixture with diesel. For ease of description, the mixtures of sur-
have adopted mixtures of n-heptane and toluene as a diesel surrogate candidates and diesel fuel were named with the surrogate
rogate fuel. However, the molecular weight of these two compo- candidates. Further, the author analyzed the main factors (injection
nents is very different from that of the diesel fuel. The researchers timing, cetane number, volatility, degree of unsaturation, carbon
chose components with higher molecular weight to represent chain length) that affect efficiency and regulated emissions and
diesel. Wang et al. [53] took n-decane and 1-methylnaphthalene as studied the effects of components on the combustion process in
diesel surrogate components and compared the ignition delay of detail. Then, the relationship between efficiency, engine emissions
their mixtures in different mole ratios (the cetane number is and the above main factors were systematically analyzed. Finally,
respectively 58 and 28). Ramirez et al. [54] used mixtures of 70% n- based on the overall consideration of the engine combustion pa-
decane and 30% 1-methylnaphthalene as diesel surrogate fuel and rameters and physical and chemical characteristics of the fuel, the
studied the oxidation process in JSR to set up the detailed kinetics constraint equations of surrogate fuel formulation were proposed,
mechanism. Pei et al. [55] used n-dodecane and m-xylene as diesel and three kinds of diesel surrogate fuels with different components
surrogate components and proposed a detailed mechanism with were developed to simulate engine performance and emissions.
2885 species and 11754 elementary reactions. Afterwards, they
reduced it to 163 species and 887 elementary reactions in order to 2. Testing fuels
apply it in the 3-D CFD calculation. Mati et al. [56] proposed a 5-
component diesel surrogate fuel with n-hexadecane, n-propylcy- 2.1. Commercial diesel fuel components
clohexane, n-propylbenzene, iso-octane and 1-methylnaphthalene
as components. A detailed mechanism including 377 species and The article first analyzed China stage V 0# diesel in detail. The
2755 reactions was developed. Ra et al. [57] put forward a 6- proportions of n-alkanes, iso-alkanes, aerenes and cycloalkanes are
component diesel surrogate fuel with n-tetradecane, toluene, shown in Fig. 1b, and other statistical results are shown in Fig. 3.
cyclohexane, dimethyl ether, alcohol and methyl butyrate as com- The fuel component content is determined according to ASTM
ponents. Naik et al. [58] developed a general surrogate fuel com- D2425 [60]. As can be seen in Fig. 3, the carbon atom numbers of
bustion mechanism containing 26 components that was applicable alkanes in diesel are mainly in the range of C9eC26, and even about
to nature gas, gasoline, diesel and even alcohol. Liang et al. [59] 95 w% are in the range of C11eC24. For cycloalkanes, 43.6 w% is
tried to use Naik's detailed combustion mechanism to efficiently monocycloalkanes and 41.6 w% is bicycloalkanes. Only about 14.8 w
calculate the diesel combustion process. They found that it was % is tricyclic cycloalkanes. For aromatics, 84.6 w% are monocyclic
feasible to use n-tetradecane to study the spray and combustion aromatic hydrocarbons and about 13.6 w% are bicyclic aromatics.
properties of diesel fuel or use n-tetradecane for spray prediction Only small amounts of tricyclic aromatics were detected, about 1.8
and n-heptane for ignition and combustion prediction to improve w% of the total amount of aromatics.
the calculation efficiency.
2.2. Selection of candidate components
1.3. The purpose of this paper
Farrell et al. [3] systematically summarized the significance of
As mentioned above, many researchers have noted that tradi- each component in the diesel surrogate fuel formulation. They
tional surrogate formulation was mainly from the perspective of considered that n-heptane, n-decane, n-dodecane, n-tetradecane
physical and chemical characteristics that may not reflect the per- and n-hexadecane in n-alkanes; HMN in iso-alkanes; methyl-
formance of the real engine. Moreover, the fuel components are cyclohexane, ethylcyclohexane, propylcyclohexane, butylcyclo-
very complex, and each component may impact engine perfor- hexane and decalin in cycloalkanes; and toluene, ethylbenzene,
mance and emissions. The interaction between the fuel propylbenzene, butylbenzene, xylene and decylbenzene in
Analysis of Physical and Chemical Properties of Diesel
Selections of candidate components
Evaluation of engine combustion and emissions
Constraint equation
Surrogate formulation
Engine combustion and emissions verification
Fig. 2. The methodology proposed in this paper for diesel surrogate formulation.
monocyclic aromatics were the important components in diesel methyl cyclohexane was chosen as the monocycloparaffins to set
surrogate fuel. In addition, tetralin and 1-methylnaphthalene were up surrogate fuels.
the important surrogate components.
For surrogate fuel formulation, the availability of chemical ki-
netics of components is important for the further application of 2.3. The test fuels and their properties
surrogate fuels on combustion characteristics [61]. On the other
hand, as much as possible to collect the high-purity components Because many components in diesel cannot be employed
that can be applied on engine conditions to test the effects of directly on a compression ignition engine, a relatively higher pro-
different components on engine combustion and emissions, the portion of each component was blended into the commercial diesel
final selection range for surrogate components would be broader. oil to guarantee it as the dominant component in the mixture and
Therefore, the experiment chose n-decane, n-dodecane, n-tetra- to control the physical and chemical characteristics of mixed fuel.
decane and n-hexadecane as n-alkanes, iso-octane, iso-dodecane Then, a series of experiments were conducted on the engine, and
and iso-dodecane as iso-alkanes, methylcyclohexane and decalin as the effects of each component on the engine were studied by
cycloalkanes, and toluene, butylbenzene, tetrahydronaphthalene comparing the combustion and emission characteristics of mixed
and 1-methylnaphthalene as arenes to study the effects of surro- fuel and commercial diesel oil.
gate candidate components on the engine performance and emis- Based on the analysis of diesel components (Figs. 1 and 3), al-
sions. The thirteen components adopted in this experiment are kanes account for about 40e50% and cycloalkanes and arenes ac-
listed in Table 1. The main properties are from Ref. [62] except ce- count for about 10e30%, respectively. Thus, in this paper, 50% n-
tane number. It should be noted that the selection of the candidate alkanes and 30% iso-alkanes, cycloalkanes and arenes were blended
components was based on two principles. First, candidate compo- into commercial diesel oil to formulate 13 kinds of mixed fuels. In
nents can be run in the engine after blending with diesel. Second, studies of the physical and chemical characteristics of multi-
the amount of candidate components needs to meet the engine test component surrogate fuels, many researchers adopted a linear
requirements. In the beginning, n-octadecane was selected as a interpolation method based on volume or mole fraction to predict
candidate component. Due to its higher boiling point, blending of some parameters. For example, Kim et al. [63] used it to calculate
50 vol% n-octadecane leads to the crystallization of the fuel in the cetane number, LHV, density, surface tension and distillation range.
high pressure common rail system. So, the results of n-octadecane/ Narayanaswamy et al. [64] applied it to calculate H/C ratio, cetane
diesel were not given. But the components with higher boiling number, smoke point, etc. Actually, some of the physical and
point were very important to make the volatile much closer to chemical characteristics of multi-component surrogate fuel are
diesel. So, n-octadecane were added in the final diesel surrogates. difficult to predict by the linear interpolation method [65,66],
As for n-butylcyclohexane, it is too expensive and its availability is especially cetane number, distillation range, smoke point, etc.
not enough. Due to its boiling point higher than methyl cyclo- Therefore, the study measured some properties including density,
hexane and closer to diesel components (methylcyclohexane is viscosity, surface tension and distillation range but used the linear
obviously not suitable for diesel surrogate components). So, in the interpolation method to calculate the LHV and H/C ratio [63,64].
final steps, the more expensive n-butylcyclohexane instead of The formula is shown as the following:
15
X
N
LHVmix ¼ xi LHVi (2)
1
12
Fig. 4 shows the distillation curves of test fuels and commercial
diesel oil. The distillation range of mixed fuel were measured ac-
Mass fraction (%)
9 cording to ASTM D86-17 [67]. It can be concluded that: (1) mixed

with surrogate components which have a lower boil point resulted
in higher T10 deviation; (2) the boiling point of n-hexadecane is
6 close to that of diesel T50, resulting in little difference in T50 after
blending 50% n-hexadecane; (3) the 30% volumetric blending ratio
has little effect on T90; (4) the polar property of some components
3 resulted in a nonlinear phenomenon of the fuel distillation range;
(5) for the cetane index, based on the European standard and China
national standards for diesel, it can be calculated according to ASTM
0
8 10 12 14 16 18 20 22 24 26 D976-06 [68].
Number of carbon atoms in alkanes
3. Experimental procedures and analysis methods
(a) The distribution of the carbon number of alkanes 3.1. The main properties that affect engine performance
60
The efficiency (h) and normal emissions (CO, HC, NOx and PM)
50 of the internal combustion engine are the most concerned pa-
rameters. But the physical characteristics (density, viscosity, surface
tension and distillation range), combustion characteristics (octane
40
Mass fraction (%)
number/cetane number, adiabatic flame temperature and flame

propagation speed) and molecular structures (carbon chain length,
30 saturation, cycloalkanes ring number and arenes ring number) have
too complicated relationship with the emissions and efficiency of
20 the engine. In this part, the main parameters affecting efficiency
and emissions are discussed in detail.
In conditions of a certain engine speed and load, the spray at-
10 omization, droplet evaporation, fuel-air mixing and the ignition
and combustion are the core parameters affecting emissions and
0 efficiency. The test system for fuel spray characteristics was intro-
Monocyclic Bicyclic Tricyclic duced in Ref. [69]. The camera used in the test was a high-speed
(b) Compositions of cycloalkanes CMOS camera. The exposure time was 1.1 ms. The nozzle hole
number and diameter of the injector were 1 and 0.176 mm,
100
respectively. The injection pressure was fixed at 120 MPa. The
ambient temperature was controlled at 293 C and the ambient
80 pressure was 1 bar. A comparison of the diesel spray cone angle and
spray penetration with mixed fuels is shown in Fig. 5. It can be seen
that the spray properties obviously changed because of the physical
Mass fraction (%)
60 properties changing. According to Table 1, in terms of density,

decalin diesel > n-tetradecane; in viscosity respect,
diesel [ decalin ¼ n-tetradecane; as for surface tension,
40 diesel ¼ decalin > n-tetradecane. But the real spray penetration is
decalin > diesel > n-tetradecane, and the real spray cone angle is n-
tetradecane > diesel > decalin. The empirical equations for jet flow
20 atomization are as follows [70,71]:
0:26
0 q ¼ a ra=r (3)
l
Monocyclic Bicyclic Tricyclic
(c) Compositions of aromatics 0:47 0:26
SMD ¼ bRe0:25 We0:32 ra=r ul=u (4)
Fig. 3. Detailed analysis of diesel fuel components. l a
S ¼ gðDP=ra Þ0:25 (5)

P
N
xi n H
nH
mix 1
i q, SMD and S indicate the spray cone angle, Sauter mean diameter
ðH=CÞmix ¼ ¼ (1)
nCmix P
N and completely developed spray penetration, respectively; a;   b;   g
xi nCi are the constants about geometry of the nozzle. Re is Reynolds
1
number; We is Weber number; ra ;   ua ;   rl ;   ul are environmental
air density, viscosity, liquid density and viscosity, respectively; DP is
Table 1
The molecular structure and physicochemical properties of the selected candidate surrogate components.
Name Molecular structure Purity Molar Weight Density CN ST BP LHV H/C ratio
% g/cm3 103 N/m
C MJ/kg
@25 C @101 kPa
n-Decane 99 142.29 0.732 76 23.41 174 44.236 2.20
n-Dodecane 98 170.34 0.748 80 24.94 216 43.949 2.17
n-Tetradecane 98 198.39 0.761 95 26.15 253 44.020 2.14
n-Hexadecane 99 226.45 0.773 100 27.09 287 43.952 2.13
iso-Octane 99 114.23 0.695 14 18.32 99 44.342 2.25
iso-Dodecane 95 170.34 0.750 e 170 46.1 2.17
HMN 98 226.45 0.793 15 240 43.4 2.13
Methylcyclohexane 98 98.19 0.77 20 23.30 101 43.356 2.00
trans-Decalin 98 138.25 0.896 48 29.38 197 42.618 1.80
Toluene 99.9 92.14 0.87 9 27.93 111 40.523 1.14
n-Butylbenzene 96 134.22 0.860 e 28.63 183.1 41.457 1.40
1-Methylnaphthalene 97 142.20 1.02 0 39.71 240 39.259 0.91
Tetralin 97 132.21 0.970 e 33.16 207.2 40.524 1.20
Table 2
The mixed fuel used in this study and their main physical and chemical properties.
Components Blending ratio with Viscosity Cetane index Density LHV Distillation curves ( C)
diesel (Vol %) mm2/s g/cm3 MJ/kg
T10 T30 T50 T90
@20 C
Diesel e 4.22 52.8 0.830 42.5 214 243 267 338

n-Decane 50 2.11 48.8 0.781 43.3 175 182 195 329
n-Dodecane 50 2.74 56.6 0.790 43.1 210 217 226 314
n-Tetradecane 50 3.21 61.9 0.793 43.4 222 238 247 292
n-Hexadecane 50 4.17 65.0 0.801 43.2 225 264 274 304
iso-Octane 30 2.06 e 0.787 43.1 123 153 225 328
iso-Dodecane 30 2.84 48.5 0.802 43.7 185 197 216 325
HMN 30 3.84 50.4 0.819 42.8 213 229 242 325
Methylcyclohexane 30 2.42 e 0.812 42.7 115 154 243 327
Decalin 30 3.18 e 0.839 42.9 190 202 216 326
Toluene 30 1.66 32.5 0.832 42.4 121 149 203 322
n-Butylbenzene 30 2.56 40.0 0.838 42.2 189 202 229 331
1-Methylnaphthalene 30 3.06 29.73 0.882 41.6 209 228 239 327
Tetralin 30 2.9 29.2 0.865 42.2 198 211 222 327
100 100
80 80
Percent recovery (vol %)

Diesel
60 n-Decane 60
n-Dodecane
n-Tetradecane
Diesel
40 n-Hexadedane 40
iso-Octane
iso-Dodecane
20 20 HMN
0 0
150 200 250 300 350 400 100 150 200 250 300 350 400
o o
Temperature ( C) Temperature ( C)
(a) n-Alkanes (b) iso-Alkanes
100 100
80 80

60 60
40 40
Diesel
20 Diesel 20 Toluene
Methylcyclohexane n-Butylbenzene
Decalin 1-Methylnaphthalene
0 0 Tetralin
100 150 200 250 300 350 400 100 150 200 250 300 350 400
o o
Temperature ( C) Temperature ( C)
(c) Cycloalkanes (d) Aromatics
Fig. 4. The distillation range of ten mixed fuels and diesel.
the difference between injection pressure and ambient pressure.

Based on the empirical formula and real spray properties, the
H
factors dominating the spray penetration, spray cone angle and CO ¼ fð∅;  TmaxÞ ¼ f ∅;  CN;  Dist;   (7)
C
droplet average diameter are liquid density, viscosity, viscosity and
surface tension, respectively. Thus, the key physical parameters are
density, viscosity and surface tension.
H
As studied by Heywood [72], the engine efficiency and emis- HC ¼ f CN; ;  Dist (8)
sions depend entirely on the mixture and its ignition and com- C
bustion processes. The main factors affecting efficiency are CA50
and in-cylinder average temperature; CA50 is mainly determined
H
by cetane number, fuel volatility or distillation range, adiabatic Soot ¼ f ∅; Dist;    (9)
flame temperature (H/C), injection timing and equivalence ratio. C
For emissions, CO emissions are dominated by the equivalence ratio
and combustion temperature. HC emissions are mainly determined
by the mixture reaction reactivity, fuel volatility or distillation H
NOx ¼ f ðTmax;  ½OÞ ¼ f CN;  Dist; ; ∅ (10)
range, and molecular structure. NOx emissions are mainly C
controlled by the maximum in-cylinder combustion temperature,
the duration of high temperature and oxygen concentration. Soot h is the indicates thermal efficiency of the internal combustion
emissions are related to the equivalence ratio, fuel volatility and engine. CA50 is defined as the crank angle position at which
molecule structure (molecular saturation). Therefore, the efficiency 50% of the fuel is burned. Tmax is the maximum in-cylinder
and four kinds of emissions can be expressed as the following temperature. CN is the cetane number. Dist is the distillation
formulas: range of fuel. H/C is the hydrogen atoms and carbon atoms
ratio of the fuel. DIT is the fuel direct injection timing. ∅ is the
equivalence ratio of the in-cylinder fuel and air. ½O is the in-
cylinder local oxygen concentration. It can be seen that ce-
H
h ¼ f ðCA50; Tmax;Þ ¼ f CN; Dist; ; DIT; ∅ (6) tane number, H/C ratio, density, viscosity, surface tension and
C
distillation range are the primary properties of engine perfor-
mance and emission.
characteristics in the same injection timing (fixed at -7oCA ATDC)

and in the same ignition timing (fixed at 3oCA ATDC). In a general
way, under the same injection timing condition, the cetane number
is the main parameter to determine the ignition timing. Under the
same ignition timing condition, the impacts of fuel volatilities on
the combustion and emission characteristics are enhanced. Thus,
factors like cetane number, distillation range, molecule structure
and carbon chain length can be discussed relatively independently
by the above two control strategies.
The in-cylinder gas pressure was measured by a pressure
transducer (Kistler model 6125B), the charge output from which
was converted to an amplified voltage using an amplifier (Kistler
model 5015A). The heat release rate and in-cylinder temperature
(a) Spray of diesel fuel (up) and n-tetradecane/diesel blend fuel (down) are calculated by a combustion analyzer. CO emissions were
70 measured by non-dispersive infra-red analyzer (NIDR CAI 602P),
60
total hydrocarbon (THC), and NOx emissions were measured by a
heated flame ionization detector (HFID CAI 600) and heat chemi-
Spary tip penetration (mm)
50
luminescent analyzer (HCLD CAI 600), respectively. For the mea-
40 surement of gas emissions, the sample were recorded at a rate of
30
1 Hz for 5 min. The particulate matter distribution was measured by
a fast particulate analyzer (DMS500) at a rate of 1 Hz for 5 min. The
20 Diesel
n-Tetradecane
exhaust temperature was measured by Omega K type thermo-
10 Decalin couple. For the test results, 95% confidence error bars were used to
0
highlight the test repeatability.
0.0 0.1 0.2 0.3 0.4
In this experiment, the engine speed was maintained at
Time after start of injection (ms)
1500 rpm. The directly injection pressure was fixed at 120 MPa.
(b) Spray penetration of diesel, n-tetradecane/diesel and decalin/diesel
50 There was no EGR strategy during this experiment. Before the tests
started, the engine was considered warm up once the coolant and
40 lubrication oil met 85 C. The intake temperature and pressure
were 15 C ± 1 and 1 atm (absolute pressure), respectively. The
Spary cone angle ( )
o
30 exhaust pressure was the ambient pressure. And the temperature

of coolant and lubrication oil were controlled at 80 C ± 1 and
20 85 C ± 1 by two circulation systems, respectively. All data were
Diesel measured under the stable conditions. The in-cylinder gas pressure
n-Tetradecane
10
Decalin is averaged from 200 consecutive cycles for each operating
condition.
0
10 20 30 40 50 60 70
Spray tip penetration (mm) 4. The effects of diesel surrogate candidates on engine
(c) Spray cone angle of diesel, n-tetradecane/diesel and decalin/diesel combustion and emission
Fig. 5. Spray properties of diesel and mixed fuels. 4.1. Constant injection timing vs. fixed ignition timing
Fig. 6 compares the effects of surrogate components with

3.2. Experimental system and methodology
different molecular structures blended with commercial diesel on
the in-cylinder pressure and heat release rate under fixed engine
The experiment was conducted on a single cylinder diesel en-
speed and load. In Fig. 6a, the direct injection timing is fixed at 7oCA
gine with a high-pressure common rail. The test system has been
BTDC, and in Fig. 6b, the ignition timing is fixed at 3oCA ATDC. For
introduced in Refs. [73,74], and the engine properties are listed in
fixed injection timing, the combustion process is mainly controlled
Table 3. As is known, it is difficult to study the effects of each
by the cetane number and physical properties (distillation curve or
parameter independently on engine operation; in particular, the
volatility). When the cetane number is higher, the ignition time is
distillation range, cetane number, molecular structures and other
earlier, but the maximum in-cylinder pressure, and the peak value
parameters are interconnected with each other and work together
of heat release were decreased. After blending 1-
on engine efficiency and emissions. Therefore, two kinds of control
methylnaphthalene, which lowers the cetane number of mixed
strategies were applied in the course of this experiment: load
fuel, the ignition timing was delayed, but the peak value of the heat
release rate was significantly higher than that of other fuels. In
Table 3 Fig. 6b, to weaken the effects of the cetane number on the com-
The engine properties. bustion phase, the ignition timing was fixed. For mixed fuel, the
order of T50 was cetane > diesel > HMN > 1-
Displacement 1.325 L Exhaust Valve open 337.5 ATDC
methylnaphthalene > decalin, so the burning rate was ce-
Stroke 130 mm Exhaust Valve Close 144.5 ATDC
tane < diesel < HMN < decalin<1-methylnaphthalene. Since the
Cylinder Diameter 114 mm Inlet Valve Open 112.5 ATDC
Connecting Rod Length 216 mm Inlet Valve Close 334.5 ATDC cetane number of 1-methylnaphthalene was the lowest, in order to
Compression Ratio 18:1 Nozzle Open 24.5 MPa ensure that the CA10 was the same, the injection time was the
Pressure earliest, which benefits the fuel-air mixing, so the peak values of
Common Rail Pressure 120 MPa Nozzle Hole 7 0.176 mm the heat release rate and the maximum pressure were the highest.
Number and diameter
Fig. 7 compares several gas emissions of diesel with those of
350 350
Diesel Diesel
n-Hexadecane n-Hexadecane
80 80
HMN 300 HMN 300
Decalin Decalin
1-Methylnaphthalene 1-Methylnaphthalene
Heat release rate (J/ CA)

250 250

60 60
o
Pressure (bar)
Pressure (bar)
o
200 200
40 150 40 150
100 100
20 20
50 50
0 0 0
0
-5 0 5 10 15 -5 0 5 10 15
o
Crank Angle ( CA) Crank Angle ( CA)
o
(a) Fixed fuel injection timing (b) Fixed ignition timing (CA10)
Fig. 6. The in-cylinder pressure and heat release rate of mixed fuels (IMEP ¼ 6 bar, 1500r/min).
mixed fuels, in which Fig 7a shows gas emissions under fixed in- NOx emission increased with the load increase. But there were
jection timing and Fig. 7b illustrates the gas emissions under fixed certain differences in NOx emissions when fueled with different
ignition timing. Under different control strategies, the CO emissions mixed fuels.
decreased rapidly with the increase of loads. For HC emissions, It is worth noting that comparing the emissions in Fig. 7a and b,
whether it is fixed injection timing or fixed ignition timing, the HC it can be found that for the same fuel under the same load, gas
emissions do not change much in the whole load range except in emissions were roughly the same except 1-methylnaphthalene/
the case of 1-methylnaphthalene/diesel. When fueled with 1- diesel. Moreover, the HC emissions and CO emissions of 1-
methylnaphthalene/diesel, the HC emissions were relatively high methylnaphthalene/diesel were significantly higher than other
at low loads and decreased rapidly with the increase of loads. The fuels, especially under low loads. NOx emissions of 1-
1600 800 1000

Diesel Diesel
Diesel
n-Hexadecane n-Hexadecane
n-Hexadecane
HMN 800 HMN
HMN
1200 Decalin 600 Decalin
Decalin
1-Methylnaphthalene
CO (ppm)
NOx (ppm)
HC (ppm)
600
800 400
400
400 200
200
0 0 0
3 4 5 6 7 8 9 3 4 5 6 7 8 9 3 4 5 6 7 8 9
IMEP (bar) IMEP (bar) IMEP (bar)
(a) Constant fuel injection timing

1600 800 1000
Diesel Diesel
Diesel n-Hexadecane
n-Hexadecane
n-Hexadecane HMN
HMN 800
HMN Decalin
1200 Decalin 600
Decalin 1-Methylnaphthalene
1-Methylnaphthalene
1-Methylnaphthalene
CO (ppm)
NOx (ppm)
HC (ppm)
600
800 400
400
400 200
200
0 0 0
3 4 5 6 7 8 9 3 4 5 6 7 8 9 3 4 5 6 7 8 9
IMEP (bar) IMEP (bar) IMEP (bar)
(b) Constant ignition timing (CA10)

Fig. 7. The gas emissions of mixed fuels under different loads (1500 rpm).
1600 800 55
n-Decane n-Decane iso-Octane n-Decane
IMEP =6bar IMEP=6bar
n-Dodecane n-Dodecane iso-Dodecane n-Dodecane
n-Tetradecane n-Tetradecane HMN n-Tetradecane
700
n-Hexadecane n-Hexadecane Decalin n-Hexadecane
iso-Octane MCH iso-Dodecane
iso-Dodecane Toluene iso-Octane
1500 HMN 600 50
n-Butylbenzene HMN
Tetralin Decalin
NOx (ppm)
Tmax ( C)
MCH
ITE (%)
1-Methylnaphthalene
500
o
Decalin
1400 400 45
MCH
Toluene Toluene
n-Butylbenzene Diesel
n-Butylbenzene
Tetralin 300
Tetralin
Diesel
1300 200 40
6.0 6.5 o
7.0 7.5 8.0 6.0 6.5 7.0 7.5 8.0 6.0 6.5 7.0 7.5 8.0
CA50 ( CA ATDC) o o
CA50 ( CA ATDC) CA50 ( CA ATDC)
(a) Constant fuel injection timing

1600 800 55
n-Decane n-Decane iso-Dodecane n-Decane
IMEP =6bar IMEP=6bar
n-Dodecane n-Dodecane iso-Octane n-Dodecane
n-Tetradecane n-Tetradecane HMN n-Tetradecane
700
n-Hexadecane n-Hexadecane Decalin n-Hexadecane
iso-Dodecane MCH iso-Dodecane
iso-Octane Toluene iso-Octane
1500 HMN 600 50
Tetralin HMN
n-Butylbenzene Decalin
NOx (ppm)
Tmax ( C)
MCH
ITE (%)
1-Methylnaphthalene
500
o
Decalin Diesel
1400 400 Diesel 45
MCH
Toluene Toluene
Diesel
Tetralin Tetralin
n-Butylbenzene 300
n-Butylbenzene
Diesel
1300 200 40
5 6 7 8 5.0 5.5 6.0 6.5 7.0 7.5 8.0 5.0 5.5 6.0 6.5 7.0 7.5 8.0
o
CA50 ( CA ATDC) o o
CA50 ( CA ATDC) CA50 ( CA ATDC)
(b) Constant ignition timing (CA10)

Fig. 8. The relationship of combustion temperature, NOx and indicated thermal efficiency to CA50 (1500 rpm, 6 bar).
methylnaphthalene/diesel were obviously higher than diesel and ATDC. Only the CA50 of n-decane was earlier, and the CA50 of
other fuels in the experimental conditions. tetralin and 1-methylnaphthalene were slightly later, so the ther-
CA50 is a key factor that reflects the efficiency of the engine. mal efficiency did not change significantly with CA50. In the case of
Fig. 8 shows the relationship between CA50 and the combustion fixed CA10, a significant relationship was found between CA50 and
parameter including maximum in-cylinder average temperature, ITE. When CA50 was about 7 oCA ATDC, the ITE was maximal. CA50
NOx emissions and indicated thermal efficiency under fixed in- that was too early or too late would result in a decrease in the
jection timing (in Fig. 8a) and fixed ignition timing (in Fig. 8b). For indicated thermal efficiency.
maximum in-cylinder average temperature, with the high pro- From the previous analysis, by comparing the in-cylinder pres-
portion of surrogate components blended with diesel, the sure, heat release rate, gas emissions and key combustion param-
maximum in-cylinder average temperature would be significantly eters between fixed injection timing and fixed ignition timing, it
higher than commercial diesel. can be concluded that the effects of cetane number on the com-
When the injection timing is fixed, CA50 for diesel blended with bustion process can be reduced to a certain extent by fixing the
different surrogate components was concentrated between 6.5 and ignition timing, but the influence of the physical characteristics of
7.5oCA ATDC. For constant ignition timing, the distribution range of the fuel, especially the volatility, on the combustion process and the
CA50 was quite wide, especially for n-alkanes even between 6.5 generation of gas emissions can be strengthened. Therefore, the
and 8.0 oCA. This may indicate that the volatility of fuel had a sig- effects of cetane number, carbon chain length and distillation range
nificant influence on the combustion rate and CA50. When the on engine combustion and emissions were studied by using the
injection time is fixed, however, the CA50 of tetralin and 1- above research and analytical methods.
methylnaphthalene are relatively late, 0.5e1 oCA than other fuels.
The maximum in-cylinder average temperature is maintained at 4.2. The effects of distillation of fuels
the same level as methyl cyclohexane (MCH). For constant ignition
timing, the CA50 of 1-methylnaphthalene was the earliest; the It is difficult to characterize the volatility of diesel through a
maximum in-cylinder average temperature was the same as MCH. specific parameter, usually described as T10, T50 and T90. In order
In Fig. 8, it can be found that NOx emissions of aromatics were to investigate the effects of fuel volatility on combustion and
significantly higher than those of other mixed fuels under both emission characteristics, in this section, the relationship between
fixed injection timing and fixed ignition timing. This may be due to the distillation range parameters and the combustion parameters is
the earlier CA50 of aromatics/diesel, which leads to a longer high- discussed in detail under fixed ignition timing. As shown in Fig. 4,
temperature reaction time. In addition, when the ignition timing is the T10 of HMN/diesel and 1-methylnaphthalene/diesel was
fixed, NOx emissions decreased with the delay of CA50. There was similar to that of diesel fuel. When diesel is mixed with n-hex-
no obvious decline in maximum average temperature. So, this was adecane, the T10 increased, but when mixed with decalin, the T10
probably caused by shortening the duration of high-temperature decreased significantly. There was no obvious relationship between
combustion. the combustion parameters of the above fuels (reflected by the
For the indicated thermal efficiency (ITE), the CA50 of most maximum in-cylinder pressure and the peak values of the heat
mixed fuels for fixed injection timing was concentrated at 7 oCA release rate) and T10. In addition, T90 is more closely related to the
final heat release and soot emissions. For the reasons above, Fig. 9 diesel. That is, the law of T50 was not consistent with ordering by
compares the in-cylinder pressure and heat release rate of fuels the burning rate. In Fig. 9c, the T70 of several fuels are as follows:
with similar T30, T50 and T70. diesel (294 C)> iso-dodecane (258 C)> toluene (257 C)> n-tet-
In Fig. 9a, the order of T30 for mixed fuel from the highest to radecane (256 C)> decalin (253 C), but the maximum in-cylinder
lowest was as follows: diesel (243 C)> n-tetradecane (238 C)> pressure or peak value of the heat release rate is: toluene > iso-
HMN (229 C)> 1-methylnaphthalene (228 C). The trend was dodecane > decalin > n-tetradecane. So, T70 cannot reflect the
completely consistent with the maximum in-cylinder pressure and impact of fuel volatility on the burning rate. From the above anal-
maximum heat release rate. In Fig. 9b, the T50 of mixed fuels was ysis, it can be seen that T30 can not only reflect the physical
diesel (267 C)> n-butylbenzene (229 C)> n-dodecane (226 C)> properties of mixed fuels but also reflected the effects of fuel
iso-octane (225 C)> tetralin (222 C). The maximum in-cylinder volatility on the combustion rate, which was mainly characterized
pressure and the peak value of the heat release rate were as fol- as the maximum in-cylinder pressure and peak value of the heat
lows: the tetralin > n-butylbenzene > isooctane > n-dodecane ¼ release rate.
Fig. 10 shows the relationship between combustion parame-
ters and T30 under fixed ignition timing. As discussed above, the
effects of the cetane number on the combustion and emission
characteristics had been weakened under fixed ignition timing. In
350
Fig. 10a, the CA50 for all fuels except 1-methylnaphthalene was
Diesel
80
n-Tetradecane 300 significantly delayed with increasing T30, whereas the CA50 of
HMN the n-alkanes was delayed by 0.5oCA relative to other fuels, which
1-Methylnaphthalene
IMEP=9bar
250 may be due to the slightly higher proportion of n-alkanes. In
60
o
Fig. 10b, the NOx emissions decreased significantly with the in-
Pressure (bar)
200
crease in T30. But the NOx emissions of aromatics were signifi-
40 150 cantly higher than those of other fuels, consistent with the results
in Fig. 8. This may be due to the higher pyrolysis temperatures of
100 aromatic hydrocarbons, which led to the faster heat release,
20
finally resulting in a higher in-cylinder gas temperature. Fig. 10c
50
shows the relationship between thermal efficiency and T30. It can
0 0 be seen that the highest point of thermal efficiency occurs at
-5 0 5 10 15 around 220 C (T30).
o
Crank Angle ( CA) Fig. 11 shows the relationship between combustion parameters
(CA50, NOx emissions and the thermal efficiency) and the T30 in
(a) Similar T30 the case of low load and heavy load. As can be seen in Fig. 11a,
Diesel
350 CA50 maintains at a stable value of about 6oCA ATDC under low
80
n-Dodecane loads. Under heavy loads, CA50 was delayed with the increase of
iso-Octane 300
n-Butylbenzene T30, except aromatic hydrocarbons. NOx emissions were reduced
Tetralin
250 with the increase of T30. Compared with other fuels, NOx emis-
60 IMEP=9bar
o
sions of aromatics were higher at high loads. The thermal effi-

Pressure (bar)
200 ciencies under heavy loads were not affected by T30. From this
40
analysis, T30 can be used as a key parameter of fuel volatility on
150
combustion, efficiency, and emissions in the full load range of the
100 engine.
20
50 4.3. The effects of cetane number
0 0
-5 0 5 10 15 Cetane number is an important factor affecting ignition timing,
o
Crank Angle ( CA) and it also has a strong influence on the maximum in-cylinder
(b) Similar T50 pressure, the peak value of the heat release rate. Based on the
350 research above, the main parameter that reflects the effects of fuel
n-Tetradecane volatility on the combustion process is T30. Therefore, in this sec-
80 iso-Dodecane
300
Decalin tion, mixed fuels with similar T30 were selected to study the effects
Methylbenzene
of cetane number on combustion under fixed injection timing. As
IMEP=9bar 250
mentioned above, when T30 is at about 220 C, the indicated ther-
o
60
Pressure (bar)
200 mal efficiency would be higher. To this end, fuels with T30 between
202 and 243 C were selected, including diesel (243 C), n-dodec-
40 150
ane/diesel (217 C), HMN/diesel (229 C), n-butylbenzene/diesel
100
(202 C), tetralin/diesel (211 C). And their cetane indexes were 52.8,
20 56.6, 50.4, 40 and 29.2, respectively. Fig. 12a compared the in-
50 cylinder pressure history and heat release rate. The maximum in-
cylinder pressure of tetralin was similar to that of n-butylbenzene,
0 0
-5 0 5 10 15 which was higher than that of other fuels. While the maximum in-
o
Crank Angle ( CA) cylinder pressure of diesel and n-dodecane were comparable. The
peak value of the heat release rate for mixed fuel was tetralin > n-
(c) Similar T70
butylbenzene > HMN > diesel > n-dodecane. Fig. 12b shows the
Fig. 9. In-cylinder pressure and heat release rate in similar T30, T50, T70 (1500 rpm, variation of CA10 for mixed fuels with different cetane numbers. As
9 bar, constant CA10). can be seen from the figure, when the cetane index was greater than
9
n-Decane IMEP =6bar
n-Dodecane
n-Tetradecane
8 n-Hexadecane
iso-Octane
iso-Dodecane
CA50 ( CA ATDC)
HMN Diesel
7
MCH
Decalin
o
Toluene
5
n-Butylbenzene
Tetralin
1-Methylnaphthalene
4
140 160 180 200 220 240 260
o
T30 ( C)
(a) CA50
800
n-Decane MCH
n-Dodecane Decalin
700 n-Tetradecane Toluene
n-Hexadecane n-Butylbenzene
iso-Octane Tetralin
iso-Dodecane 1-Methylnaphthalene
600
HMN
NOx (ppm)
500
400 Diesel
300
200
140 160 180 200 220 240 260
o
T30 ( C)
(b) NOx
55
n-Decane
n-Dodecane IMEP=6bar
n-Tetradecane
n-Hexadecane
iso-Octane
iso-Dodecane
50
HMN
MCH
Decalin
ITE (%)
Diesel
45
Toluene
n-Butylbenzene
Tetralin
1-Methylnaphthalene
40
140 160 180 200 220 240 260
o
T30 ( C)
(c) Indicated thermal efficiency

Fig. 10. CA50, NOx, indicated thermal efficiency changing with T30 under medium load (1500 rpm, 6 bar, constant CA10).
12 50, their impact on the ignition delay was very small. In fact, the
n-Decane MCH IMEP=3bar
n-Dodecane
n-Tetradecane
Decalin
IMEP=9bar ignition delay is mainly determined by the molecular structure and
n-Hexadecane
10 in-cylinder temperature/pressure conditions; even if the two fuels
have the same cetane number, low temperature reactions would not
CA50 ( CA ATDC)
8 Diesel 9bar
be the same, which may lead to differences in ignition delay [75].
When fuel reactivity is strong, the difference in ignition delay
6 resulting from the cetane number would be very small.
o
Diesel 3bar
Fig. 13 shows the effects of the cetane index on the maximum in-
4
iso-Octane
Toluene
n-Butylbenzene
cylinder pressure and maximum in-cylinder temperature. In
iso-Dodecane
HMN
Tetralin
1-Methylnaphthalene
Fig. 13a, under low loads, when the cetane index is less than 50, the
2
140 160 180 200 220 240 260 maximum in-cylinder pressure increased with the increase of the
o
T30 ( C) cetane index. With a cetane index increase higher than 50, the
maximum in-cylinder pressure decreased. Under medium and
(a) CA50 heavy loads, with an increase in the cetane index, the maximum in-
1000 n-Decane
n-Dodecane
Toluene
n-Butylbenzene
iso-Octane
iso-Dodecane
cylinder pressure decreased gradually. When the cetane index was
n-Tetradecane
n-Hexadecane
Tetralin
1-Methylnaphthalene
HMN
IMEP=3bar
about 55, the maximum in-cylinder pressures were close to the
800 MCH
Decalin IMEP=9bar diesel oil level. In Fig. 13b, the cetane index has no significant effects
Diesel 9bar on the peak values of in-cylinder temperature, which were mainly
600
NOx (ppm)
determined by the loads.

400
Fig. 14 shows the changes in the maximum pressure rise rate
and the maximum heat release rate with the increase in the cetane
200
Diesel 3bar
index. Under low loads, the maximum pressure rise rate increased
at first and then decreased. Under medium and high loads, the
0 maximum pressure rise rate decreased with the increase in the
140 160 180 200 220 240 260
T30 ( C)
o cetane index. Under low loads, the maximum heat release rate was
(b) NOx maintained at a stable level. Under medium and high loads, the
50 maximum heat release rate decreased with the increase in the
IMEP=3bar cetane index. When the cetane index was about 55, the maximum
IMEP=9bar
Diesel 9bar heat release rate of mixed fuels was close to those of diesel under
multiple loads.
45
ITE (%)
4.4. The effects of n-alkanes carbon chain length

40 Diesel 3bar
n-Decane
n-Dodecane
Toluene
The carbon chain length of n-alkanes has strong effects on the
n-Tetradecane
n-Hexadecane
MCH
n-Butylbenzene
Tetralin
iso-Octane
iso-Dodecane cetane number. On the other hand, the chain length of n-alkanes
1-Methylnaphthalene HMN
35
Decalin
also has an impact on fuel density and distillation range. In general,
140 160 180 200 220 240 260
o
with the increase of carbon chain length, the cetane number and
T30 ( C)
boiling point increase, but volatility deteriorates. The effects of the
(c) Indicated thermal efficiency
chain length (C10eC16) on combustion and emissions were dis-
Fig. 11. CA50, NOx, indicated thermal efficiency changing with T30 under low and high cussed in Fig. 15. As can be seen in Fig. 15a, there was no obvious
loads (1500 rpm, constant CA10). change in ignition timing for diesel blended with n-alkanes that
have different chain lengths. In fact, the CA10 for n-alkanes/diesel is
400
Diesel 10
80 n-Decane n-Decane Tetralin IMEP=3bar
HMN HMN IMEP=9bar
n-Butylbenzene n-Butylbenzene
Heat Release Rate(J/ CA)
300 8
Tetralin
o
60
CA10 ( CA ATDC)
Pressure (bar)
6
200
40
o
4
Diesel 9bar
100 Diesel 3bar

20 2
0 0
0 25 30 35 40 45 50 55
-5 0 5 10 15
o Cetane index
Crank Angle ( CA)
(a) In-cylinder pressure and heat release rate (IMEP=6bar) (b) Ignition timing
Fig. 12. The effects of cetane index on in-cylinder pressure, heat release rate and ignition timing (1500 rpm).
n-Decane Tetralin IMEP=3bar

2000 HMN IMEP=6bar
HMN IMEP=6bar
90 n-Butylbenzene IMEP=9bar
n-Butylbenzene IMEP=9bar
1800
Diesel 9bar
Diesel 9bar
Tmax ( C)
Pmax (bar)
80 1600
o
Diesel 6bar
1400 Diesel 6bar
70
Diesel 3bar 1200
Diesel 3bar
1000
60 0 25 30 35 40 45 50 55
25 30 35 40 45 50 55
Cetane index
Cetane index
(a)The maximum in-cylinder pressure (b)The peak value of in-cylinder average temperature
Fig. 13. The effects of cetane index on maximum in-cylinder pressure and peak value of in-cylinder average temperature (1500 rpm).
20 270
n-Decane Tetralin IMEP=3bar n-Decane Tetralin IMEP=3bar
HMN IMEP=6bar HMN IMEP=6bar
n-Butylbenzene IMEP=9bar n-Butylbenzene IMEP=9bar
PRRmax (bar/ CA)
15 210
HRRmax (J/ CA)
o
Diesel 6bar Diesel 9bar

10 150
Diesel 3bar Diesel 6bar
Diesel 9bar
Diesel 3bar
5 90
25 30 35 40 45 50 55 25 30 35 40 45 50 55
Cetane index Cetane index
(a) Maximum pressure rise rate (b) Peak value of heat release rate
Fig. 14. The effects of cetane index on peak pressure rise rate and peak heat release rate (1500 rpm).
350
Diesel
80
n-Decane 300
Heat Release Rate (J/ CA)
n-Dodecane
n-Tetradecane
250
o
n-Hexadecane
60
Pressure (bar)
200
40 150
100
20
50
0 0
-5 0 5 10 15
o
Crank Angle ( CA)
(a) In-cylinder pressure and heat release rate

(b) Ignition delay [76]
(IMEP=6bar, constant injection timing)
Fig. 15. The effects of n-alkanes carbon chain length on in-cylinder pressure, heat release rate and ignition timing (1500 rpm).
2.5oCA ATDC, and the CA10 for diesel is 3.0 oCA ATDC. But there was volatility. As can be seen in Table 2, the key parameters of the
a slight difference in the peak values of the heat release rate, which distillation curves were that for T10, n-decane/diesel (175 C) « n-
is mainly due to the difference in physical properties of n-alkanes of dodecane/diesel (210 C) < diesel (214 C) < n-tetradecane/diesel
different chain lengths. In fact, basic studies on the combustion (222 C) < n-hexadecane/diesel (225 C) and for T30, n-decane/
parameters (shock tube and chemical kinetics calculations) have diesel (182 C) « n-dodecane/diesel (217 C) < n-tetradecane/diesel
shown that the ignition delays are approximately the same for n- (238 C) < diesel (243) < n-hexadecane/diesel (264 C). But for HC
alkanes from C7 to C16 over a wide temperature range, as shown in emissions in the full load, n-hexadecane < n-tetradecane < n-
Fig. 15b [76]. dodecane < n-decane ¼ diesel, and the difference in HC emissions
However, as can be seen in Fig. 16, the carbon chain length had a between different mixed fuels was lower than 50 ppm. For NOx
certain influence on CO emissions. Under low loads (<5 bar), the CO emissions, diesel oil and the mixture of n-decane/diesel were
emissions of n-decane/diesel were similar to those of diesel oil. And slightly higher, while n-hexadecane was slightly lower. And their
the CO emissions of n-dodecane/diesel, n-tetradecane/diesel and n- tendencies changed with loads that were similar to each other.
hexadecane/diesel were lower than those of diesel fuel. Under high
loads (>7 bar), diesel had the lowest CO emissions, and the CO
emissions of n-hexadecane and n-tetradecane were higher. For
direct injection engines, the HC emissions are relatively low at
about 50e100 ppm and were mainly determined by the fuel 4.5. The effects of H/C ratio
The fuel unsaturation directly or indirectly determines the

500 physical (density, distillation range) and chemical properties (ce-
Diesel tane number, ignition delay) of the fuel. And the fuel unsaturation
n-Decane
400 n-Dodecane
also has a significant impact on the combustion emissions. In order
n-Tetradecane to discuss the effects of fuel unsaturation on combustion and
n-Hexadecane emission properties, surrogate components with different H/C ra-
CO (ppm)
300
tios were selected to blended with diesel. Their combustion and
emission characteristics were studied.
200
The H/C ratios of chosen components were n-decane (2.2),
decalin (1.8), n-butylbenzene (1.4), tetralin (1.2) and 1-
100 methylnaphthalene (0.91). Fig. 17 shows the in-cylinder pressure
and heat release rate for fuels blended with surrogate components
0 with different H/C ratios at IMEP 3, 6 and 9 bar. It can be seen that at
3 4 5 6 7 8 9
low loads, the H/C ratio has limited effects on the in-cylinder
IMEP (bar)
pressure and heat release rate. The difference in peak values was
(a) CO within 5 bars, and the difference in peak values of the heat release
300 rate was within about 10 J/oCA. However, under medium and high
Diesel loads, with the increase of H/C ratio, the peak values of in-cylinder
n-Decane pressure and the peak values of the heat release rate increased
n-Dodecane
gradually. In particular, for tetralin/diesel, it was about 10 bar and
n-Tetradecane
200
n-Hexadecane 40 J/oCA higher than those of n-decane/diesel in peak values of in-
HC (ppm)
cylinder pressure and heat release rate, respectively. This indicated

that the greater the degree of unsaturation, the faster the com-
bustion rate under medium and high loads.
100
Fig. 18 shows the relationship between gas emissions and the H/
C ratio. The CO emissions decreased with the increase in the H/C
ratio in the whole load, especially under low loads. When the H/C
0 ratio reached about 2.0, CO emissions were close to actual diesel
3 4 5 6 7 8 9
emission levels (250e300 ppm). In Fig. 18b, as the H/C ratio varies,
IMEP (bar)
the changes in HC emissions were similar to those of CO emissions
(b) HC
1000
(100 ppm). But when the H/C ratio was about 1.8, the HC emissions
Diesel of all mixed fuel reached the minimum value and recovered to the
n-Decane value of diesel fuel HC emissions when the H/C ratio was about 2.0.
800 n-Dodecane
n-Tetradecane When the H/C ratio was about 1.8, the lower reactivity of the sur-
n-Hexadecane rogate components led to longer ignition delay and maintained a
NOx (ppm)
600
higher maximum in-cylinder heat release rate as shown in Fig. 17b
and c. Both of the above two factors were conducive to the reduc-
400 tion of HC emissions and resulted in the lowest HC emissions when
the H/C ratio was about 1.8. In Fig. 18c, NOx emissions for diesel
200 were plotted in dashed lines under different loads. Compared with
medium and high loads, the difference in NOx emissions between
0 mixed fuels and diesel were smaller under low loads. When under
3 4 5 6 7 8 9
medium and high loads, NOx emissions declined with the H/C ratio
IMEP (bar)
increasing and exhibited some difference with diesel oil. The above
(c) NOx
analysis showed that the H/C ratio had a strict influence on CO and
Fig. 16. The effects of n-alkane carbon chain length on gas emissions (1500 rpm, HC emissions. But the relationship with NOx emissions was
IMEP ¼ 6 bar, constant injection timing). uncertain.
90 90 90
80 80 80
Pressure (bar)
Pressure (bar)
Pressure (bar)
70 70 70
60 60 60
Diesel
50 50 Diesel 50 n-Decane
n-Decane
Decalin
40 40 Decalin 40 n-Butylbenzene
Diesel n-Butylbenzene
Tetralin
n-Decane Tetralin
200 200 200
Decalin


n-Butylbenzene
160 160 160
o
o
Tetralin
o
120 120 120
80 80 80
40 40 40
0 0 0
-5 0 5 10 15 -5 0 5 10 15 -5 0 5 10 15
o o o
Crank Angle ( CA ATDC) Crank Angle ( CA ATDC) Crank Angle ( CA ATDC)
(a) IMEP=3bar (b) IMEP=6bar (c) IMEP=9bar
Fig. 17. Combustion characteristics with mixed fuels which blended with candidates of different H/C ratios (1500 rpm, constant CA10).
5. Diesel surrogate formulation

X
ST ¼ xi STi (13)
The purpose of diesel surrogate fuel is to simulate the key
properties of target commercial diesel fuel. Based on the above X
analysis, cetane number, H/C ratio, density, viscosity, surface ten- CN ¼ CNi vi (14)
sion and distillation range are the key properties of diesel fuel.
Fig. 19 shows the relationship of the number of components and Combined with the volume conservation equation,
distillation range. As shown in Fig. 19, the more components in the X
surrogate fuel, the more precisely the distillation range can be vi ¼ 1 (15)
simulated. In addition, the relatively even distribution of the boiling
point of the selected components in the distillation range would In the formula, vi is the volume fraction of number i
help to reproduce the diesel distillation curve. On the other hand, hydrocarbon;
considering the complexity of the combustion mechanism when
surrogate fuel has too many components and the limitation of ri is the density of number i hydrocarbon;
computing resources when surrogate fuel is applied to engine mi is the number of hydrogen atoms in number i hydrocarbon;
performance calculation, a 3e7 component surrogate fuel would be ni is the number of carbon atoms of number i hydrocarbon;
suitable. xi is the mole fraction of number i hydrocarbon;
As shown in Fig. 12b, when cetane index was higher than a STi is the surface tension of number i hydrocarbon;
certain value (50 in this article), there would be no obvious changes CNi is the cetane number of number i hydrocarbon.
in CA10 with the increase in the fuel cetane number. According to
Fig. 20, the author considered that cetane index of diesel surrogate It is worth noting that previous studies have shown that aro-
fuel should be within a certain range, about 52 C 54 (original matics have significant effects on NOx emissions. And as studied by
diesel was about 53). In addition, with the strong effect of degree of other researchers, the high proportion of aromatics would result in
unsaturation on HC and CO emission, the H/C ratio was between 1.8 higher soot emissions [3]. Therefore, the proportion of aromatics is
and 2.2 in Fig. 21 to ensure that the two gas emissions were similar defined in the relevant fuel standard, which needs to be used as a
to diesel fuel. The other properties such as density and surface constraint in the surrogate fuel formulation. Last but not least, in a
tension can be calculated by the following formula: view of the uncertainty in components and physicochemical
properties of the actual diesel fuel, the final proportion of candidate
X components in diesel surrogate should be a range rather than a
r¼ ri vi (11) fixed value which means that the construction of surrogate fuel has
only reasonable solution, there is no optimal solution.
Based on the studies above, the distillation range of diesel is
P extremely important for fuel atomization and subsequent com-
mx
H=C ¼ P i i (12) bustion. For the building of a 3-Components surrogate, the pref-
ni xi
erable boiling points of the compositions are between diesel T20
and T80 (about 220 Ce290 C in Fig. 19) to avoid the excessive
1200 8
Decalin 1-Methylnaphthalene IMEP=6bar
900 6
CA10 ( CA ATDC)
CO (ppm)
600 4 n-Decane
o
Diesel
HMN
n-Butylbenzene
300
2 Tetralin
0
0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 0
25 30 35 40 45 50 55
H/C ratio
Cetane index
(a) CO
400
Fig. 20. The relationship of ignition timing and cetane number.
300
HC (ppm)
200
400
100 300
HC (ppm)
200
0
0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2
H/C ratio 100
(b) HC
800
n-Decane Tetralin IMEP=3bar 0 1000
800
600
CO (ppm)
Diesel 9bar
n-Decane 600
Decalin
NOx (ppm)
400 n-Butylbenzene 400

Diesel 6bar
Tetralin
1-Methylnaphthalene 200
200
Diesel 3bar 0
0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2
H/C ratio
0
0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2
H/C ratio
(c) NOx Fig. 21. Relationship between H/C ratio and CO, HC emissions.
Fig. 18. Effects of H/C ratio on gas emissions (1500 rpm, constant CA10).
100 100
v1
Percent Recovery (% Volume)
80 80
Volume Percentage(%)
v2
60 60
40 v3 40
20 20
v4
0 0
160 180 200 220 240 260 280 300 320 340 360
o
T( C)
Fig. 19. The relationship of the number of components and distillation range.
Table 4 cycloparaffin and aromatics, respectively. According to the

Compositions of multicomponent diesel surrogate fuels. constraint equations from Equation (11) to Equation (15), the
Diesel surrogates Components Mole fraction % compositions of multicomponent surrogate fuels are listed in
3-Components surrogate fuel n-Hexadecane 41.3
Table 4.
HMN 36.8
1-Methylnaphthalene 21.9
5-Components surrogate fuel n-Hexadecane 21.6 6. The verification of diesel surrogates based on real engine
HMN 26.0 combustion and emission
1-Methylnaphthalene 20.7
Decalin 16.2
n-Octadecane 15.5
Fig. 22 compared the in-cylinder pressure and heat release rate
of the multicomponent surrogate fuels with diesel oil under low,
7-Components surrogate fuel n-Hexadecane 21.5
medium and high loads under fixed injection timing 7oCA BTDC. It
HMN 25.8
1-Methylnaphthalene 13.7 can be seen that the pressure and heat release rate curves of sur-
Decalin 8.1 rogate fuels are roughly identical to those of diesel oil under the
n-Octadecane 15.4 same operating conditions, which means the proposed diesel sur-
n-Butylbenzene 8.1 rogate fuels successfully reproduce the diesel combustion phase
n-Butylcyclohexane 7.4
and the combustion rate.
Fig. 23 presents a comparison of the gas emissions of multi-
component surrogate fuels and commercial diesel oil under various
light or heavy components in the surrogate. In addition, the sur- loads. In the figure, the gas emissions of diesel oil under each load
rogate fuel should include, as far as possible, the major carbon were normalized to 1, and the gas emissions of surrogate fuels were
species in diesel. So, n-hexadecane (boiling point 287 C), HMN expressed as the ratio of surrogate fuel emissions to diesel oil
(boiling point 240 C) and 1-methylnaphthalene (boiling point emissions. As can be seen in Fig. 23, the NOx emissions of surrogate
240 C) were chosen as the components of 3-Components surro- fuel were roughly identical to diesel oil. But the CO emissions of
gate fuel. Cycloparaffin is one type of the important components of surrogate fuels were about 5e25% lower than those of diesel fuel in
diesel, and for this purpose, decalin was chosen to represent most operating conditions. To comprehensively assess the emis-
cycloparaffin in the 5-Components surrogate fuel. For the overall sions of surrogate fuels under a whole load range, the parameters
consideration of components with large molecules in commercial were defined according to the following formula:
diesel and the difference in distillation curves caused by decalin

added in surrogate fuel, n-octadecane was chosen as another n- X9
COsur;i
alkane in the 5-Components surrogate fuel. So, n-hexadecane, n- DCO ¼ 1 (16)
COdies;i
octadecane, HMN, 1-methylnaphthalene and decalin were chosen i¼3
as the components of the 5-Components surrogate fuel. Further-

more, in order to meet the distribution of carbon atoms of cyclo- X9
HCsur;i
paraffin and aromatics in the commercial diesel oil, 7-Components DHC ¼ 1 (17)
HCdies;i
surrogate fuel was proposed in which n-butylcyclohexane, decalin i¼3
and n-butylbenzene, 1-methylnaphthalene were selected as
90 90 90
80 80 80
IMEP=3bar IMEP=9bar
70 IMEP=6bar
Pressure (bar)
Pressure (bar)
Pressure (bar)
70 70
60 60 60
50 50 50
40 40 Diesel 40 Diesel
Diesel 3-Components 3-Components
3-Components 5-Components 5-Components
200 200 200
5-Components 7-Components 7-Components
160 7-Components 160 160

o
o
o
120 120 120
80 80 80
40 40 40
0 0 0
-5 0 5 10 15 -5 0 5 10 15 -5 0 5 10 15
o o o
Crank Angle ( CA) Crank Angle ( CA) Crank Angle ( CA)
(a) IMEP=3bar (b) IMEP=6bar (c) IMEP=9bar
Fig. 22. In-cylinder pressure and heat release rate of multicomponent surrogate fuels and diesel (1500 rpm).
1.6 Table 5
3-Components Accumulation deviations of emissions for different surrogate fuels.
5-Components
7-Components DCO DHC DNOx
1.2
3-Components surrogate fuel 0.811 0.514 0.140
0.8
CO
0.4
surrogate fuels, and the deviation values of NOx emissions were the
lowest. Through the above comparison, it can be found that the gas
0.0 emissions of 5-Components surrogate fuel were the nearest to
3 4 5 6 7 8 9
those of diesel.
IMEP (bar)
In this study, the smoke emissions are nearly zero under
most loads due to the application of high fuel injection pres-
(a) CO sure. But the particulate emissions still exist. Therefore, the
1.6
3-Components particle size distributions of surrogate fuel and diesel fuel were
5-Components compared as shown in Fig. 24. The mechanism of particle
7-Components
1.2 generation was quite complex [77]. Many factors affect the
generation and evolution of particulate matter including the
molecular structure of fuel, molecular oxygen content, vola-
0.8 tility, combustion mode, fuel injection strategy, operating con-
HC
ditions (speed and load), and equivalence ratio [78,79]. In

Fig. 24, under low loads, the particle size distributions were
0.4 unimodal mode, and the peak particle size was about 10 nm.
Under medium loads, the particle size distribution changed to
bimodal mode, and the corresponding peak particle sizes were
0.0
3 4 5 6 7 8 9
about 10 nm and 50e60 nm, which corresponded to nucleation
IMEP (bar) mode and accumulation mode, respectively. Under high loads,
(b) HC the curves reverted back to unimodal mode. The proportion of
1.6 accumulation mode particles increased, and the peak particle
3-Components
5-Components
size moved to 40 nm. For multicomponent surrogate fuels, the
7-Components peak values of particle size distributions were obviously higher
1.2
than those of diesel under low loads, especially 3-Components
surrogate fuel and 5-Components surrogate fuel. But under
medium loads, the peak values of particle size distributions of
0.8
NOx
surrogate fuels declined dramatically, and only about 30% of

that of diesel fuel. Under heavy loads, the particle size distri-
butions of multicomponent surrogate fuels were similar to
0.4
those of diesel oil. The small particles were very sensitive to
fuel evaporation. Under the low load of the engine, the tem-
0.0
perature of the piston and the cylinder wall was slightly lower.
3 4 5 6 7 8 9
During fuel injection progress, part of the fuel may hit the
IMEP (bar)
piston or the cylinder wall. Compared to diesel, the higher
(c) NOx boiling point of the components in surrogate fuel would
weaken the fuel gasification, which may lead to higher particle
Fig. 23. Comparison of gas emissions of multicomponent surrogate fuels and diesel oil
(1500 rpm).
emissions. With the increase of engine loads, the temperature
of the piston and cylinder wall would be higher, which may
benefit the progress of fuel gasification and weaken the effects
of the boiling point of the components on the particle emis-
sions. Then, the difference in particle size distribution between
X9
NOxsur;i
DNOx ¼ 1 (18) surrogate fuel and diesel would decrease.
NOxdies;i
i¼3
In the formula, DCO ; DHC ; DNOx are the cumulative deviation of 7. Conclusion
the CO, HC, NOx emissions of surrogate fuels in the full load range,
respectively; COsur;i ; HCsur;i ; NOxsur;i represent the CO, HC and NOx Due to the complexity and uncertainty of components in com-
emissions of surrogate fuels in a certain IMEP; COdies;i , mercial diesel oil and the existence of a certain proportion of
HCdies;i ; NOxdies;i are defined as the CO, HC and NOx emissions of components with lower saturated vapor pressure, for which it is
diesel oil in a certain IMEP; i is the indicated mean effective pres- difficult to perform studies of typical gas-phase combustion ex-
sure (IMEP). periments, the formulation of diesel surrogate fuel is limited by the
According to the above definition, the accumulation deviations choice of surrogate components and the lack of kinetic parameters
of emissions for different surrogate fuels are listed in Table 5. It can of components with carbon atoms above 12. Based on detailed
be seen that among the three kind of deviation values, the deviation component analysis for China 0# diesel oil, in this paper, 13
values of CO emissions were the highest for whatever different candidate components have been selected to represent n-alkanes,
8
1.5x10
Diesel
3-Components
5-Components
8 7-Components
1.0x10
dN/dlogDp (#/cm )
-3
7
5.0x10
0.0
5 50 500
Particle Diameter (nm)
(a) IMEP=3bar
7
4x10
Diesel
3-Components
5-Components
7 7-Components
3x10
dN/dlogDp (#/cm )
-3
7
2x10
7
1x10
0
5 50 500
(b) IMEP=6bar
8
1.0x10
Diesel
3-Components
7
8.0x10 5-Components
7-Components
dN/dlogDp (#/cm )
-3
7
6.0x10
7
4.0x10
7
2.0x10
0.0
5 50 500
(c) IMEP=9bar
Fig. 24. Particle size distributions of multicomponent surrogate fuels and diesel oil (1500 rpm).
iso-alkanes, cycloalkanes and aromatics components. (5) There was no obvious difference in the ignition time for n-
Traditional methodologies for the formulation of surrogate alkanes selected in this paper (C10 ~ C16). But n-alkanes had
fuel were based on the accurate simulation of oil key parameters a certain influence on gas emissions because volatility
including average molecular weight, density, viscosity, cetane changes with an increase in the carbon chain length.
number/octane number and HC ratio. Then, the appropriate (6) Fuel molecular unsaturation has a significant impact on the
components were chosen to formulate diesel surrogate fuel in an engine combustion process and emissions. As molecular
attempt to achieve the target key combustion properties. Actu- saturation decreases, the combustion rate increases and the
ally, the physical and chemical properties of commercial oil are maximum in-cylinder pressure and peak value of the heat
usually complicated, and each property may affect the engine release rate increase. With respect to emissions, HC and CO
operation. But the target properties are much less than proper- emissions were significantly reduced (especially at low
ties affecting engine performance. Therefore, in this paper, a new loads) as the molecular saturation increased. When the H/C
methodology for diesel surrogate fuel formulation was proposed ratio was greater than 1.8, the CO and HC emissions under
that was not only based on key parameters of commercial oil but various loads were close to those of the original diesel en-
also based on the actual engine ignition, combustion and emis- gine. Molecular saturation also has a significant effect on NOx
sion properties. emissions, especially at medium and high loads; with a
Considering that many pure surrogate components cannot be decrease in the H/C ratio, the NOx emissions obviously
directly applied to compression ignition engines, a high proportion increased.
of surrogate components (50% n-alkanes and 30% other compo- (7) Based on the key physical and chemical parameters of the
nents) were added to the diesel fuel, making the surrogate commercial fuel and the influence of the surrogate compo-
component the dominant component in the blended fuel. First, the nents mixed with diesel on the combustion and emissions of
key physical and chemical parameters that had an impact on fuel the actual engine, the key physical and chemical parameters
injection, atomization, the ignition and combustion phase, effi- of the surrogate fuel are defined, and the constraint equa-
ciency and emissions were discussed in detail. Second, the com- tions were constructed for different parameters. Three kinds
bustion and emission characteristics of each mixed fuel were of diesel surrogate fuel were established including 3-
studied based on a single-cylinder engine. The influence of cetane Components surrogate fuel (41.3% n-hexadecane, 36.8%
number and the volatility of the fuel were studied by using the HMN, 21.9% 1-methylnaphthalene), 5-Components surrogate
same injection timing and the same ignition timing under various fuel (21.6% n-hexadecane, 15.5% n-octadecane, 26.0% HMN,
loads for each fuel. Then, the key physical and chemical parameters 20.7% 1-methylnaphthalene, 16.2% decalin) and 7-
of the surrogate fuel are defined, and the constraint equations are Components surrogate fuel (21.5% n-hexadecane, 15.4% n-
constructed for different parameters. Finally, based on the studies octadecane, 25.8% HMN, 13.7% 1-methylnaphthalene, 8.1%
above, three kinds of diesel surrogate fuel were established and decalin, 8.1% n-butylbenzene, 7.4% n-butylcyclohexane).
verified. Through this study, the following conclusions can be (8) The proposed multicomponent surrogate fuel can accurately
reached: simulate the engine combustion process. In the aspect of
emissions, NOx emissions of surrogate fuel were identical to
(1) The density, viscosity and surface tension of the fuel are the those of a commercial diesel engine. Under medium and high
key parameters that affect the spray characteristics of the loads, the CO emissions were lower than those of commercial
direct injection compression ignition engine. The distillation diesel oil. The gas emissions of 5-Components surrogate fuel
range (volatility) is the core parameter that determines the were the nearest to those of diesel.
mixing and combustion rate. The cetane number determines
the ignition delay. The degree of unsaturation is the domi- Acknowledgement
nant factor in leading CO and HC emissions. The above six
parameters are the key parameters in the diesel surrogate This work was supported by the Foundation for Distinguished
fuel construction process. Young Scholars of China (Grant No. 51425602) and the National
(2) For the same fuel, the load characteristics were studied with Basic Research Program of China (Grant No. 2013CB228405)
the control strategies, including the same injection timing
and the same ignition timing, to isolate the effects of cetane
References
number and volatility on engine combustion and emission
characteristics, respectively. Regardless of the injection [1] Lu XC, Han D, Huang Z. Fuel design and management for the control of
strategy, the NOx emissions of aromatics were significantly advanced compression-ignition combustion modes. Prog Energy Combust Sci
higher than those of other fuels. In addition, the indicated 2011;37:741e83.
[2] Guibet J-C, Faure-Birchem E. Fuels and engines: technology, energy, envi-
thermal efficiency reached the maximum when CA50 was
ronment. Editions Technip; 1999.
about 7.5 oCA ATDC. [3] Farrell JT, Cernansky NP, Dryer FL, Law CK, Friend DG, Hergart CA, et al.
(3) T30 can reflect the influence of the fuel distillation range Development of an experimental database and kinetic models for surrogate
diesel fuels. 2017. SAE Technical Paper 2007-01-0201.
(volatility) on the engine combustion process under various [4] William P, Charles M. Recent progress in the development of diesel surrogate
load conditions. CA50 is delayed with the increase of T30 fuels. Prog Energy Combust Sci 2011;37:330e50.
except in the case of aromatics/diesel. NOx emissions [5] Dryer FL. Chemical kinetic and combustion characteristics of transportation
fuels. Proc Combust Inst 2015;35:117e44.
decreased with the increase in T30 except in the case of ar- [6] Zhu LF, Qian Y, Wang XL, Lu XC. Effects of direct injection timing and pre-
omatics/diesel. There is a clear relationship between the mixed ratio on combustion and emissions characteristics of RCCI (Reactivity
indicated thermal efficiency and T30. In small and medium Controlled Compression Ignition) with N-heptane/gasoline-like fuels. Energy
2015;93:383e92.
loads, when T30 is between 220 and 230 C, the indicated
[7] Qian Y, Li H, Han D, Ji LB, Huang Z, Lu XC. Octane rating effects of direct in-
thermal efficiency is higher. jection fuels on dual fuel HCCI-DI stratified combustion mode with port in-
(4) When the cetane index was higher than a certain value, with jection of n-heptane. Energy 2016;111:1003e16.
a further increase in the cetane number, there was no [8] Qian Y, Wang XL, Zhu LF, Lu XC. Experimental studies on combustion and
emissions of RCCI (reactivity controlled compression ignition) with gasoline/
obvious change in ignition delay. n-heptane and ethanol/n-heptane as fuels. Energy 2015;88:584e94.
[9] Battin-Leclerc F. Detailed chemical kinetic models for the low-temperature
combustion of hydrocarbons with application to gasoline and diesel fuel [42] Zhu Y, Davidson D, Hanson R. Pyrolysis and oxidation of decalin at elevated
surrogates. Prog Energy Combust Sci 2008;34:440e98. pressures: a shock-tube study. Combust Flame 2014;161:371e83.
[10] Edwards T, Maurice LQ. Surrogate mixtures to represent complex aviation and [43] Li B, Zhang Y, Zhang H, Egolfopoulos FN. Extinction studies of non-premixed
rocket fuels. J Propuls Power 2001;17:461e6. iso-cetane and decalin flames. Proc Combust Inst 2015;35:965e72.
[11] Li H, Yu L, Lu XC, Ouyang LQ, Sun SZ, Huang Z. Autoignition of ternary blends for [44] Qian Y, Qiu Y, Zhang YH, Lu XC. Effects of different aromatics blended with
gasoline surrogate at wide temperature ranges and at elevated pressure: shock diesel on combustion and emission characteristics with a common rail diesel
tube measurements and detailed kinetic modeling. Fuel 2016;181:916e25. engine. Appl Therm Eng 2017;125:1530e8.
[12] Mueller CJ, Cannella WJ, Bruno TJ, Bunting B, Dettman HD, Franz JA, et al. [45] An YZ, Pei YQ, Qin J, Zhao H, Teng SP, Li B, et al. Development of a PAH
Methodology for formulating diesel surrogate fuels with accurate composi- (polycyclic aromatic hydrocarbon) formation model for gasoline surrogates
tional, ignition-quality, and volatility characteristics. Energy & Fuel 2012;26: and its application for GDI (gasoline direct injection) engine CFD (computa-
3284e303. tional fluid dynamics) simulation. Energy 2016;94:367e79.
[13] Mueller CJ, Cannella WJ, Bays JT, Bruno TJ, Defabio K, Dettman HD, et al. Diesel [46] Davis S, Wang H, Breinsky K, Law C. Laminar flame speeds and oxidation ki-
surrogate fuels for engine testing and chemical-kinetic modeling: composi- netics of benene-air and toluene-air flames. Symp Combust 1996;26:
tions and properties. Energy & Fuel 2016;30:1445e61. 1025e33.
[14] Colket M, Edwards T, Williams S, Cernansky NP, Miller DL, Egolfopoulos F, [47] Darcy D, Nakamura H, Tobin CJ, Mehl M, Metcalf WK, Pitz WJ, et al. A high-
et al. Development of an experimental database and kinetic models for sur- pressure rapid compression machine study of n-propylbenzene ignition.
rogate jet fuels. AIAA 2007-770. Combust Flame 2014;161:65e74.
[15] Lippert AM, Stanton DW, Reitz RD, Rutland CJ, Hallett WL. Investigating the [48] Darcy D, Mehl M, Simmie JM, Wu€rmel J, Metcalfe WK, Westbrook CK, et al. An
effect of spray targeting and impingement on diesel engine cold start. 2000. experimental and modeling study of the shock tube ignition of a mixture of n-
SAE Technical Paper 2000-01-0269. heptane and n-propylbenzene as a surrogate for a large alkyl benzene. Proc
[16] Pastor J, García-Oliver J, Nerva J-G, Gimenez B. Fuel effect on the liquid-phase Combust Inst 2013;34:411e8.
penetration of an evaporating spray under transient diesel-like conditions. [49] Nakamura H, Darcy D, Mehl M, Tobin CJ, Metcalfe WK, Pitz WJ, et al. An
Fuel 2011;90:3369e81. experimental and modeling study of shock tube and rapid compression ma-
[17] Kook S, Pickett LM. Liquid length and vapor penetration of conventional, chine ignition of n-butylbenzene/air mixtures. Combust Flame 2014;161:
Fischer-Tropsch, coal-derived, and surrogate fuel sprays at high-temperature 49e64.
and high-pressure ambient conditions. Fuel 2012;93:539e48. [50] Dagaut P, Ristori A, Frassoldati A, Faravelli T, Dayma G, Ranzi E. Experimental
[18] Lemoine F, Castanet G. Temperature and chemical composition of droplets by study of tetralin oxidation and kinetic modeling of its pyrolysis and oxidation.
optical measurement techniques: a state-of-the-art review. Exp Fluid Energy & Fuel 2013;27:1576e85.
2013;54:1e34. [51] Wang H, Gerken WJ, Wang W, Oehlschlaeger MA. Experimental study of the
[19] Basha SA, Gopal KR. In-cylinder fluid flow, turbulence and spray modelsda high-temperature autoignition of tetralin. Energy & Fuel 2013;27:5483e7.
review. Renew Sustain Energy Rev 2009;13:1620e7. [52] Mati K, Ristori A, Pengloan G, Dagaut P. Oxidation of 1-Methylnaphthalene at
[20] Myong KJ, Suzuki H, Senda J, Fujimoto H. Evaporation characteristics of multi- 1e13 atm: experimental study in a JSR and detailed chemical kinetic
component fuel. Fuel 2006;85:2632e9. modeling. Combust Sci Technol 2007;179:1261e85.
[21] Myong K-J, Suzuki H, Senda J, Fujimoto H. Spray inner structure of evaporating [53] Wang HW, Warner SJ, Oehlschlaeger MA, Bounaceur R, Biet J, Glaude PA, et al.
multi-component fuel. Fuel 2008;87:202e10. An experimental and kinetic modeling study of the autoignition of a-meth-
[22] Aye MM, Beeckmann J, Vanegas A, Peters N, Pitsch H. Experimental investi- ylnaphthalene/air and a-methylnaphthalene/n-decane/air mixtures at
gation of diesel and surrogate fuels: spray and ignition behavior. 2011. SAE elevated pressures. Combust Flame 2010;157:1976e88.
Technical Paper 2011-01-1921. [54] Ramirez HP, Hadj-Ali K, Dievart P, Moreac G, Dagaut P. Kinetics of oxidation of
[23] Zhang L, Kong S-C. Modeling of multi-component fuel vaporization and commercial and surrogate diesel fuels in a jet-stirred reactor: experimental
combustion for gasoline and diesel spray. Chem Eng Sci 2009;64:3688e96. and modeling studies. Energy & Fuel 2010;24:1668e76.
[24] Siebers DL. Liquid-phase fuel penetration in diesel sprays. 1998. SAE Technical [55] Pei YJ, Mehl M, Liu W, Lu TF, Pitz WJ, Som S. A multicomponent blend as a
Paper 980809. diesel fuel surrogate for compression ignition engine applications. J Eng Gas
[25] Siebers DL. Scaling liquid-phase fuel penetration in diesel sprays based on Turbines Power 2015;137, 111502.
mixing-limited vaporization. 1999. SAE Technical Paper 1999-01-0528. [56] Mati K, Ristori A, Gail S, Pengloan G, Dagaut P. The oxidation of a diesel fuel at
[26] Iyengar SV, Rutland C. Effect of physical properties on spray models. 2013. SAE 1-10 atm: experimental study in a JSR and detailed chemical kinetic modeling.
Technical Paper 2013-01-1601. Proc Combust Inst 2007;31:2939e46.
[27] Srivastava S, Schock H, Jaberi F. Numerical simulations of turbulent sprays [57] Ra Y, Reitz RD. A combustion model for IC engine combustion simulations
with a multicomponent evaporation model. 2013. SAE Technical Paper 2013- with multi-component fuels. Combust Flame 2011;158:69e90.
01-1603. [58] Naik CV, Puduppakkam K, Wang C, Kottalam J, Liang L, Hodgson D, et al.
[28] Vasu SS, Davidson DF, Hong Z, Vasudevan V, Hanson RK. N-Dodecane oxida- Applying detailed kinetics to realistic engine simulation: the surrogate blend
tion at high-pressures: measurements of ignition delay times and OH con- optimizer and mechanism reduction strategies. 2010. SAE Technical Paper
centration time-histories. Proc Combust Inst 2009;32:173e80. 2010-01-0541.
[29] Narayanaswamy K, Pitsch H, Pepiot P. A chemical mechanism for low to high [59] Liang L, Naik CV, Puduppakkam K, Wang C, Modak A, Meeks E, et al. Efficient
temperature oxidation of methylcyclohexane as a component of trans- simulation of diesel engine combustion using realistic chemical kinetics in
portation fuel surrogates. Combust Flame 2015;162:1193e213. CFD. 2010. SAE Technical Paper 2010-01-0178.
[30] Curran HJ, Gaffuri P, Pitz WJ, Westbrook CK. A comprehensive modeling study [60] Standard test method for hydrocarbon types in middle distillates by mass
of n-heptane oxidation. Combust Flame 1998;114:149e77. spectrometry. ASTM D 2425-04.
[31] Kelley A, Smallbone A, Zhu D, Law C. Laminar flame speeds of C 5 to C 8 n- [61] Violi A, Yan S, Eddings EG, Sarofim AF, Granata S, Faravelli T, et al. Experi-
alkanes at elevated pressures: experimental determination, fuel similarity, mental formulation and kinetic model for JP-8 surrogate mixtures. Combust
and stretch sensitivity. Proc Combust Inst 2011;33:963e70. Sci Technol 2002;174:399e417.
[32] Ji C, Dames E, Wang YL, Wang H, Egolfopoulos FN. Propagation and extinction [62] Yaws C. Chemical properties handbook. Osborne McGraw-Hill; 1999.
of premixed C5-C12 n-alkane flames. Combust Flame 2010;157:277e87. [63] Kim D, Martz J, Violi A. A surrogate for emulating the physical and chemical
[33] Shen HP, Steinberg J, Vanderover J, Oehlschlaeger MA. A shock tube study of properties of conventional jet fuel. Combust Flame 2014;161:1489e98.
the ignition of n-heptane, n-decane, n-dodecane, and n-tetradecane at [64] Narayanaswamy K, Pitsch H, Pepiot P. A component library frame work for
elevated pressures. Energy & Fuel 2009;23:1367e75. deriving kinetic mechanisms for multi-component fuel surrogates: applica-
[34] Ranzi E, Frassoldati A, Granata S, Faravelli T. Wide-range kinetic modeling tion for jet fuel surrogates. Combust Flame 2016;165:288e309.
study of the pyrolysis, partial oxidation, and combustion of heavy n-alkanes. [65] Bruno TJ, Smith BL. Evaluation of the physicochemical authenticity of aviation
Ind Eng Chem Res 2005;44:5170e83. kerosene surrogate mixtures. Part 1: analysis of volatility with the advanced
[35] Kirchen P, Shahbakhti M, Koch CR. A skeletal kinetic mechanism for PRF distillation curve. Energy & Fuel 2010;24:4266e76.
combustion in HCCI engines. Combust Sci Technol 2007;179:1059e83. [66] Bruno TJ, Smith BL. Evaluation of the physicochemical authenticity of aviation
[36] Tsurushima T. A new skeletal PRF kinetic model for HCCI combustion. Proc kerosene surrogate mixtures. Part 2: analysis and prediction of thermophys-
Combust Inst 2009;32:2835e41. ical properties. Energy & Fuel 2010;24:4277e84.
[37] Curran HJ, Gaffuri P, Pitz W, Westbrook C. A comprehensive modeling study of [67] Standard test method for distillation of petroleum products and liquid fuels at
iso-octane oxidation. Combust Flame 2002;129:253e80. atmospheric pressure. ASTM D86e17.
[38] Sarathy SM, Westbrook CK, Mehl M, Pitz WJ, Togbe C, Dagaut P, et al. [68] Standard test method for calculated cetane index of distillate fuels. 2016.
Comprehensive chemical kinetic modeling of the oxidation of 2- ASTM D976e06, Reapproved.
methylalkanes from C 7 to C 20. Combust Flame 2011;158:2338e57. [69] Li T, Zhang XQ, Wang B, Guo T, Shi Q, Zheng M. Characteristics of non-
[39] Ji C, Sarathy SM, Veloo PS, Westbrook CK, Egolfopoulos FN. Effects of fuel evaporating, evaporating and burning sprays of hydrous ethanol diesel
branching on the propagation of octane isomers flames. Combust Flame emulsified fuels. Fuel 2017;191:251e65.
2012;159:1426e36. [70] Hiroyasu H, Arai M. Structures of fuel sprays in diesel engines. 1990. SAE
[40] Sivaramakrishnan R, Michael J. Shock tube measurements of high tempera- Technical Paper 900475.
ture rate constants for OH with cycloalkanes and methylcycloalkanes. [71] Hiroyasu H, Arai M, Tabata M. Empirical equations for the sauter mean
Combust Flame 2009;156:1126e34. diameter of a diesel spray. 1989. SAE Technical Paper 890464.
[41] Natelson RH, Kurman MS, Cernansky NP, Miller DL. Low temperature oxida- [72] Heywood JB. Internal combustion engine fundamentals. McGraw-Hill Book
tion of n-butylcyclohexane. Combust Flame 2011;158:2325e37. Company; 1988.
[73] Shan XX, Qian Y, Zhu LF, Lu XC. Effects of EGR rate and hydrogen/carbon [76] Westbrook C. Recent advances in chemical kinetics of jet fuels. 52nd Aero-
monoxide ratio on combustion and emission characteristics of biogas/diesel space sciences meeting: American institute of aeronautics and astronautics
dual fuel combustion engine. Fuel 2016;181:1050e7. 2014-0306.
[74] Qian Y, Zhou QY, Wang XL, Zhu LF, Lu XC. Enabling dual fuel sequential [77] Kittelson DB. Engines and nanoparticles: a review. J Aerosol Sci 1998;29:
combustion using port fuel injection of high reactivity fuel combined with 575e88.
direct injection of low reactivity fuels. Appl Therm Eng 2016;103:399e410. [78] Tree DR, Svensson KI. Soot processes in compression ignition engines. Prog
[75] Suzuki S, Hori M, Nakamura H, Tezuka T, Hasegawa S, Maruta K. Study on Energy Combust Sci 2007;33:272e309.
cetane number dependence of diesel surrogates/air weak flames in a micro [79] Merkisz J, Pielecha J. Nanoparticle emissions from combustion engines.
flow reactor with a controlled temperature profile. Proc Combust Inst Springer International Publishing; 2015.
2013;34:3411e7.

Diesel Surrogate

Uploaded by

Copyright:

Available Formats

Diesel Surrogate

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Diesel Surrogate

Uploaded by

Copyright:

Available Formats

Energy 148 (2018) 424e447

Contents lists available at ScienceDirect

A new methodology for diesel surrogate fuel formulation: Bridging

1. Introduction cylinder combustion process. It is well known that the physical

Abbreviations PM Particulate matter

Fig. 1. Components of US #2 ULSD fuel and China stage V diesel fuel.

Analysis of Physical and Chemical Properties of Diesel

Selections of candidate components

Evaluation of engine combustion and emissions

Engine combustion and emissions verification

9 cording to ASTM D86-17 [67]. It can be concluded that: (1) mixed

number/cetane number, adiabatic ﬂame temperature and ﬂame

60 properties changing. According to Table 1, in terms of density,

S ¼ gðDP=ra Þ0:25 (5)

n-Decane 99 142.29 0.732 76 23.41 174 44.236 2.20

n-Dodecane 98 170.34 0.748 80 24.94 216 43.949 2.17

n-Tetradecane 98 198.39 0.761 95 26.15 253 44.020 2.14

n-Hexadecane 99 226.45 0.773 100 27.09 287 43.952 2.13

iso-Octane 99 114.23 0.695 14 18.32 99 44.342 2.25

iso-Dodecane 95 170.34 0.750 e 170 46.1 2.17

HMN 98 226.45 0.793 15 240 43.4 2.13

Methylcyclohexane 98 98.19 0.77 20 23.30 101 43.356 2.00

trans-Decalin 98 138.25 0.896 48 29.38 197 42.618 1.80

Toluene 99.9 92.14 0.87 9 27.93 111 40.523 1.14

n-Butylbenzene 96 134.22 0.860 e 28.63 183.1 41.457 1.40

1-Methylnaphthalene 97 142.20 1.02 0 39.71 240 39.259 0.91

Tetralin 97 132.21 0.970 e 33.16 207.2 40.524 1.20

Diesel e 4.22 52.8 0.830 42.5 214 243 267 338

Percent recovery (vol %)

(a) n-Alkanes (b) iso-Alkanes

Percent recovery (vol %)

(c) Cycloalkanes (d) Aromatics

Fig. 4. The distillation range of ten mixed fuels and diesel.

the difference between injection pressure and ambient pressure.

characteristics in the same injection timing (ﬁxed at -7oCA ATDC)

30 exhaust pressure was the ambient pressure. And the temperature

Fig. 6 compares the effects of surrogate components with

Heat release rate (J/ CA)

Heat release rate (J/ CA)

1600 800 1000

(a) Constant fuel injection timing

(b) Constant ignition timing (CA10)

(a) Constant fuel injection timing

(b) Constant ignition timing (CA10)

sions of aromatics were higher at high loads. The thermal efﬁ-

(c) Indicated thermal efficiency

determined by the loads.

4.4. The effects of n-alkanes carbon chain length

100 Diesel 3bar

n-Decane Tetralin IMEP=3bar

Diesel 6bar Diesel 9bar

(a) In-cylinder pressure and heat release rate

The fuel unsaturation directly or indirectly determines the

cylinder pressure and heat release rate, respectively. This indicated

Heat Release Rate (J/ CA)

Heat Release Rate (J/ CA)

(a) IMEP=3bar (b) IMEP=6bar (c) IMEP=9bar

5. Diesel surrogate formulation

400 n-Butylbenzene 400