Available online at www.sciencedirect.

com

Proceedings of the Combustion Institute 36 (2017) 3389–3413

www.elsevier.com/locate/proci

Alternative fuels for internal combustion engines

Choongsik Bae∗, Jaeheun Kim
Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology, 291 Daehak-ro, Yuseong-gu,
Daejeon 305-701, Republic of Korea

Received 4 December 2015; accepted 5 September 2016

Available online 13 October 2016

Abstract

This review paper covers potential alternative fuels for automotive engine application for both spark igni-
tion (SI) and compression ignition (CI) engines. It also includes applications of alternative fuels in advanced
combustion research applications. The representative alternative fuels for SI engines include compressed nat-
ural gas (CNG), hydrogen (H2 ) liquefied petroleum gas (LPG), and alcohol fuels (methanol and ethanol);
while for CI engines, they include biodiesel, di-methyl ether (DME), and jet propellent-8 (JP-8). Naphtha
is introduced as an alternative fuel for advanced combustion in premixed charge compression ignition. The
production, storage, and the supply chain of each alternative fuel are briefly summarized, and are followed
by discussions on the main research motivations for such alternative fuels. Literature surveys are presented
that investigate the relative advantages and disadvantages of these alternative fuels for application to en-
gine combustion. The contents of engine combustion basically consist of the combustion process from spray

Abbreviations: BSFC, brake specific fuel consumption; Btu, British thermal unit; CAI, controlled auto-ignition; CFR,
cooperative fuel research; CI, compression ignition; CN, cetane number; CNG, compressed natural gas; CRDI, common-
rail direct injection; CO, carbon monoxide; CO2 , carbon dioxide; CTL, coal to liquid; DME, di-methyl ether; DI, direct
injection; EA, elemental analysis; EGR, exhaust gas recirculation; FIE, fuel injection equipment; FSII, fuel system icing
inhibitor; FT, Fischer Tropsch; GCI, gasoline compression ignition; GDI, gasoline direct injection; GHG, greenhouse gas;
GTL, gas to liquid; HC, hydrocarbon; HCCI, homogeneous charge compression ignition; HTR, high temperature reaction;
IEA, International Energy Agency; ICE, internal combustion engines; ISFC, indicated specific fuel consumption; JP-8,
jet propellant-8; LCA, life cycle assessment; LHV, lower heating value; LPDI, liquefied petroleum gas (LPG) direction
injection; LPG, liquefied petroleum gas; LPLi, liquid phase LPG (liquefied petroleum gas) injection; LTC, low temperature
combustion; LTR, low temperature reaction; MPCI, multiple premixed compression ignition; NA, naturally aspirated;
NOx , nitrogen oxides; PCCI, partially-premixed charge compression ignition; PDPA, phase Doppler particle analyzer;
PFI, port fuel injection; PM, particulate matter; PN, particulate number; RCCI, reactivity controlled compression ignition;
RME, rapeseed methyl ester; RON, research octane number; R/P, reserves-to-production; SI, spark ignition; SMD, Sauter
mean diameter; SOC, start of combustion; SOI, start of injection; SFC, single fuel concept/specific fuel consumption;
TEM, transmission electron microscopy; TDC, top dead center; TGA, thermogravimetric analysis; TTW, tank-to-wheel;
TWC, three-way catalyst; WCO, waste cooking oil; WTT, well-to-tank; WTW, well-to-wheel; XPCI, X-ray phase-contrast
imaging.
∗ Corresponding author. Fax: +82 42 350 5044.

E-mail addresses: csbae@kaist.ac.kr (C. Bae), jaeheun.kim@kaist.ac.kr (J. Kim).

http://dx.doi.org/10.1016/j.proci.2016.09.009
1540-7489 © 2016 The Combustion Institute. Published by Elsevier Inc. All rights reserved.
3390 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
development, air–fuel mixing characteristics, to the final combustion product formation process, which is
analyzed for each alternative fuel. An overview is provided for alternative fuels together with summaries
of engine combustion characteristics for each fuel, in addition to its current distribution status and future
prospects.
© 2016 The Combustion Institute. Published by Elsevier Inc. All rights reserved.
Keywords: Alternative fuels; SI engines; CI engines; Advanced combustion
1. Introduction
1.1. Background
Internal combustion engines (ICEs) are ma-
chines that convert the heat produced from
combustion into mechanical work. The main sub-
jects of this paper are reciprocating engines, such
as spark-ignition (SI) and compression-ignition
(CI) engines. They have been widely adopted
as power sources for passenger and commercial
vehicles, electricity power generation, and in other
industrial fields, due to their high power density
and high efficiency. The combustion process is one
of the most important energy conversion methods
where the chemical energy of fuel is directly con-
verted into heat. Therefore, it is possible to say that
human activities are greatly driven by and rely on
fossil fuel energy.
Population growth over the last decades has led
to tremendous growth in fossil energy demand. For-
tunately, predictions of fossil fuel exhaustion keep
expanding, owing to the improvement of drilling
technologies, and the emergence of large quantities
of shale gas (natural gas) reserves. Therefore,
despite the evolution of modern and renewable
energy sources, such as nuclear, solar, and wind en-
ergy, combustion technologies will continue to play
an important role in the energy conversion field.
The two major fuels that had been developed
and widely used along with the development of the
ICEs and the automotive industries over the past
century are gasoline and diesel. The combustion of
gasoline fuel undergoes flame propagation after an
initial spark event igniting the homogeneous air–
fuel mixture in SI engines, while the combustion of
diesel fuel is driven by the auto-ignition of the fuel
exposed to high temperature gas, heated by com-
pression in CI engines. Despite the long history of
a steady supply chain and the determined position
of gasoline and diesel as conventional automotive
fuels in the market, the search for alternative fuels
gradually started to emerge back in the 1980s [1].
1.2. Definition of alternative fuels and their
importance
The definition of alternative fuels may vary de-
pending on the context. The current study defines
alternative fuels as those other than conventional
gasoline and diesel fuels, covering a wide variety in
terms of final forms and manufacturing sources.
For example, ethanol fuel is considered an alterna-
tive for SI engines, regardless of its original source
from either conventional crude oil or any renewable
biomass. The alternative fuels defined by the En-
ergy Policy Act (EPAct) also cover a vast amount
of non-conventional fuels, including alcohols, such
as ethanol (including blends with gasoline over
85%); natural gas and liquefied fuels domestically
derived from natural gas; liquefied petroleum gas
(LPG); coal-derived liquid fuels (CTL); hydrogen
(H2 ); biodiesel (B100); fuels, other than alcohol,
derived from biological materials; and fuel that is
substantially non-petroleum that yields substantial
energy security and environmental benefits.
The significance of using alternative fuels can
be attributed to the following aspects:
(1) pursuing energy sustainability through the
extended usage of those alternative fuels de-
rived from renewable energy sources and
mitigating the concerns of limited fossil fuel
energy;
(2) improving engine efficiency and engine-out
emissions with the aid of superior physical or
chemical properties of alternative fuels com-
pared to those of conventional fuels; and
(3) relieving the unbalanced usage of conven-
tional petroleum-based fossil fuels.
The search for stable energy supplies from
various energy resources together with the associ-
ated shift toward renewable energies is necessary,
because most energy resources currently rely on
fossil fuels, which are of finite availability. Despite
the projection that conventional liquid petroleum-
based fuel would still dominate transportation
fuels for the next 50 years [2], the desire for energy
security (energy sustainability) and concern for re-
ducing greenhouse gas (GHG) emission have also
led to the increased usage of renewable biofuels.
The use of fossil fuels is a major contributor to
CO2 emissions, where human activities generate
about 25 billion tons of CO2 annually [3]. The use
of renewable biofuels can form a CO2 life cycle,
which from a social perspective can contribute to
total CO2 emission mitigation. For example, a full
life cycle assessment (LCA) that was performed
 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
Fig. 1. Global transportation demand by fuel [6].
to compare the GHG impacts of conventional
fuels and biofuels showed that the biofuels derived
from vegetable oil, as well as corn- and wheat-based
ethanol when they are produced using biomass as a
primary energy source, would cut GHG emissions
on a well-to-wheel (WTW) basis to about half of
that of conventional fuels [4]. This analysis took
into account all the GHG emissions generated from
their production, transport and storage, as well
as emissions associated with their use in vehicles,
subtracted by the amounts of CO2 taken in from
the atmosphere by the biomass in its growth phase.
Biofuels accounted for around 3% of the total
global transportation fuels by 2012 [5]. Figure 1
shows that as the energy demand in the transporta-
tion sector keeps increasing in the near term, the
ratio of biofuel to the total amount of fuel is also
expected to increase [6], which from a long-term
perspective may eventually lead to considerable
mitigation of the net GHG emission (Fig. 2) [3].
Implementing low-carbon transportation fuels,
such as second-generation biofuels in addition to
Fig. 2. Well-to-wheel emissions reductions from transport in 4DS and 2DS [3].
3391
the efficiency improvement of vehicles, is expected
to contribute to lowering future GHG emission to
the current level in the best scenario case (referred
as the ‘2DS Scenario’ [3]).
From a technical point of view, the wide use
of renewable biofuels or alternative fuels can also
directly contribute to improvements in engine
performance and emission characteristics. The
balance between fuel efficiency and hazardous
emissions, such as unburned hydrocarbon (HC),
carbon monoxide (CO), nitrogen oxides (NOx ) and
particulate matter (PM), has always been an im-
portant issue in the engine research field. Since the
combustion of fuel is determined by the air–fuel
interaction, the unique properties of alternative
fuels can be advantageously and directly utilized
during the engine combustion process. Research is
being conducted worldwide on the application of
alternative fuels to ICEs to reveal their potentials.
The worldwide demand for energy in the trans-
portation sector is expected to grow continuously
at between 1.2% and 1.4% per annum [7]. At the
same time, the unbalanced growth of demands
for derivatives from fossil fuel may induce issues
of concern. The ratio of middle distillate, such
as diesel, to that of gasoline fuels is expected to
increase to around 2.4 by 2040 from the present
ratio of around 1.5 [8]. This higher demand for
diesel fuel could result in unbalanced energy dis-
tribution. Novel combustion concepts, such as
gasoline compression ignition (GCI), which was
initially proposed to realize compression ignition
with gasoline to improve the drawbacks of other
advanced combustion concepts, proved to also be
feasible with low-grade gasoline-like alternative
fuels, such as naphtha. In other words, these low-
grade by-products during refinery, which are often
treated as waste or that require further processing,
can become alternative fuels for advanced com-
bustion concepts, and contribute to mitigating the
unbalanced usage of conventional fuels without
additional cost [7]. Other major refinery by-
product fuels include light component distillates,
such as propane and butane (which are the major
components for LPG), and methane (natural gas).
3392 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413

Table 1
Average fuel prices in the regional market, as well as the fuel prices on an energy equivalent basis [11].
Fuels Nationwide Nationwide average price in Nationwide average price in
average price (US gasoline gallon equivalents (US dollars per million Btu (US
$/gallon) $/gasoline gallon)b $/Mil Btu)c
Gasoline (regular) 3.34 3.34 28.95
Diesel 3.89 3.49 30.26
CNG 2.09a 2.09a 18.09
Ethanol (E85) 3.04 4.29 37.22
Propane 3.12 4.31 37.35
Biodiesel (B20) 3.97 3.62 31.40
Biodiesel 4.28 4.22 36.53
(B99–B100)
a The price of CNG was normalized to a price per gasoline gallon equivalent (GGE).
b The price per gasoline gallon equivalent (GGE) indicates the prices of fuels normalized to the energy contents of
gasoline fuel per unit gallon.
c The calculation uses the nominal lower heating values (LHV) in British thermal unit (Btu) per gallon of fuel from the
Oak Ridge National Laboratory’s Transportation Energy Data Book.

The extended usage of these fuels in ICEs can
contribute to relieving issues of imbalance.
1.3. Considerations for alternative fuels
Some of the common critical considerations
for alternative fuels for both CI and SI engines are
summarized below:
• Combustion properties (chemical properties,
such as octane and cetane number)
• Physical properties (spray or mixture forma-
tion for combustion, and engine operability
over a wide range of temperatures)
• Lower heating value (LHV)
• Compatibility (including approval by engine
and vehicle manufacturers and costs)
• Manufacturing cost and infrastructures
The combustion properties directly determine
whether or not the given alternative fuels are suit-
able for engine operation. The physical properties
are also important, as they determine the formation
of combustible mixture. The octane and cetane
number will be discussed in details in Sections 2 and
3. The LHV determines the effectiveness of the fuel
as an energy carrier. The comparable level of LHV
to conventional fuels is preferred, otherwise penal-
ties in fuel consumption may become problematic.
Material compatibility with current engine
hardware or fuel-supply infrastructure is important
for the penetration of alternative fuels into the mar-
ket. Otherwise, a great amount of additional cost
would need to be expended for hardware modifica-
tions. Lubrication differences are one of the typ-
ical examples when discussing the compatibilities.
Low lubrication of the fuel, such as di-methyl ether
(DME) [9] or jet propellant-8 (JP-8) [10], may re-
sult in malfunction of sophisticated fuel injection
equipment (FIE), such as common-rail direct injec-
tion system (CRDI) in diesel engines. Thus, lubric-
ity improvers and additives are required to prevent
possible damage to hardware.
Manufacturing cost is one of the practical is-
sues for both manufacturers and consumers. Over
the last few decades, manufacturing costs have
decreased as technologies have improved; however,
the price of alternative fuels is still high compared
to conventional gasoline and diesel. Prices of ma-
jor alternative fuels in the regional markets indicate
that they are more expensive when normalized to
an equal energy contents basis [11]. Table 1 shows
that ethanol price was as high as $37.22/million
British thermal unit (Btu), while the gasoline price
was $28.95/million Btu. The cost of producing fuels
can vary considerably both over time and region-
ally, depending on factors such as the local market
price of feedstocks, and the scale of production [4].
1.4. Current status of production, impact and
future vision
Figure 3 summarizes all the available pathways
for producing liquid and some gaseous fuels from
either fossil sources or biomass [12]. The majority
of transportation fuels today derive from the re-
fining of crude petroleum oil with well-established
technologies [4], though the production from un-
conventional feedstocks, such as oil sands, oil shale
and shale gas, keeps increasing. Synthetic liquid
fuels, which have properties similar to those of
petroleum-derived products, can be produced from
the gasification and Fischer–Tropsch (FT) process
of any hydrocarbon or coal feedstock. Table 2 sum-
marizes the important features of various fuels, in-
cluding conventional and alternative fuels [4].
A number of familiar alternative fuels, such
as ethanol (bio-ethanol) and advanced biodiesel,
which show moderate compatibility with existing
ICE vehicles, are in moderate-scale production and
fulfill moderately-low GHG emissions. The ad-
vancement of production technology from conven-
tional methods (or 1st generation biofuels) to ad-
vanced methods (2nd generation biofuels) has en-
 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
Fig. 3. Manufacturing paths of fossil and regenerative fuels [12].
hanced the efficiency of production, and promoted
the utilization of inedible bio-feedstocks. The use
of edible sources can lead to increases in food
price, and has been widely criticized in the past.
Table 3 summarizes the annual worldwide ethanol
production by countries [13]. Ethanol is regarded
as an eco-friendly fuel, because traditional feed-
stocks, like molasses, sugarcane juice and corn, can
function as a CO2 sink [14]. In particular, corn and
sugar cane are the main feedstocks of the U.S. and
Brazil, which are the largest ethanol producers in
the world. The U.S. is also the leading producer of
biodiesel, producing 940 thousand barrels per day
in 2012 [15].
Table 2 shows that compressed natural gas
(CNG) and LPG also feature better compatibility
with the existing distribution infrastructure than
those alternative fuels derived from CNG (such as
DME and H2 ). These fuels are still attractive for
application to vehicles with certain modifications.
Besides, interest in CNG as an alternative fuel has
renewed recently, due to the accessibility of shale
gas reserves. The reserves-to-production (R/P) ratio
for natural gas in 2015 has become about 54, while
that for crude oil is about 52 [16]. Figure 4 shows
that the share of shale gas of the total U.S. natu-
ral gas production has increased from 1% in 2000,
to 40% in 2012; and is expected to increase to 53%
in 2040 [17]. Shale gas has yielded a stable natural
gas price in the market, while the oil prices showed
fluctuations over the last few years due to the shale
gas production.
3393
Several researchers have carried out extensive
WTW CO2 emissions and efficiency research based
on the LCA of major alternative fuels. Even though
the WTW energy requirements for biofuels are
greater than those for conventional gasoline and
diesel, their WTW GHG emissions are comparably
lower [18]. For example, the GHG reduction range
indicated for soy methyl-ester derived biodiesel
was around 50% that of conventional diesel [19].
However, the magnitude of CO2 reduction for
bio-ethanol was lower compared to other biofuels.
Transition from grain- or seed-based biofuels to
high-yielding lignocellulosic energy crop derived
biofuels was suggested to further mitigate the
GHG emission impacts of biofuels [19].
1.5. Objectives of the research
Even though the rest of the alternative fuels do
not appear to be attractive in terms of compati-
bility with currently existing systems, this does not
mean that they are inapplicable. Attention has been
paid to other alternative fuels as well, including
H2 , other alcohol fuels, DME, and JP-8. Besides,
derivative products from the refinery process, such
as naphtha, which is inapplicable for conventional
SI engines, were also studied for engine application
with novel combustion concepts. Other alternative
fuels which have not been mentioned in Table 2 in-
clude, ammonia (NH3 ) and various ethers, such as
methyl tertiary butyl ether (MTBE), ethyl tertiary
butyl ether (ETBE) and tertiary amyl methyl ether
3394 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413

Table 2
Comparison of the production process and characteristics of various fuels [4].

(TAME) for SI engines; diethyl ether (DEE) and
various esters, such as soybean methyl ester (SME),
rapeseed methyl ester (RME) and hydrogenated
soy ethyl ester (HySEE) for CI engines [20]. It is
necessary to individually assess each fuel, due to
its distinctive attributions when used on ICEs.
The current paper offers a brief introduction to
the background, current production methods, and
infrastructure, such as the supply chain, of repre-
sentative alternative fuels. The potential and fea-
sibility of implementation of each alternative fuel
in ICEs were analyzed in terms of spray and com-
bustion characteristics. It is important to focus on
these two aspects, because the combustion of fuel
is closely related to the air–fuel interaction. The
advantages and disadvantages of each alternative
fuel are analyzed in terms of engine efficiency and
emission formation. The list of representative al-
ternative fuels is as follows: the fuels for SI en-
gines are CNG, alcohol fuels such as methanol and
ethanol, LPG and H2 ; while the fuels for CI engines
are biodiesel, DME, and jet propellant-8 (JP-8).
Naphtha is included in the advanced combustion
field.
 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
Table 3
Annual worldwide ethanol production by country (Million Gallons) [13].
Country 2007 2008 2009
United States 6521 9309 10,938
Brazil 5019 6472 6578
Europe 570 734 1040
China 486 502 542
Canada 211 238 291
Rest of World 315 389 914
Total 13,123 17,644 20,303
2. Alternative fuels for SI engines
2.1. Fuel requirements for SI engines
In conventional SI engines, the air–fuel mixture
is inducted into the cylinder, and compressed to
the vicinity of top dead center (TDC), ready to
be ignited. Combustion is initiated by an elec-
tric discharge at the spark plug. The distinctive
characteristic of the combustion phenomena in
SI engines is the flame propagation of a premixed
air–fuel mixture [21]. Auto-ignition may occur in
SI engines, referred to as ‘knock’; however, it is an
undesirable abnormal combustion phenomenon.
Knock can be multi-point spontaneous auto-
ignition throughout the entire premixed air–fuel
mixture in the end gas region, accompanying ex-
tremely rapid heat release and damage to the engine
(end gas is the unburned gas inside the cylinder,
which is ahead of the propagating flame, ready to
be burned). Knock can be caused by fuel properties
as well as by engine design parameters. There are
several requirements for the fuels for SI engines,
including the prevention of knock. The following
are some of the requirements for gasoline fuels, as
well as for alternative fuels for SI applications [22].
• Octane number
• Flammability related to lean limit and com-
bustion stability
• Laminar burning velocity
• LHV (of air–fuel mixture)
• Volatility; boiling curve; vapor pressure
The octane number determines the knock re-
sistance of a given fuel, and it is a decisive factor
that determines whether the given alternative fuel
is applicable to SI combustion or not. Iso-octane,
which is extremely knock-resistant, is given the
octane number of 100, while n-heptane, which is
knock-susceptible, is assigned as 0. The octane
number of an unknown fuel is determined with
a cooperative fuel research (CFR) engine, where
its octane number is equal to the numerical value
that corresponds to the proportion of iso-octane
in an iso-octane/n-heptane mixture which demon-
strates the same knock characteristics. The octane
number can be displayed by either Research Oc-
tane Number (RON) or Motor Octane Number
3395
2010 2011 2012 2013 2014
13,298 13,948 13,768 13,300 14,300
6922 5573 5577 6267 6190
1209 1168 1179 1371 1445
542 555 555 696 635
357 462 449 523 510
985 698 752 1272 1490
23,311 22,404 21,812 23,429 24,570
(MON), according to the measurement method.
Normal (regular) gasoline has a RON and MON
value of 91 and 82.5 for the particular market
[22]. The occurrence of knock, or auto-ignition,
in gasoline engines can be quantified through the
Livengood–Wu integral equation [23]. The decisive
factor for such undesired knock phenomena can
be attributed to the auto-ignition delay time of
the fuel, where the ignition delay itself is also
determined by the ambient temperature and fuel
reactivity. The octane number of gasoline hinders
further compression ratio increment, thus limiting
the efficiency of the current engine. The octane
number of alternative fuels may play a dominant
role in future gasoline engine development.
Lean combustion concepts have been explored
to mitigate the pumping loss of SI engines at
part-load conditions. However, the tail-pipe NOx
emissions have been pointed out as one of the
main drawbacks, due to the deteriorated NOx
conversion efficiency of the after-treatment system
(three-way catalyst (TWC)) under lean operation.
The typical solution for mitigating high NOx
emissions is to apply exhaust gas recirculation
(EGR) to substitute for the excessive amount of
O2 . Since the lean limit of an air–fuel mixture is
determined by the flammability of the fuel itself,
alternative fuels with wider flammability windows
can be helpful in increasing the EGR tolerance,
leading to further reductions of NOx emissions.
Although the combustion duration of the mix-
ture is greatly governed by the turbulence intensity
of the mixture, the laminar burning velocity acts
as an important property of the fuel. An alterna-
tive fuel that has the attribute of faster burning
velocity can mitigate the necessity of increasing
the turbulent intensity of the mixture with com-
plicated port design. Besides, the improvement of
knock occurrence can also be expected, due to the
end-gas participating in flame propagation before
auto-ignition occurs.
The volatility or evaporation characteristics
of liquid fuel is important, due to the intrinsic
characteristics of the FIE system of conventional
port fuel injection (PFI) SI engines. Since the
fuel needs to be premixed prior to spark ignition,
good fuel volatility ensures reliable startability
at cold condition, and homogeneous mixture
3396 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
formation for combustion stability. Nowadays,
the FIE system is shifting from PFI toward direct
injection (DI), due to the ability to improve engine
efficiency by increasing the compression ratio with
the charge cooling effect of the fuel. The charge
cooling effect improves the knocking tolerance
due to charge temperature reduction, and provides
room for a compression ratio increment. However,
the time for mixture formation (the time between
fuel injection and ignition) becomes shorter with
DI, especially for the stratified DI concept, which
injects the fuel late during the compression stroke.
This may yield a locally inhomogeneous mixture
that is prone to deteriorated combustion efficiency.
The fuel volatility of alternative fuels can also
result in different schemes of mixture formation,
compared to conventional fuels.
2.2. Compressed natural gas
2.2.1. Background, current production, and supply
Natural gas was usually obtained as a by-
product when refining crude oil. The unwanted
gas was usually burned at the oil fields from the
early era until the early 20th century. However,
nowadays the gas is gathered, and transported
by pipeline for power generation, domestic use,
or delivered to other end users, including export
in the form of liquefied natural gas (LNG), and
conversion into liquid forms through GTL tech-
nologies. LNG is generally re-gasified into natural
gas, before being fed through the pipelines into
the distribution grid. The gas is also in direct use
in the transportation sector, due to its lower price
compared to other conventional fuels (Table 1).
The abundance of natural gas, followed by the
realization of extracting technologies from shale
gas, has become another attracting factor for the
wider usage of the fuel in ICEs. The sources of
natural gas now include oil fields, coal beds, and
gas fields where shale gas is stored. This abundance
can ensure further reduction in the fuel cost, and
fulfill a stable energy supply. It was reported that
the WTW GHG emission for CNG lies between
gasoline and diesel [24]. From a long-term per-
spective, CNG vehicle emissions from both conven-
tional sources and shale gas would be lower than
that of gasoline, ranging around 220 g CO2 eq/km
[25]. The WTW GHG emission would approach to
that of diesel, while WTW energy use would remain
slightly higher than that of gasoline beyond 2020
[24].
2.2.2. Fuel properties and their effects on
combustion
Natural gas primarily consists of methane up
to 94.4%, followed by other minor substances,
such as ethane and propane [26]. Table 4 compares
conventional gasoline, diesel, and other alternative
fuels for SI engines (methane, H2 , LPG) [21,27,28].
Natural gas has a higher LHV per unit mass than
any other fuel [21]. The low carbon/hydrogen ratio
Fig. 4. U.S. natural gas production by source, 1990–2040
(trillion cubic feet) [17].
in the fuel molecule also allows natural gas to emit
smaller amount of CO2 emission compared to any
other hydrocarbon fuels, when equal amounts of
mass of fuels are burnt. However, transportation
usage needs vessels for the high-pressure storage of
natural gas. The fuel is used in the form of CNG.
Even though natural gas has high energy density
on a mass basis, the energy density on a volume
basis is lower than those of gasoline and diesel.
Thus, the travel distance of a natural gas vehicle is
shorter than that of conventional vehicles. There-
fore, CNG-driven vehicles can be easily adopted in
cities or suburban areas, where travel distances are
relatively short, and good supply infrastructures
are available [29]. For vehicles needing to travel
long distances, LNG is a good alternative in some
commercial vehicles.
Methane has a higher octane number than
gasoline. A higher octane number fuel allows
higher compression ratios to be used in the engines
to increase efficiency. The lack of a long carbon–
carbon chains, in addition to simple and stable
molecular structure consisting only of carbon–
hydrogen bonds, allows strong resistance to the
knocking phenomenon.
The gas phase of the fuel can be advanta-
geous for the air–fuel mixing process; therefore, re-
lated emissions reduction, such as PM, can be ex-
pected, due to more homogeneous mixture forma-
tion [26,30,31]. However, the gas phase acts as an
obstacle when dealing volumetric efficiency, unless
DI technology is implemented. The torque with
natural gas can be decreased by more than 16%
compared to gasoline, due to this issue [30,31]. The
brake specific fuel consumption (BSFC) for natural
gas is less than gasoline throughout the speed range
under full throttle, because the LHV of natural
gas is greater than that of gasoline by around 14%
[31].
The direct injection of CNG into the cylinder
after the intake valve closes could improve the
 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413 3397

Table 4
Comparison of alternative fuels for SI engines [21,28,53] and conventional fuels.
Item Gasoline Diesel Methane Hydrogen LPG [28] Alcohol [53]
Propane Butane Methanol Ethanol
Formula (phase) Cn H1.87 n (l) Cn H1.8 n (l) CH4 (g) H2 (g) C3 H8 C4 H10 CH4 O C2 H6 O
Density (kg/m3 ) 0.72–0.78 0.84–0.88 0.72a 0.09a 0.51 0.58 0.792 0.785
Low heating 44.0 42.5 50.0 120.0 46.1 45.5 19.9 26.9
value (MJ/kg)
Octane number 92-98 – 120 ≥120 111 103 109 107
(RON)
Cetane number – 52 – – – – –
Stoichiometric 14.6 14.5 17.23 34.3 15.6 15.4 6.5 9.0
air–fuel ratio
(A/F)
Latent heat of 305 270 509 – 426 385 1168 840
vaporization
(kJ/kg)
Boiling point 27–225 – −161.5 – −42 −0.5 64.7 78
(°C)
Flame speed 0.37–0.43 – 0.38 1.85 0.38 0.37 – 0.39 [13]
(m/s) [27]

(l) liquid phase; (g) gaseous phase.

a Density in kg/m3 at 0 °C and 1 atm.

torque at low engine speeds [32]. The engine torque
of a CNG PFI engine was lower by 30%, compared
to the gasoline DI mode. It was reported that the
use of a CNG DI system and optimum spark
timing could compensate approximately 2/3 of the
torque loss at low engine speeds.
Engines with natural gas generally produce
lower CO and non-methane HC emissions, com-
pared to conventional gasoline engines [24].
Typically, more than 90% of the total unburned
HC is methane. Methane has a global warming
potential that is 20 times that of CO2 . The issue of
methane emission in the engine tailpipe should be
taken care of, because the benefits of CO2 emission
reduction can be otherwise cancelled out due to
the substantial increase in methane emissions.
2.3. Hydrogen
2.3.1. Background, production, and supply
Hydrogen (H2 ) is widely used as the fuel for
combustion over a wide range of industries, in-
cluding the aerospace field, due to its high energy
density in mass basis, high flame speed, and wide
flammability range (Table 4). There is no doubt
that it can also be used in the transportation
sectors. Hydrogen can be used under various com-
bustion regimes, from conventional to advanced
combustion concept [33-35]. Hydrogen can be
produced from fossil resources, such as natural gas
and coal, as well as renewable resources, such as
biomass. Hydrogen can also be produced through
electrolysis of water with solar-generated electric-
ity in terms of CO2 emission reduction; however,
the production scale is very small currently. To-
day, more than 90% of the hydrogen produced is
derived from fossil fuels, including natural gas, oil
and coal, via a steam reforming process [36].
2.3.2. Fuel properties and their corresponding
combustion characteristics
The high octane number of hydrogen also pro-
vides the opportunity to increase the compres-
sion ratio to increase the engine efficiency. Its wide
flammability limits also enable un-throttled opera-
tion with lean combustion. The high laminar flame
speed, an order of magnitude higher than hydro-
carbons, is a distinct combustion characteristic of
H2 . The laminar flame speed is about 1.85 m/s for
H2 , while the speed for hydrocarbon fuels range
from 0.3 to 0.5 m/s. Therefore, a PFI engine oper-
ating with H2 has a higher thermal efficiency over
the conventional gasoline-fueled engine, due to
the availability of lean unthrottled operation, high
flame speed, and potential for higher compression
ratio. However, attentions should be paid on the
possibility of greater heat transfer losses through
the cylinder-wall with H2 fueled engines [37].
The auto-ignition temperature of H2 is 858
K, which is higher than that of gasoline fuel
(around 550 K). However, the minimum ignition
energy of hydrogen is 0.02 mJ (while that for
equivalent gasoline–air mixture is 0.24 mJ) and
may cause backfire and pre-ignition, which are
critical issues for H2 -fueled engines. Pre-ignition is
a phenomenon whereby an ignition of the air–fuel
mixture occurs before the spark plug fires, leading
to engine damage. Pre-ignition mostly takes place
at engine hot spots, such as spark electrodes, valves,
or oil contaminants [38]. Local hot spots can easily
3398 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
be created when the compression ratio of the
engine, or the engine load (namely the equivalence
ratio of the mixture) are raised too much. Such con-
cerns bring boundaries to the operation windows,
and thus limit the maximum power of the engines.
H2 combustion does not produce carbon-
related emissions, such as CO, CO2 and PM, due
to the absence of C atoms. However, note the car-
bon footprint if carbon-based feedstocks are used
to produce H2 . NOx is the only non-trivial engine-
out emission pollutant. There are some available
approaches to reduce NOx emissions [37]. The first
approach is to adopt an ultra-lean combustion with
an equivalence ratio (ϕ) lower than 0.45, in order to
reduce the peak combustion temperature; however,
this limits available engine torque. The low energy
density on a volumetric basis also hinders the max-
imum torque characteristics. Another approach
is to install an after-treatment downstream of the
engine. A lean NOx trap (LNT) would be needed
when operating at a moderately lean condition
with equivalence ratio below 1 at higher cost; while
a lower cost TWC is required when operating at
stoichiometric conditions. However, research on
the prevention of abnormal combustion, such
as pre-ignition, needs to be resolved, prior to
operation under the stoichiometric condition.
2.4. Liquefied petroleum gas
2.4.1. Background, production, and supply
Currently, there are three distinct sources of
LPG production [39]. The first source is as a by-
product during the refining of crude oil, where light
component fuels such as propane and butane are
produced through distillation, reforming, cracking,
and other processes. These gases are gathered and
have already become widely used in certain regional
markets as alternative fuels in the passenger vehicle
sector. The second source is extraction from the
processing during natural gas (or non-associated
gas) extraction. The LPG is extracted to prevent
the gas from condensing, and causing problems
with natural gas transportation. The last source
comes from the process during oil extraction
(source in the associated gas) [39]. Around 60%
of LPG worldwide is produced with the recovery
process from fuel extraction, while the remaining
40% comes from refining the crude oil [39,40].
2.4.2. Fuel properties and their effects on spray and
combustion characteristics
By virtue of the variation of its sources and
processing processes, Table 5 shows that the com-
position of LPG is highly variable both across the
globe, and during the year [41]. While LPG content
varies significantly across the globe, it is primarily
a mixture of four species: propane, propylene,
iso-butane, and n-butane [41]. Each component
of the four species has an octane number over
RON 100, which is higher than that of gasoline.
Therefore, increase of compression ratio of the
engine is also possible with LPG fuel. The high
octane number of LPG is beneficial in increasing
the brake thermal efficiency, by optimized ignition
timing without knock occurrence, and increase of
compression ratio [42].
Table 4 summarizes the properties of the main
components of LPG, which are propane and
butane. Since at the standard condition it is in
gas phase, the LPG FIE for SI engines has four
different types, namely mixer type, gas phase PFI,
liquid phase PFI, and liquid phase DI. The unique
injector geometries are determined according to
the fuel injection types. Mixer and gas phase PFI
are not dealt with hereinafter, because those types
work in a similar manner to the conventional
carburetor system.
Although PFI in gas phase is advantageous for
the mixing process with air, it suffers from low
volumetric efficiency in a similar manner to other
gaseous fuels, such as CNG and H2 . Higher vol-
umetric efficiency can be achieved through liquid
injection at the manifold [43].
In general, the injection of liquid phase fuel
with high injection pressure generates a fine at-
omization of the spray with shorter liquid length
penetration. The spray pattern of liquid phase
LPG PFI (LPLI) under atmospheric pressure con-
ditions also revealed that high injection pressure
led to wider spray angles, and better atomization of
droplets with shorter liquid length [44]. Numerical
results show that the injection velocity primarily
determined the penetration at a very early stage
of injection. However, faster evaporation of the
fine droplets into the gas phase due to low boiling
point resulted in vigorous momentum exchange
with entrained ambient air. Rapid loss of its initial
momentum resulted in shorter penetration and
wider spray angle [45].
The boiling point and vapor pressure of LPG
are lower than those of gasoline fuel [46]. Thus, the
flash boiling phenomenon can be utilized for high
injection pressure LPG sprays under part-load
condition (throttled operation), where the ambient
pressure is low. The LPG spray characteristics with
swirl injectors exhibited much higher spray velocity
under flash boiling conditions, due to the expan-
sion of the fuel vapor followed by the flash boiling.
DI systems for LPG (LPDI system) are under
research, along with the trend toward gasoline
DI technologies. One of the problems that the
gasoline direct injection (GDI) system faces is oil
dilution, due to the cylinder wall wetting of the
direct-injected liquid fuel. Substantial reduction
in fuel wetting in the combustion chamber was
achieved due to the shorter liquid length of LPG.
HC emissions were lower than that of gasoline
by up to 45%. The enhanced mixture formation
due to wider spray angle and flash boiling of LPG
provided room for HC and CO emission reduction,
regardless of either gas or liquid phase injection,
 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413 3399

Table 5
Variable composition of LPG for auto-gas by country [41].
Country Fuel composition (%, vol/vol)
Ethane Propane Propylene Butane
Australia 0–10 0–100 0–45 0–50
Belgium 0–5 50–100 0–50 0–10
Canada – 92.5–100 0–5 0–2.5
France 0–5 50–100 0–50 0–5
Hong Kong 0–5 20–40 – 60–80
Italy 0–5 40–100 0–50 0–50
South Korea – 10–35 – 65–90
Mexico – 60 – 40
Netherlands 0–5 50–100 0–50 0–50
New Zealand – 60–70 – 30–40
Poland – 20–60 – 40–80
Singapore 0–5 20–40 – 60–80
Spain 0–5 87.5–100 0–5 0–5
United Kingdom 0–5 50–100 0–50 0–5
United States – 92.5–100 0–5 0–2.5

compared to gasoline fuel. In addition, particulate
number (PN) emissions with LPDI application
was reduced by 94% compared to that of gasoline
in NEDC due to better mixture formation [47].
The stratified combustion mode is regarded as
a further means to reduce fuel consumption at
part load conditions, due to the reduced pumping
and heat losses. The key for stable ignition is the
preparation of ignitable mixture close to the spark
location. Although flash boiling happens to be
advantageous in homogeneous mixture forma-
tion, the spray structure deformation due to flash
boiling may become an obstacle and challenge for
spray targeting near the vicinity of spark location
[48,49]. Low injection pressure or change of spark
plug location might be required for stable stratified
combustion of LPG.
2.5. Alcohol fuels—methanol and ethanol
2.5.1. Background, production, and supply
Alcohol has accompanied the history of
mankind from ancient civilization to the present.
Through the industrial revolution in the 19th
and early 20th centuries, alcohol began to gain
attention as an automotive fuel. The potential
to achieve higher thermal efficiency and lower
GHG emission due to the unique properties and
its renewable production by plants have gained
attention. Among the alcohol fuels, it is known
that methanol (CH3 OH), ethanol (C2 H5 OH),
propanol (C3 H7 OH) and butanol (C4 H9 OH) can
be used as automotive fuels. Methanol and ethanol
were successfully introduced as alternative fuels
for conventional spark ignition engines in the form
of fuel additives, blend fuels and bi-fuels, due to
technical and economic reasons.
Methanol can be manufactured by coal, natural
gas, coke-oven gas, hydrogen and biomass. How-
ever, most methanol is produced by synthesis gas,
which is composed of carbon monoxide and hy-
drogen [50]. Production of ethanol has two major
industrial production pathways. One pathway is
the reaction of ethane with steam, while the other
pathway is alcoholic fermentation [51]. The reason
why ethanol is often called bio-ethanol is that
ethanol is produced from renewable bio-organic
materials, and these bio-organic materials can act
as a carbon sink in its lifetime. The ethanol can
be easily delivered and supplied with some minor
modifications using the current infrastructure,
because it is in liquid phase and easy to handle,
compared to gaseous alternative fuels.
Ethanol demand and supply almost tripled
over the last decade [50]. Table 3 also shows the
increasing trend of annual worldwide ethanol
production. However, the low cost at present of
conventional petroleum-derived fuel mitigates the
profitability of the production of alcohol fuels.
A careful selection of reasonable feedstock is
required, taking the production cost and energy
efficiency of production into account. Numerous
studies showed quite contradictory and debatable
results through the LCA of ethanol on the benefit
and cost of ethanol production, considering its
use in internal combustion engines. One of the
studies indicates that the GHG emission reduction,
energy efficiency and environmental sustainability
were low, when taking the air pollution, soil and
groundwater contamination issues [52].
2.5.2. Fuel properties and their effects on spray and
combustion
Table 4 lists several important properties of
pure methanol and ethanol as fuels [53]. Methanol
and ethanol share similar properties as alcohol
fuels. They are both oxygenated liquid hydro-
carbon fuels. It is known that the distinctive
3400 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
characteristics of these alcohol fuels basically
arise because of the existence of the hydroxyl
group in the carbon chain. This hydroxyl moiety
makes methanol and ethanol hydrophilic, and
has a dipole moment. This polarization by dipole
moment affects the inter-molecular interactions,
and makes them hydrophilic and behave close
to an azeotrope when blended with gasoline.
The vapor pressure and distillation properties
are affected by those intermolecular interactions.
Therefore, the Reid vapor pressure and distillation
property changes do not show a linear corre-
lation as a function of the ethanol blend ratio,
which can become an important issue in GDI
engines. The spray structure is directly influenced
by such evaporation characteristics, in a similar
manner to that in LPG, and alters the preparation
of air–fuel mixture under different engine load
conditions.
The LHVs of methanol and ethanol are sig-
nificantly lower than that of gasoline. However,
the alcohol fuels show higher octane numbers,
which is favorable to achieve higher thermal effi-
ciency. Studies on the fuel consumption behavior
between increased compression ratio and octane
number fuel [54] showed that an octane num-
ber increment of approximately 4–6 [55,56] was
required per unit increase in compression ratio.
The benefits gained through increased thermal
efficiency and the loss for increased fuel con-
sumption should be taken into account when
calculating the overall fuel economy. Alcohol fuels
have higher latent heats of vaporization than that
of gasoline; therefore, enhanced charge cooling
effects can be expected. The cooling effect itself
can also contribute to increased knock tolerance
and volumetric efficiency. Studies on the relation
between the heat of vaporization and the poten-
tial compression ratio increment require further
investigation.
The vapor pressure of gasoline is higher and
the boiling point is lower than those of ethanol be-
tween the temperature ranges of 300–410 K. There-
fore, engine stability and emission problems under
cold-start condition need to be taken care of when
implementing ethanol fuel. The Ohnesorge dia-
gram of Reitz and Bracco’s [57] break-up regimes
revealed that the calculated value for ethanol at
low temperature (280 K) was within the boundary
of second wind induced regime [58], which implies
the concern of bad atomization of the fuel during
cold-start. However, when taking the in-cylinder
pressure and temperature during the compression
stroke into account, the vapor pressures of gaso-
line and ethanol vary. The vaporization curves
of gasoline, ethanol and iso-octane drawn by Oh
et al. [59] with the help of the Peng–Robinson
equation [60] showed that the vapor pressure of
ethanol increased beyond that of gasoline when
the ambient temperature increased above 400 K.
This indicates that ethanol spray showed faster
diffusion and evaporation under an elevated tem-
perature condition, which is a simulated condition
at spark timing under stratified combustion mode.
The results suggested that the injection timing of
ethanol blended fuel could be retarded close to
TDC as the ethanol content increased, due to faster
evaporation and enhanced local homogeneity [59].
Various ratios of ethanol blended fuels and gaso-
line sprays also showed that the spray liquid length
decreases and the spray angle increases as the
ethanol contents increase at low ambient pressure
[61].
The increase of droplet size after the injection
event is attributed to the higher kinematic viscosity
and surface tension of the ethanol [61]. Higher
Sauter mean diameter (SMD) of ethanol droplet
size was measured through the phase Doppler
particle analyzer (PDPA) techniques. When con-
sidering the effects of fuel temperature on spray
atomization, one might expect flash boiling spray
(in a similar manner mentioned previously in
respect to LPG fuel) when the fuel temperature
was beyond the boiling temperature of ethanol (78
°C at ambient pressure). However, there were no
significant differences of macroscopic spray struc-
ture between gasoline and ethanol–gasoline blend
fuels (namely E85) under both flash boiling con-
dition (fuel temperature of 100 °C) and non-flash
boiling condition [58], since gasoline is a multi-
component fuel with a wide range of distillation
curves. It was reported that flash boiling did not
occur until the excess temperature reached beyond
the boiling temperature by 14 K [62]. The higher
volatility and lower boiling point of methanol
(64.7 °C) made methanol and methanol blended
with gasoline exhibit slightly more distinct flash
boiling behavior under an elevated temperature
condition [50].
It is generally accepted that ethanol blended fuel
exhibited higher PM and PN emissions than that
of gasoline, mainly due to the lower evaporation
characteristics of ethanol at ambient temperature
below 300 K. However, the PM emission shows
controversial results under warm-up conditions.
The total PM emission with E85 showed eight
times more than that of gasoline using a spray-
guided single cylinder engine [63], while in other
cases, the PM emission reduced by about 30%
and 42% under the FTP and US06 driving cycles,
respectively [64]. It was reported that the higher
enthalpy of vaporization and deteriorated spray
break-up led to increase of the PM emission
[63]. Since the PM formation involves parame-
ters such as engine load condition, temperature,
and mixture formation, further research in this
field is required, in order to fully understand the
formation of particulate emission from ethanol
blend fuel, with respect to engine operating
conditions.
 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
3. Alternative fuels for CI engines
3.1. Fuel requirements for CI engines
Diesel fuels are the conventional fuels for CI
engines. The distinct difference of combustion
in CI compared to SI engines is that the fuel is
directly injected into the cylinder, and auto-ignites
due to the high ambient temperature toward the
end of the compression stroke. Therefore, the
auto-ignition ability of the fuel is crucial for its use
in CI engines. Several important criteria exist to
measure the quality of the diesel fuels, as well as
alternative fuels applied to CI engines [65], namely:
• Cetane number
• Boiling point
• Narrow density and viscosity spread
• Low aromatic compounds (particularly pol-
yaromatic compounds) content
The cetane number (CN) strongly determines
the ignition quality of the fuel. It is also a primary
decisive factor for whether a given alternative fuel
is applicable to CI engine application. A cetane
number of 100 is assigned to n-hexadecane, which
ignites easily, while cetane number of 0 is assigned
to methyl naphthalene, which features slow-
ignition. The cetane number of an unknown fuel
is measured in a CFR single-cylinder test engine
through a standard American Society of Testing
Materials (ASTM) procedure. The proportion of
n-hexadecane in the mixture is varied, until the
same ignition delay as that of unknown fuel is ob-
tained. The proportion of n-hexadecane becomes,
by definition, the cetane number. It is accepted that
a cetane number that exceeds 50 is desirable for
optimized operation in modern CI engines [65].
The cetane number is an engineering measure to
classify the ignition quality of the fuel. The cetane
number takes both the physical properties and
chemical properties of the fuels into account. At
the time when the cetane number was developed,
the chemical composition of the fuel had a greater
relevance on the ignition delay rather than the
diesel engine technologies. The advent of novel
technologies, such as CRDI, yielded a technical
gap against the CFR engine; thus the ignition char-
acteristics determined by the traditional method
do not always correlate well with the ignition
delays in production engines, particularly for some
alternative fuels [66]. However, the scale is still an
important measure when evaluating alternative
fuels for CI engines, since it directly measures the
quality of the auto-ignition.
The fuel and air mix vigorously, and form a
combustible mixture during the ignition delay
period. The physical delay includes heating and
evaporation of the liquid fuel, forming a com-
bustible mixture. The chemical delay includes
pre-ignition reactions that break down the hy-
drocarbon fuel and generate radicals, followed by
3401
Fig. 5. Conceptual schematic of conventional diesel com-
bustion [67,68].
highly exothermic reaction leading to the premixed
phase of diesel spray combustion [67]. The heat of
vaporization and auto-ignition temperature play
major roles in these stages.
Figure 5 shows a conceptual model of the
conventional diesel spray combustion during the
quasi-steady period proposed by Dec [68]. The
diffusion flame remains a distance away from the
nozzle downstream, where the distance is referred
to as the lift-off length (LOL) [67]. Formaldehyde
(HCHO), which is a marker for the cool-flame
during the transient ignition process, is initially
formed upstream of the LOL, and is consumed
downstream of the LOL in the fuel-rich premixed
reaction zone [69]. Soot is formed as a result of
both rich the premixed reaction zone and the hot
fuel-rich core of the diesel jet.
In general, smoke emission acquired from the
tail-pipe is a final result of competition between
soot formation and the oxidation process. The
soot formation process can be divided into par-
ticle formation and particle growth. The prod-
ucts after oxidation and/or pyrolysis from the fuel
molecule mainly consist of various unsaturated hy-
drocarbons, particularly acetylene and its higher
analogs (C2 n H2 ), and polycyclic aromatic hydrocar-
bons (PAH). These two types of molecules are con-
sidered the most likely precursors of soot in flames
[21]. The amount of air entrainment, which is cru-
cial for lowering the locally fuel-rich region and
suppressing soot formation, increases with increas-
ing LOL [70-72]. The equivalence ratio at the LOL
is determined by both engine parameters, such as
ambient O2 concentration etc., and also the proper-
ties of the fuel itself, such as the auto-ignition tem-
perature and stoichiometric air–fuel ratio.
The formation of NOx is largely produced
around the jet periphery, where the equivalence
ratio just after where diffusion flame forms is near
stoichiometric [73]. The measured NO emission
with PLIF signal indicated that no measurable NO
was formed in the premixed burn region of the con-
ceptual model (Fig. 5), due to too rich combustible
mixture with an equivalence ratio of greater than
three. The NOx emission is significantly related
3402 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
to the peak combustion temperature, and the
residence time for formation [21]. The generation
of nitrogen oxide is described by the Zeldovich
mechanism. Simple correlations of flame adiabatic
flame temperature versus NOx emission also sup-
port such a theory [74,75]. The fuel properties, such
as the LHV of the fuel, heat of vaporization, and
oxygen contents in the fuel molecule, can influence
the adiabatic flame temperature.
Overall, the alternative fuels of various physic-
ochemical properties may result in different engine
performance and engine-out emissions, due to
different mixture formation processes. The con-
tribution of the unique characteristics of the
alternative fuels will be discussed.
3.2. Biodiesel
3.2.1. Background, production, and supply
Biodiesel is an alternative fuel based on bio-
originated feedstock. A variety of oil feedstocks
can be processed into fuel. These include vegetable
oil, animal fat, and waste cooking oil. Rapeseed
and soybean oils are the most commonly used
raw materials for biodiesel fuel. Soybean oil alone
accounted for about 65% of the U.S. production
in 2013 [76]. The reutilization of waste cooking
oil has attracted attention, because it can elimi-
nate disposal problems [77,78]. Waste cooking oil
derived biodiesel showed the lowest WTW GHG
emissions, with around 82% GHG emissions saving
compared to conventional diesel [24]. Preferences
of feedstocks differ in different countries, based on
regional production cost, environmental impact,
and agricultural strategies.
The focus of second generation biodiesel at first
stage is the use of agricultural residues. Successful
examples of sustainable commercial biodiesel pro-
duction from residues currently exist in developed
countries such as the U.S. and Europe [79]. Some
prospective feedstocks like algae show technical
promise, in terms of the efficient use of resources,
and minimal competition with the food industry.
However, such production methods remain costly,
and are not yet effectively commercialized [80].
Due to the complexity of biodiesel production
cycles, the WTW GHG emissions depend on the
feedstock and the production technology. How-
ever, current estimates show that if the production
is not on grasslands and/or deforested lands, WTW
GHG emissions can be reduced to about 50% com-
pared to conventional diesel [24]. Biodiesel only,
or blending with conventional diesel, has already
been mandated by regulation in some countries
[81]. Blends with small portions of biodiesel do not
require major modifications in the engine, because
the physiochemical characteristics of biodiesel are
generally similar to those of diesel. However, some
differences do exist, and should be considered, as
discussed in the following section.
3.2.2. Fuel properties and their effects on spray and
combustion characteristics
The LHV of biodiesel is less than that of con-
ventional diesel fuel [82], while it generally has a
higher cetane number [83]. Biodiesel also exhibits
a higher flash point compared to conventional
diesel [84], which is advantageous for fuel storage
and transportation safety. Biodiesel has higher
cloud and pour points compared to conventional
diesel, due to the higher portion of saturated fatty
acids. Cloud and pour points show the lowest tem-
peratures at which a fuel can be pumped, before
turning into a wax of crystals [85]. Higher cloud
and pour points mean disadvantages in cold start,
and represent an obstacle to the use of blends with
large biodiesel portions.
Biodiesel is denser and more viscous, and has
higher surface tension, compared to conventional
diesel. Microscopic images of biodiesel sprays
from rapeseed methyl ester (RME) revealed longer
injection delays than for conventional diesel due
to the viscosity [86]. X-ray phase-contrast imaging
(XPCI) technique was applied to analyze the
needle-lift profile and near-exit flow characteristics
[87]. The viscosity of the biodiesel slowed down the
needle speed, and decreased the flow performance
of the injector during the transient injection behav-
ior. It retarded the start of closing-transient, and
resulted in longer total injection duration under
identical energizing pulse duration.
Macroscopic spray images of waste cooking oil
(WCO) biodiesel and palm biodiesel in a constant
volume chamber revealed that the biodiesels had
longer injection delay and liquid length, while
the spray angle, area and volume were smaller
than those of conventional diesel [88]. The liquid
length increased and spray angle decreased with
the increase in mixing ratio of biodiesels [84].
The liquid length of WCO biodiesel observed
in an optical diesel engine was also greater than
that of conventional diesel [89]. The calculated
Reynolds number (Re) of WCO biodiesel was
roughly half that of diesel, due to its doubled
viscosity. The higher viscosity of biodiesel led to a
smaller spray angle, and thus poor air entrainment,
and higher SMD. However, the local equivalence
ratio, which is dominant for soot formation, was
lower than that of diesel, despite the fact men-
tioned above. The equivalence ratio along the axial
direction of biodiesel spray showed leaner value
compared to that of diesel, due to the presence
of inherent oxygen atoms in the fuel molecule
[90].
The lower LHV of biodiesel may become an
obstacle to achieving maximum torque under
full load condition. Tests performed with various
soybean biodiesel blends from B10 to B100 showed
that the average maximum brake torques decreased
by 1.57%–4.7% [91]. The indicated specific fuel
consumption (ISFC) of the biodiesel was up to
15% higher than that of the diesel fuel, solely due
 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
Fig. 6. Average emission impacts of biodiesel fuels in
compression ignition engines [96].
to the lower LHV in mass basis, despite the fact of
unaffected engine efficiency [92].
It is generally accepted that the start of combus-
tion (SOC) for pure biodiesel or biodiesel blends
is advanced compared to conventional diesel
fuel, due to their higher cetane number [93,94].
However, a longer ignition delay was observed
with WCO biodiesel when the cetane number
difference between the fuel and conventional diesel
was less than 1 [95]. The poor evaporation charac-
teristics hindered the early formation of ignitable
air–fuel mixture. As was mentioned previously,
the measurement process of cetane number in a
CFR engine means that discrepancies in ignition
delays are to be expected for modern FIE system.
Measured ignition delays of fuels of similar cetane
number can be influenced by the physical delay,
which the engine FIE system can easily influence.
The peak heat release rate of biodiesel blends was
also lower than that of diesel fuel at higher engine
load. This result was also attributed to the poor
atomization and evaporation of biodiesel blends.
The U.S. EPA produced a review of published
biodiesel emissions data for heavy-duty engines.
Figure 6 summarizes the overall results for CO,
HC, NOx and PM emissions [96]. As the biodiesel
concentration in diesel increased, a clear trend of
reduced CO and HC emissions was observed which
means higher combustion efficiency. These results
were due to the oxygenated nature of biodiesel,
where more oxygen was available for burning, and
for reducing the emissions in the exhaust [97]. It
also helped to promote stable and complete com-
bustion by delivering oxygen to the pyrolysis zone.
The oxygen content can reduce locally over-rich
regions and limit primary particle formation, as
was mentioned previously in regard to fuel spray.
Further, the kinetic work of Westbrook et al.
showed that the strong oxygen and carbon bond
remained intact, preventing it from becoming
available for soot production [98]. The C atoms
with bonded O atoms in fuel molecules could not
contribute to soot production. An investigation on
3403
the morphological and compositional character-
istics of WCO biodiesel in a compression ignition
engine was conducted through transmission
electron microscopy (TEM), thermogravimetric
analysis (TGA) and elemental analysis (EA) [99].
Figure 7 shows the comparison of primary particle
size from WCO biodiesel and diesel soot. The soot
particles from the biodiesel were composed of
smaller primary particles than those from conven-
tional diesel. Both TGA and EA results showed
that the biodiesel soot contained more soluble
organic fraction (SOF) than diesel, by showing a
faster oxidation process, and less carbonaceous
species. While WCO biodiesel soot particles con-
tained a large amount of oxygen content up to
9.44%, diesel soot showed only 3.09% of oxygen
content.
Various studies have attempted to explain the
increase in NOx emissions when using biodiesel
fuels [100]. One persuasive explanation is the
increased flame temperature with biodiesel, caused
by a reduction in the heat dissipation through
radiation as a consequence of the lower amount
of soot emitted [101]. There is no significant dif-
ference in the calculated stoichiometric adiabatic
flame temperature between biodiesel and diesel
[102]. However, the actual flame temperatures
can be affected by soot emissions, because radia-
tion from hot PM as a dominant source of heat
transfer from the in-cylinder gases, and this can
be responsible for significant cooling of sooty
gas jet flames [103]. It was discovered that, the
radiative cooling by PM reduced flame tempera-
ture by 25–50 K, corresponding to an estimated
NOx reduction of 12%–25% [104]. Therefore, the
combustion of biodiesel could be maintained at
elevated in-cylinder temperature, due to the lower
PM emissions, and finally, higher NOx emissions.
3.3. Di-methyl ether
3.3.1. Background, production, and supply
DME has been used for decades as an aerosol
propellant in the personal care industry. It has
gained attention as a clean alternative fuel to LPG,
diesel and gasoline, with physical properties very
similar to those of LPG [105]. It offers remarkable
potential not only as an automotive fuel, but also
for electric power generation, and in domestic
applications such as heating and cooking. DME,
which also is expected to be produced from biomass
resources in the future, is attractive as a solution for
energy sustainability and environmental problems
[105].
DME can be derived from many sources, in-
cluding fossil fuels (natural gas and coal) and
renewable materials (biomass, waste and agricul-
tural products). Two methods exist for producing
DME as a fuel in large quantity with relatively low
cost. The methods include (1) a two-step process
(indirect method) of adding a dehydration step to
3404 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
Fig. 7. (a) Transmission electron microscopy image of WCO biodiesel soot particles and (b) comparison of primary par-
ticle size between WCO biodiesel and diesel [99].
the latter part of industrial methanol processing,
and (2) a one-step process (direct method) of
synthesizing DME directly from synthesis gas (H2 ,
CO). Some recent developments in the synthesis
methods of DME, conventional processes and
innovative technologies in reactor design, and em-
ployed catalysts allowed the efficient production
of DME [106]. Overall, the production amount is
small, at about 150,000 tons worldwide [105].
Although the well-to-tank (WTT) efficiency for
DME is low, due to the remotely located gas sites
and wasted heat from fuel production, the over-
all WTW efficiency is generally comparable to that
of LPG and CNG fueled vehicles, ranging around
18%–19% (while that for diesel fuel is 25.7%)
[106,107]. The WTW CO2 emissions (including all
CO2 from production to combustion) for fossil fuel
derived DME were also comparable to those of
diesel and CNG fuel, while the DME derived from
bio-feedstock was nearly 1/5 that of diesel fuel.
However, the concerns of engine hardware imple-
mentation cost dedicated to DME fuel can be an
issue due to some properties of DME, such as low
lubricity, and high bulk modulus of elasticity.
3.3.2. Fuel properties and their effects on spray and
combustion
DME is in gas phase at standard temperature
and pressure. However, it changes to liquid when
subjected to modest pressure (>6 MPa) or cooling.
This easy liquefaction makes DME easy to trans-
port and store. This and other properties, including
high oxygen content, and lack of sulfur or other
noxious compounds, make DME a versatile and
promising solution in the mixture of clean renew-
able and low-carbon fuels under consideration
worldwide.
Table 6 lists some important properties of
DME [106]. The cetane number of DME is higher
than that of diesel, which makes it a promising
alternative fuel for CI engines. Attempts to develop
DME dedicated fuel injection equipment were
reported in the late 1990s, taking the injection
strategies of DME into consideration [108]. The
bulk modulus of elasticity for typical hydrocarbon
fuel is 2–6 times greater than that for DME. This
indicates greater compression work for DME is
required. It was reported that the compression
work for DME was greater than that for dode-
cane by a factor of 4 at 50 °C in a closed system
[108].
The low vapor pressure at 298 K is advan-
tageous in liquefying the fuel for storage or
transportation; however, at modestly higher tem-
peratures of around 80–100 °C (or 353–373 K)
during engine steady operation, DME evaporates.
Malfunction of the piezo injector may occur when
the backpressure of the DME fuel line is not high
enough, since the hydraulic coupler will not work
properly if it is filled with gas phase fuel [109].
The spray structure of DME was categorized
with respect to three different ambient pressure
conditions, namely below saturated vapor pressure
of DME (or at atmospheric condition), beyond
critical pressure of DME, and finally, in between
these pressure ranges [106]. Below the saturated
vapor pressure at around 0.6 MPa, the sprays
exhibited the flash boiling phenomenon. When the
fuel is injected at ambient pressure in between the
saturated vapor pressure and critical pressure (5.37
MPa), the behavior of the DME spray is similar
 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413 3405

Table 6
Comparison of alternative fuels for CI engines and commercial diesel fuel.
Item Diesel fuel JP-8 [124] Biodiesel DME Naphtha
Chemical structure – – – CH3 –O–CH3
Cetane number 52.8 [124] 45 46–64 [83] >55 31–41 [147]
Octane number [RON] – – – – 65.6
Liquid density (kg/m3 ) 0.82 0.79 0.70–0.89 [82] 0.667 0.758
Low heating value (MJ/kg) 42.5 43.4 41–42 [82] 27.6 43.99
Kinetic viscosity (at 313 K) (cSt) 2.6 1.2 4.5 <0.1 –
Auto-ignition temperature (K) 508 483 – 523 498
Boiling point at 1 atm (K) 450–643 450–523 588–623 248.1 299.7–421.9
Vapor pressure (at 298 K) (kPa) 10 0.20 0.27 530 75.8–89.6

Fig. 8. Microscopic images of (A) DME, and (B) diesel at atmospheric chamber pressure and 55 MPa injection pressure,
of (a) 0.03 ms before start of injection (SOI); (b) 0.1 ms after SOI; (c) 0.5 ms after SOI; and (d) 0.7 ms after SOI [9].

to that of diesel in terms of general shape (with
slightly increased spray angle). The spray finally
turns into a miscible turbulent jet when the am-
bient pressure is beyond the critical pressure. The
transition from normal spray to miscible turbulent
jet may be encountered during spray development.
The in-cylinder pressure of modern diesel engines
lies within the range of 4–6 MPa prior to the
combustion. It is expected that the spray of DME
in engines would be exposed to conditions in the
vicinity of its critical pressure.
The microscopic images of Fig. 8 show the
spray development characteristics of DME and
diesel. The spray was injected into atmospheric
condition with an injection pressure of 55 MPa
for both fuels. Figure 8(A) (a) shows the DME
vapor appeared prior to a gushing liquid phase.
Afterward, DME rapidly spread in both the lon-
gitudinal and axial directions, breaking up into
small droplets, while the diesel spray maintained
a narrow edge of the spray boundary, with longer
elapsed time for break-up [9]. The superior break-
up process along with the fast evaporation of
the fuel resulted in enhanced atomization. The
results of spray velocity and the droplet SMD
distribution measurements through PDPA sup-
ported the better atomization observed for DME
[110].
Cavitation may occur inside the nozzle for
DME fuel, due to the low boiling temperature
and low vapor pressure. An investigation of the
effects of nozzle-hole shape indicated that the
tapered nozzle, which has a K-factor of 4, showed
advantages over the straight nozzle [111]. Cavita-
tion restricts the fuel flow rate across the nozzle
regardless of the injection pressure, which is highly
undesirable for maximum performance of the
engine. Therefore, this suggests that DME fuel
requires increasing or optimizing the nozzle hole to
avoid such cavitation. In addition, the low LHV of
DME also yields maximum performance penalties
and greater fuel consumption on a mass basis,
compared to that of conventional diesel.
The potential of smoke-less combustion with
DME fuel was demonstrated in the early 1990s
[112]. The soot formed in DME combustion is near
zero, due to its oxygen content of 35%, and lack
of carbon–carbon (C–C) bonds [113,114]. Even
3406 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
though from a physical property point of view, its
superior evaporation characteristics can also en-
hance mixture formation with the surrounding air
and contribute to smoke reduction (similar to that
discussed for LPG), from a chemical perspective,
the literature regarded the absence of C–C bond as
the dominant reason for low smoke emission. This
implies that DME-fueled engines would not require
particulate filters in the after-treatment system.
Some controversial results exist for NOx emis-
sion when dealing with DME combustion. The
shorter ignition delay of DME, which contributed
to a lesser amount of pre-mixed burn phase, was
one of the reasons for lowered NOx emission with
DME, compared to that with diesel [115]. Al-
though DME is an oxygenated fuel, the release
of free oxygen was found to be limited and in-
significant to NOx formation through chemical ki-
netics calculations [116]. The higher specific heat
capacity of the DME, in addition to the slower
rate of energy release due to the lower LHV than
that of diesel, can contribute to the lowered peak
adiabatic flame temperature and lesser amount of
NOx formation compared to that of diesel [117].
Higher NOx emission can be formed when the in-
jection timing of the DME is advanced. The faster
initiation of the heat release yields faster temper-
ature build-up process inside the cylinder, where
the rest of the fuel may burn in higher tempera-
ture conditions, compared to the situations with
diesel. Such high temperature environment caused
by faster heat release resulted in higher NOx forma-
tion [118].
3.4. JP-8
3.4.1. Background, production, and supply
The North Atlantic Treaty Organization
(NATO) nations established a policy after World
War II that unified all petroleum fuels used in the
military field into a single fuel, named as the single
fuel concept (SFC) [119,120]. The purpose of the
SFC is to simplify the logistics supply chain for
petroleum fuels in the military field. The move
toward the use of a single military fuel on the
battlefield began in 1970, when many NATO Air
Forces agreed to replace the fuel previously used,
which is known as F-40 (JP-4), with a safer and
less flammable fuel known as F-34 (JP-8) [120].
This movement also resulted in many studies on
the effect of JP-8 in diesel engines [121,122].
JP-8 is a commercial kerosene-type jet fuel (8–
16 carbon atoms per molecule) that includes four
additives: a static dissipater additive, corrosion in-
hibitor, lubricity improver, and fuel system icing in-
hibitor. JP-8 is produced in a crude oil refining pro-
cess, which is conventional petroleum distillation.
Its boiling point lies between that of gasoline fuel
and diesel fuel [123]. The price of JP-8 is cheaper
than that of diesel fuel, because JP-8 does not re-
quire any processes to enhance its cetane number.
3.4.2. Fuel properties and their effects on spray and
combustion characteristics
Table 6 compares the commercial diesel fuel and
commercial JP-8 available in the specific regional
market [124]. The lower liquid density of JP-8
could result in lower maximum engine power, and
affect specific fuel consumption [125,126]. JP-8
also has a lower distillation temperature, which
means superior evaporation characteristics [124].
Therefore, the liquid-phase penetration of JP-8
was approximately 10%–15% less than that of
diesel fuel [124,127]. However, the vapor-phase
penetration and spreading angle of both fuels were
of similar structure, despite the faster evaporation
rate of JP-8. It was expected that there would
be similar levels of air entrainment into diesel
fuel and JP-8 fuel jets prior to the combustion
[127].
The cetane number of JP-8 ranges from 39 to
45, depending on the production region [127,128],
which is generally lower than that of diesel fuel.
The ignition delay of JP-8 (cetane number of
38) measured in a constant-volume vessel was
25%–50% higher compared to diesel fuel (cetane
number of 46), although JP-8 has a greater frac-
tion of vaporized fuel [127]. The ignition delay of
JP-8 measured in a heavy-duty diesel engine or a
passenger car diesel engine was also longer than
that of diesel [121,124,129]. The chemical delay of
the JP-8 fuel molecule was more dominant over
its physical delay. However, the ignition delay was
no longer large for JP-8 if the measured cetane
number is in a similar range to that of diesel,
which resulted in the opposite result [130]. This
is likely due to the physical ignition delay being
dominant over the chemical one, when the fuels’
cetane numbers are in a similar range.
The NOx emission characteristics varied, de-
pending on the engine operating conditions. JP-8
emitted lower NOx compared with diesel fuel, due
to lower aromatic contents and flame temperature
[125,126,130]. JP-8 showed higher NOx levels
as the engine load increased to mid-load range
[124,129], due to the more combustible mixture
of near stoichiometric condition caused by the
prolonged ignition delay and better evaporation.
The smoke emission showed a consistent trend.
The combustion with JP-8 generally emitted lower
smoke than with diesel fuel [129,130]. The faster
evaporation rate of JP-8 produced a lesser fuel-rich
region, which could promote the air–fuel mixing
process, and lower the local equivalence ratio
[121,124]. The results obtained through two-color
method revealed that the JP-8 exhibited more
uniform mixture distribution inside the cylinder
(Fig. 9) [129]. The difference in the aromatic con-
tent portion between the two fuels could also affect
the smoke emission. The fuel with higher aromatic
contents produced more polycyclic HCs [131],
known as the precursor of smoke, thereby emitting
more PM [132].
 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
Fig. 9. KL factor distribution of diesel fuel and JP-8
[129].
4. Alternative fuels for advanced engine combustion
4.1. Background
The combustion of air–fuel mixtures through
flame propagation in SI engines, or mixing con-
trolled combustion between injected fuel and air in
CI engines, are techniques that provide promising
combustion stability for engine operation with
modern engine technologies. However, each com-
bustion scheme has intrinsic shortcomings. For
example, throttling is required during the part-load
operation of SI engines to control the amount of
stoichiometric air–fuel mixture inducted into the
cylinder, and thus reduces the efficiency. The high
temperature due to nearly stoichiometric mixture
combustion also generates large quantities of NOx .
Even though CI engines operate under an overall
lean condition for most of the part-load condi-
tion, NOx is also generated near the fuel spray
periphery, where the local equivalence ratio is close
to stoichiometric. The intrinsic characteristics of
diffusion-controlled combustion involve mixing
between the fuel and air, and cause a problem of
smoke generation.
An alternative combustion process was pro-
posed, which was initially intended to combine
the advantages of both SI and CI engines. A well-
3407
Fig. 10. Operating regimes of conventional SI, CI, and
the advanced combustion pathway [133].
mixed air–fuel lean mixture that can auto-ignite
throughout the mixture (the compression ignition
of air–fuel mixture that is used in SI engines) may
yield lower NOx formation and smoke emission.
This concept eliminates locally high temperature
zones, which reduces NOx emission, and also elim-
inates the fuel-rich zones that result in soot emis-
sion, as the equivalence ratio ()—temperature (T)
map of Fig. 10 depicts [133]. The combustion path
of a diesel spray in CI engines goes through soot
and NOx formation islands, while advanced com-
bustion, namely homogeneous charge compression
ignition (HCCI) and low temperature combustion
(LTC), aims to avoid these two emission islands
to suppress the formation from the beginning.
Besides, simultaneous ignition accompanying a
rapid heat release rate plus throttle-less operation
also brings increased engine efficiency.
However, an excessively increased heat release
rate as the increase of engine load is problematic,
due to the increased mechanical stress acting upon
the engine, and undesired combustion noise. Load
extension is a major drawback of advanced com-
bustion. Various studies had been conducted to
slow down the chemical reaction rate, by creating
a thermal stratification of the charge inside the
cylinder [134–136], and implementing alternative
fuels with different fuel properties.
4.2. Advanced combustion schemes
HCCI (or controlled auto-ignition (CAI)) is
theoretically realized through the compression
ignition of a homogenous mixture of fuel and air.
A mixing process can be realized that is similar to
that of the SI engine system, or by early injection
of fuel at the very beginning of the intake or com-
pression stroke in CI engines. The engine operation
under HCCI combustion is fully determined by the
chemical reaction characteristics of the fuel itself.
3408 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
Detailed chemical kinetics of transportation fuel
can be found in the literature [137]. The reaction
pathways of typical hydrocarbon fuels, which
consist of lower temperature reaction (LTR) and
high temperature reaction (HTR) [138,139], greatly
depend on the ambient temperature and the fuel
property itself. The ignition delay and the start of
combustion in the crank angle domain depend on
the pressure, temperature and local equivalence
ratio inside the cylinder. In other words, the precise
timing of the start of combustion requires fine con-
trol of pressure, temperature, and the local equiva-
lence ratio of the mixture. This results in difficulties
in controlling the right timing for combustion ini-
tiation in advanced combustion; whereas, injection
timing or spark timing directly control the com-
bustion initiation in CI and SI engines, respectively.
Later concepts of partially-premixed charge
compression ignition (PCCI), LTC, reactivity
controlled compression ignition (RCCI), and GCI
were later suggested to realize better controllability
and wider operation window than the HCCI oper-
ation. The PCCI concept utilizes a certain amount
of EGR with moderately retarded injection tim-
ing compared with that of HCCI, but still more
advanced than that of the conventional mode. It
aims to form partially premixed mixture, so that
the mixture stratification can mitigate the abrupt
heat release. At the same time, it can also resolve
the ignition controllability issue. LTC is another
concept that utilizes a large amount of EGR,
shifting the combustion pathway further to the
lower-left corner of the –T map of Fig. 10. This
can lower the adiabatic flame temperature through
the dilution effect, and prolong ignition delay.
Thus, the local rich mixture and peak combustion
temperature can be lowered.
RCCI is another advanced combustion concept
that utilizes both a high cetane number fuel and
high octane number fuel to achieve better control-
lability, compared to other advanced combustion
schemes [140,141]. This can realize fuel reactivity
stratification, in addition to mixture stratification.
The energy ratio of these fuels in conjunction
with injection timing has the potential to meet the
requirement according to the load conditions. A
higher cetane number combination is preferable at
low load condition, and vice versa.
The GCI concept is realized through the com-
pression ignition of direct injection gasoline fuel.
One of the differences between GCI and CAI is
the location of the air–fuel mixture formation. The
CAI usually utilizes hot residual gas to achieve the
auto-ignition temperature of gasoline in modified
SI engines, while GCI achieves the auto-ignition
temperature through the compression of air in
modified CI engines with direct injection. De-
tailed information on CAI can be found in the
literature [142]. The GCI concept with gasoline
fuel is reported to be promising at mid-load range
operation, with high efficiency and low emissions
[143,144]. GCI was also found to be feasible with
alternative low-grade gasoline-like fuel, such as
naphtha. This can be considered as an option to ex-
tend the lower operation limits of the GCI concept.
4.3. Alternative fuels or fuel combinations for
advanced combustion
Advanced combustion is still under active
research. It includes the implementation of
alternative fuels of different physicochemical prop-
erties, which may be considered as appropriate
for particular modes of an advanced combustion
regime, such as naphtha in GCI, as mentioned
previously. Naphtha, a low-grade gasoline-like
fuel considered inapplicable to conventional SI
engines, will be briefly introduced as an alternative
for advanced combustion, namely GCI. This will
be followed by H2 and CNG application in RCCI
or the dual fuel concept, which will be discussed to
observe their effects on combustion.
4.4. Naphtha
4.4.1. Background, production, and supply
Studies on GCI using gasoline-like fuels with
superior resistance to auto-ignition and higher
volatility, claim that naphtha is suitable for creat-
ing well-premixed or properly stratified mixture
prior to ignition, and for reducing NOx and PM
emissions, while maintaining similar efficiency
[143,144]. Research on GCI can be carried out in
both retrofitted commercial GDI engines [145],
and retrofitted diesel engines for higher compres-
sion ratio [146,147]. Due to the high auto-ignition
resistance of gasoline at low load conditions, some
researchers suggested that the optimum fuel for
GCI engines would be a much less processed fuel,
such as naphtha.
The light fraction resulting from distillation
and boiling, roughly in the temperature range of
30–200 °C, is called naphtha. It consists of a com-
plex mixture of hydrocarbon molecules, generally
having between 5 and 11 carbon atoms. It typi-
cally constitutes 15%–30% of crude oil by weight.
The research octane number (RON) of naphtha is
around 70. Naphtha is the least processed product
during refinery, and the energy required to produce
naphtha and hence its CO2 impact will be lower,
compared to conventional gasoline or diesel. The
production cost of naphtha is much lower than
that of gasoline, with similar properties [147,148].
Moreover, light fractions like naphtha will be read-
ily available, and could be in surplus, as the demand
shifts toward heavier fuels for use in commercial
transportation [147].
4.4.2. Fuel properties and their impact on engine
combustion characteristics
Table 6 lists the major properties of naphtha
used in various studies [147]. The cetane number
 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
of the fuel lies between 31 and 41. Research was
carried out to investigate the naphtha combustion
in a CI engine with a compression ratio of 19
for stable ignition [145]. Naphtha exhibited lower
smoke levels compared to that of diesel, because
its prolonged ignition delays resulted in better
mixture formation. The NOx could be controlled
by decreasing intake O2 concentration with a
moderate amount of EGR. The smoke emission
did not deteriorate, due to the above mentioned
fuel properties [149].
Abrupt heat release from advanced combus-
tion technology was always an obstacle for load
expansion in the engine operating point. A unique
advanced combustion strategy named multiple
premixed compression ignition (MPCI) attempted
to separate the combustion event of each injected
fuel amount during multiple injections [149]. The
separation of heat release was achieved with naph-
tha as a result. However with gasoline, all injected
fuel (1st and 2nd injections) released the heat at
the same time in the vicinity of the TDC, causing
an abrupt increase of the pressure and combustion
noise. Therefore, naphtha was observed to be
adequate for such strategy, even at high loads.
4.5. Hydrogen in compression ignition engines
(advanced combustion with dual fuel concept)
The octane number and the auto-ignition
temperature for H2 are too high for it to be ig-
nited solely via compressive heating in the engine.
Therefore, apart from burning the H2 with spark
ignition, it can be used as an additive for combus-
tion improvement of the main fuel, or used with an
external ignition source (pilot injected high cetane
fuel) in CI engines.
The combustion phasing of a high cetane num-
ber fueled HCCI engine could be controlled by the
addition of hydrogen, because hydrogen has a high
octane number and high auto-ignition tempera-
ture. Chemical kinetics analysis was performed to
investigate the ignition control effect of hydrogen
on DME HCCI combustion [33,150]. It was found
that the combustion phasing was retarded, due
to the consumption of OH radicals by hydrogen
during low temperature oxidation of DME. The
effect of hydrogen enrichment on a diesel HCCI
engine at different compression ratios were also
characterized [34,35]. With hydrogen enrichment,
the diesel HCCI engine could be operated at a
higher compression ratio due to the retarding
effect of combustion phasing, which resulted in a
higher power output.
4.6. Natural gas in compression-ignition engines
(advanced combustion with dual fuel concept)
Natural gas is generally injected in the man-
ifold, where the engine can inhale a well-mixed
homogeneous mixture of natural gas and air. The
3409
pilot fuel (a high cetane number fuel) is directly
injected at the vicinity of TDC. The pilot fuel is
first auto-ignited, followed by auto-ignition of the
surrounding natural gas–air mixture. The termi-
nology ‘dual-fuel’ or ‘RCCI’ may differ, according
to the injection timing of the high cetane fuel.
The objective of combustion, the method of high
cetane number fuel introduction, and the ratio
between high cetane number and high octane num-
ber fuel are some of the features to distinguish the
dual-fuel mode combustion from RCCI. However,
no concrete identifications to distinguish between
these two have yet been defined.
The pilot fuel ignition delay was prolonged
slightly during dual-fuel mode, because of the re-
activity of the high octane number fuel. The reason
for such phenomenon is similar to that explained
with H2 dual fuel combustion, where radicals
are partially consumed by the low reactivity fuel.
Dual-fueled engines are noisier than conventional
CI engines, due to spontaneous auto-ignition of
the homogeneous mixture, followed by the ignition
of the pilot fuel. In terms of the exhaust emissions,
dual fuel combustion resulted in reduced NOx ,
mainly due to the lean operation [151]. However,
CO and HC at low loads may be problematic,
due to incomplete combustion, resulting from
lower charge temperature. Soot emission was
significantly reduced compared to normal diesel
operation, owing to the homogenous mixture.
5. Conclusions
The major portion (up to 90%) of propulsion
for transportation will still rely on the internal
combustion engine (ICE), even up to 2040 [6].
Indeed, the global demand for transportation
fuels is expected to grow continuously at between
1.2% and 1.4% per annum [7]. Discussions on
the current state and the future of transportation
fuel are important. An early discussion on future
options for alternative fuels already took place
during the 1980s [1]. These attempts were made not
only to seek alternative solutions to energy security
and sustainability, but also to seek benefits that
alternative fuels can provide to engine efficiency
improvement and emission reduction.
Gradual penetration of low-carbon fuels, such
as biofuels and natural gas, is expected to con-
tribute to lowering greenhouse gas (GHG) emis-
sion in the long term. Advanced technologies in the
manufacturing process of biofuels has the potential
to decrease GHG emission compared to conven-
tional fuels on a well-to-wheel (WTW) basis. GHG
emission is expected to increase short term, due
to the increased demand for energy consumption.
However, it can be reduced again when intensive
use of biofuels and engine efficiency improvement
can be realized, as was introduced as the ‘2DS sce-
nario’ by International Energy Agency (IEA). [3].
3410 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
Gasoline Increased torque
and eff. by
Torque
increased octane
number
(e.g. ethanol)
Lower PM and PN
formation with better
mixture formation
(e.g.LPG)
Speed
Diesel Mixture formation
Torque
with rapid
evaporation
(e.g. DME)
EGR rate increase
with the aid of
oxgenated fuel
(e.g. biodiesel / DME)
Speed
Fig. 11. Challenges for gasoline and diesel engines.
Some of the typical examples of improving
the engine efficiency and emissions characteristics
include the utilization of high octane number fuel
to increase the efficiency of the spark ignition (SI)
engines by increasing the compression ratio; or
the realization of smoke reduction in compression
ignition (CI) engines with oxygenated fuel such as
biodiesel or di-methyl ether (DME). Some novel
fuels, such as low octane number fuel naphtha,
were also introduced, due to the emergence of
advanced combustion concepts, which aims at
higher efficiency and clean combustion. Advanced
combustion concepts greatly rely on the physic-
ochemical properties of fuels. Detailed chemical
kinetics of transportation fuels can be found in
the literature [1,137-139]. Such knowledge can
contribute to the direction of future alternative
fuel designs or fuel additive designs. The distinct
properties of alternative fuels can widen the poten-
tial for additional efficiency improvement of the
engine, where conventional fuels would otherwise
have faced limitations. Various obstacles hinder
the growth of alternative fuels, including technical
and social issues; and in the next few decades, they
cannot replace conventional fuels to any substan-
tial extent [7]. However, attempts can be made to
realize the advantages of specific fuels in a different
manner when combined with conventional fuels.
Figure 11 schematically depicts the possible
improvements with alternative fuels when imple-
mented in conventional gasoline and diesel engines.
These can be fulfilled by gradually increasing the
portion of the alternative fuels. For example,
ethanol greatly improves the octane number of
the fuel; however total substitution is limited,
due to its lower heating value (LHV). Therefore,
blending of ethanol as an anti-knocking additive
(where E85 is already widely distributed in some
particular markets) may be one of the solutions
for improving the octane number of the fuel, thus
increasing the engine efficiency. An example for
CI is that implementation of exhaust gas recir-
culation (EGR) affects the trade-off relationship
between NOx and smoke emission. Increasing the
portion of the oxygenated fuels, such as biodiesel
or DME, may result in tolerating a larger amount
of EGR for NOx reduction, without deteriorat-
ing the fuel efficiency or smoke emission. Even
though alternative fuels are not advantageous in
every aspect, advantageous attributes can provide
guidance on what the interaction between fuel
properties and combustion might be like in the
future.
The future trends of engine development can-
not stand against the development of the fuel
industry, nor do they oppose it. Automotive man-
ufacturers and the refinery industry should face
the challenge together for more efficient and clean
combustion engines in the future. As in the past,
the improvement of the thermal efficiency of SI
engines cannot be separated from the improvement
of fuel quality (namely the octane number) by the
refinery industry. The advent of sophisticated high
pressure common rail direct injection (CRDI) in
CI engine would not have been developed without
drastically improved qualities of diesel fuel. The
increasing energy demand of petroleum-based
liquid fuel and alternative fuels, plus the inequality
of demand among various fuels would eventually
yield friendly cooperation between the automotive
and refinery industries.
Acknowledgments
The authors greatly appreciate the invitation
from the Combustion Institute to prepare this
manuscript. The authors would like to thank all
relevant personnel for organizing the Combustion
Symposium in Seoul, and supporting the author
for the speech of Topical Review and the paper.
The contribution of the members of Engine Lab-
oratory in KAIST and sponsoring bodies is also
appreciated, which enabled the series of research
on alternative fuel applications for the last decade.
References
[1] J.B. Heywood, Prog. Energy Combust. Sci. 7 (1981)
155–184.
 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
[2] U.S. Energy Information Administration, In-
ternational Energy Outlook 2013, U.S. Depart-
ment of Energy, Washington, DC, July 2013
DOE/EIA-0484.
[3] International Energy Agency (IEA), Energy Tech-
nology Perspective (ETP) 2014, 2014, ISBN 978-92-
64-20800-1.
[4] International Energy Agency (IEA), Transport,
Energy and CO2, 2009, ISBN 978-92-64-07316-
6.
[5] International Energy Agency (IEA), Energy Tech-
nology Perspective (ETP) 2012, 2012, ISBN 978-92-
64-17488-7.
[6] The Outlook for Energy: A View to 2040, Exxon-
mobil.
[7] G.T. Kalghatgi, Proc. Combust. Inst. 35 (2015)
101–115.
[8] G.T. Kalghatgi, Int. J. Eng. Res. 15 (2014) 1–17.
[9] J. Yu, C. Bae, Proc. Inst. Mech. Eng. Part D: J. Au-
tomobile Eng. 217 (2003) 1135–1144.
[10] E.A. Frame, R.A. Alvarez, M.G. Blanks, et al., SAE
Technical Paper 2004-01-2961, 2004.
[11] U.S. Department of Energy, Clean Cities Alterna-
tive Fuel Price Report, January 2014.
[12] K. Reif, Gasoline Engine Management: Systems
and Components, Springer Vieweg, 2015 ISBN
978-3-648-03964-6.
[13] B.M. Masum, H.H. Masjuki, M.A. Kalam,
I.M. Rizwanul Fattah, Renew. Sustainable Energy
Rev. 24 (2013) 209–222.
[14] A.C. Hansen, Q. Zhang, P.W. Lyne, Bioresour. Tech-
nol. 96 (2005) 277–285.
[15] U.S. Energy information administration (EIA), In-
ternational Energy Statistics, 2012.
[16] British Petroleum, BP Statistical Review of World
Energy, 2015.
[17] U.S. Energy Information Administration (EIA),
Annual Energy Outlook 2014 with Projections to
2040, DOE/EIA-0383(2014), April 2014.
[18] A.K. Agarwal, Prog. Energy Combust. Sci. 33
(2007) 233–271.
[19] E.D. Larson, Energy Sustainable Dev 10 (2006)
109–126.
[20] SAE International, Alternative Automotive Fuels
J1297_200707, 2007.
[21] J.B. Heywood, Internal Combustion Engine Funda-
mentals, McGraw-Hill, New York, 1988.
[22] Robert Bosch GmbH, Gasoline-Engine Manage-
ment: System and Components, third edition, John
Wiley & Sons Ltd, West Sussex, England, 2006.
[23] J.C. Livengood, P.C. Wu, Symp. Int. Combust. 5
(1955) 347–356.
[24] R. Edwards, H. Hass, J. Larive, L. Lonza, H. Maas,
D. Rickeard, Well-to-Wheels Report Version 4.a,
JEC Well-to-Wheels Analysis, European Commis-
sion Joint Research CentreInstitute for Energy and
Transport, 2014 ISBN 978-92-79-33887-8.
[25] A. Burnham, J. Han, C.E. Clark, M. Wang,
J.B. Dunn, I. Palou-Rivera, Environ. Sci. Technol.
46 (2012) 619–627.
[26] M.U. Aslam, H.H. Masjuki, M.A. Kalam, H. Ab-
desselam, T.M.I. Mahlia, M.A. Amalina, Fuel 85
(2006) 717–724.
[27] S. Verhelst, T. Wallner, Prog. Energy Combust. Sci.
35 (2009) 490–527.
[28] C. Gong, B. Deng, S. Wang, Y. Su, Q. Gao, X. Liu,
Energy Fuels 22 (2008) 3779–3784.
3411
[29] T. Korakianitis, A.M. Namasivayam, R.J. Crookes,
Prog. Energy Combust. Sci. 37 (2011) 89–112.
[30] P. Mello, G. Pelliza, R. Cataluna, R. da Silva, Fuel
85 (2006) 2180–2186.
[31] H.M. Cho, B.Q. He, Energy Convers. Manage. 48
(2007) 608–618.
[32] H.L. Husted, G. Karl, S. Schilling, C. Weber, 23rd
Aachen Colloquium Automobile and Engine Tech-
nology, 2014.
[33] J. Jeon, C. Bae, Int. J. Hydrogen Energy 38 (2013)
265–273.
[34] H. Guo, V. Hosseini, W.S. Neill, W.L. Chippior,
C.E. Dumitrescu, Int. J. Hydrogen Energy 36 (2011)
13820–13830.
[35] H. Guo, W.S. Neill, Int. J Hydrogen Energy 38
(2013) 11429–11437.
[36] S. Niaz, T. Manzoor, A.H. Pandith, Renewable Sus-
tainable Energy Rev 50 (2015) 457–469.
[37] C.M. White, R.R. Steeper, A.E. Lutz, Int. J Hydro-
gen Energy 31 (2006) 1292–1305.
[38] S. Verhelst, T. Wallner, Prog. Energy Combust. Sci.
35 (2009) 490–527.
[39] European LPG Association, Autogas in Europe,
the Sustainable Alternative: An LPG Industry
Roadmap, 2009.
[40] L. Raslavičius, A. Keršys, S. Mockus, N. Keršienė,
M. Starevičius, Renewable Sustainable Energy Rev
32 (2014) 513–525.
[41] K.J. Morganti, T.M. Foong, M.J. Brear, G. da Silva,
Y. Yang, F.L. Dryer, Fuel 108 (2013) 797–811.
[42] S. Lawankar, SAE Technical Paper 2013-01-2889,
2013.
[43] P. Baker, H. Watson, SAE Technical Paper 2005-01-
0246, 2005.
[44] K. Kim, Y. Kim, K. Lee, K. Lee, Atomization Spray
20 (2010) 553–564.
[45] E. Lee, J. Park, K.Y. Huh, J. Choi, C. Bae, SAE
Technical Paper 2003-01-0636, 2003.
[46] P. Mesman, B. Veenhuizen, SAE Technical Paper
2009-01-1834, 2009.
[47] J. Ariztegui, J. Gutierrez, A.A. Fürhapter,
H. Friedl, MTZ 76 (2015) 10–14.
[48] C. Park, Y. Park, S. Oh, et al., SAE Technical Paper
2013-01-1323, 2013.
[49] F. Zhao, M.C. Lai, D.L. Harrington, Prog. Energy
Combust. Sci. 25 (1999) 437–562.
[50] X. Zhen, Y. Wang, Renewable Sustainable Energy
Rev 52 (2015) 477–493.
[51] R.C. Costa, J.R. Sodre, Fuel 89 (2010) 287–293.
[52] R.K. Niven, Renewable Sustainable Energy Rev 9
(2005) 535–555.
[53] S. Mani Sarathy, P. Oßwald, N. Hansen,
K. Kohse-Höinghaus, Prog. Energy Combust.
Sci. 44 (2014) 40–102.
[54] R.L. Speth, E.W. Chow, R. Malina, S.T.H. Barett,
J.B. Heywood, W.H. Environ, Sci. Technol. 48
(2014) 6561–6568.
[55] K. Okamoto, T. Ichikawa, K. Saitoh, K. Oyama, K.
Hiraya, T. Urushihara, SAE Technical paper 2003-
01-1804, 2003.
[56] G.T. Kalghatgi, K. Nakata, K. Mogi, SAE Techni-
cal Paper 2005-01-0244, 2005.
[57] R.D. Reitz, F.V. Bracco, Phys. Fluid 25 (1982)
1730–1742.
[58] P.G. Aleiferis, J. Serras-Pereira, Z. van Romunde,
J. Caine, M. Wirth, Combust. Flame 157 (2010)
735–756.
3412 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
[59] H. Oh, C. Bae, K. Min, SAE Int. J. Engine 119
(2010) 213–222.
[60] D. Han, R.R. Steeper, SAE Technical Paper 2002-
01-0837, 2002.
[61] S. Park, H. Kim, H. Suh, C. Lee, Int. J. Heat Fluid
Flow 30 (2009) 1183–1192.
[62] Y. Huang, S. Huang, R. Huang, G. Hong, Energy
Convers. Manage. 108 (2016) 68–77.
[63] L. Chen, R. Stone, D. Richardson, Fuel 96 (2012)
120–130.
[64] J. Storey, T. Barone, K. Norman, S. Lewis, SAE Int.
J. Fuels Lubr. 3 (2010) 650–659.
[65] Robert Bosch GmbH, Diesel-Engine Management:
System and Components, fourth ed., Bently Publish-
ers, Cambridge, MA, USA, 2005.
[66] P. Richards, Automotive Fuels Reference Book,
15096, third ed., SAE International, Warrendale,
PA, 2014.
[67] M.P.B Musculus, P.C. Miles, L.M. Pickett, Prog.
Energy Combust. Sci. 39 (2013) 246–283.
[68] J.E. Dec, SAE Technical Paper 970873, 1997.
[69] C.A. Idicheria, L.M. Pickett, SAE Technical Paper
2006-01-3434, 2006.
[70] L.M. Pickett, D.L. Siebers, Non-Sooting, SAE
Technical Paper 2004-01-1399, 2004.
[71] L.M. Pickett, J. Manin, C.L. Genzale, D.L. Siebers,
M.P.B. Musculus, C.A. Idicheria, SAE Int. J. En-
gines 4 (2011) 764–799.
[72] D.L. Siebers, B.S. Higgins. SAE Technical Paper
2011-01-0530, 2011.
[73] J.E. Dec, R.E. Canaan, SAE Technical Paper,
980147, 1998.
[74] S.L. Plee, T. Ahmad, J.P. Myers, in: 19th In-
ternational Symposium on Combustion, 1982,
pp. 1495–1502.
[75] S. Kook, C. Bae, P.C. Miles, D. Choi, L.M. Pickett,
SAE Int. J. Engines 114 (2005) 1575–1595.
[76] January 2016, U.S. Department of Energy.
[77] W. Ghazali, R. Mamat, H.H. Masjuki, G. Na-
jafi, Renewable Sustainable Energy Rev 51 (2015)
585–602.
[78] M.G. Varanda, G. Pinto, F. Martins, Fuel Process.
Technol. 92 (2011) 1087–1094.
[79] R. Albrecht, J. Robinson, Bioenergy, N.B.B.,
USA/Academic Press, Boston, 2015„ pp. 561–
563.
[80] Y. Chisti, Trends Biotechnol 26 (2008) 126–131.
[81] A. Eisentraut, International Energy Agency (2010).
[82] L.E. Oliveira, M. DaSilva, in: International Con-
ference on Renewable Energies and Power Quality
(ICREPQ’13 ) Bilbao, Spain, 2013.
[83] A. Gopinath, S. Puhan, G. Nagarajan, Int. J. Energy
Environ. 1 (2003) 295–306.
[84] N. Carareto, C. Kimura, E. Oliveira, M. Costa,
A. Meirelles, Fuel 96 (2012) 319–326.
[85] I. Lee, L. Johnson, E. Hammond, J. Am. Oil Chem.
Soc. 72 (1995) 1155–1160.
[86] T. Shoba, C. Crua, M. Heikal, M. Gold, in: 24th
European conference on Liquid Atomization and
Spray Systems, Estoril, Portugal, September 5-7,
2011.
[87] S. Moon, T. Tsujimura, Y. Gao, S. Park, J. Wang,
N. Kurimoto, Y. Nishijima, M. Oguma, Int. J. En-
gine Res. 15 (2014) 504–518.
[88] X. Wang, Z. Huang, O.A. Kuti, W. Zhang,
K. Nishida, Int. J. Heat Fluid Flow 31 (2010)
659–666.
[89] J. Hwang, Y. Jung, C. Bae, Int. J. Engine Res. 16
(2015) 664–679.
[90] B. Mohan, W. Yang, K.L. Tay, W. Yu, Energy Con-
vers. Manage. 88 (2014) 622–632.
[91] O. Ozener, L. Yuksek, A.T. Ergenc, M. Ozkan, Fuel
115 (2014) 875–883.
[92] M. Lapuerta, J. Rodríguez-Fernández, J.R.
Agudelo, Bioresour. Technol. 99 (2008) 731–740.
[93] A. Senatore, M. Cardone, V. Rocco, M.V. Prati,
SAE Technical Paper 2000-01-0691, 2000.
[94] X. Zhang, G. Gao, L. Li, Z. Wu, Z. Hu, J. Deng,
SAE Technical Paper 2008-01-1832, 2008.
[95] J. Hwang, D. Qi, Y. Jung, C. Bae, Renewable Energy
63 (2014) 9–17.
[96] United States Environmental Protection Agency
(EPA), Comprehensive Analysis of Biodiesel Im-
pacts on Exhaust Emissions, EPA420-02-001, 2002.
[97] S. Kalligeros, F. Zannikos, S. Stournas, et al.,
Biomass Bioenergy 24 (2003) 141–149.
[98] C.K. Westbrook, W.J. Pitz, H.J. Curran, J. Phys.
Chem. A 110 (2006) 6912–6922.
[99] J. Hwang, Y. Jung, C. Bae, SAE Int. J. Fuels Lubr. 8
(2015) 290–297.
[100] A.K. Agarwal, L.M. Das, J. Eng. Gas Turbines
Power 123 (2001) 440–447.
[101] M. Lapuerta, O. Armas, J. Rodriguez-Fernan-
dez, Prog. Energy Combust. Sci. 34 (2008) 198–
223.
[102] A.S. Cheng, A. Upatnieks, C.J. Mueller, Int. J. En-
gine Res. 7 (2006) 297–318.
[103] S.R. Turns, Prog. Energy Combust. Sci. 21 (1995)
361–385.
[104] M.P. Musculus, SAE Technical Paper 2005-01-
0925, 2005.
[105] DME Handbook, Japan DME Forum, 2001.
[106] C. Arcoumanis, C. Bae, R. Crookes, E. Kinoshita,
Fuel 87 (2008) 1014–1030.
[107] R. Verbeek, J. van der Weide, SAE Technical Paper
971607, 1997.
[108] S.C. Sorenson, M. Glensvig, D.L. Abata, SAE
Technical Paper 981159, 1998.
[109] J. Kim, H. Park, C. Bae, in: 21st International Sym-
posium on Alcohol Fuels (ISAF), Gwangju, Korea,
March 10-13, 2015.
[110] S.H Park, C.S. Lee, Prog. Energy Combust. Sci. 39
(2013) 147–168.
[111] S. Sasaki, M. Kato, T. Yokota, M. Konno, D. Gill,
SAE Technical Paper, 2015-01-0932, 2015.
[112] S. Sorenson, S. Mikkelsen, SAE Technical Paper
950064, 1995.
[113] S.H. Park, C.S. Lee, Energy Convers. Manage. 86
(2014) 848–863.
[114] H.J. Curran, E.M. Fisher, P.A. Glaude, et al., SAE
Technical Paper 2001-01-0653, 2001.
[115] P. Kapus, H. Ofner, SAE Technical Paper 950062,
1995.
[116] H. Teng, J.C. McCandless, J.B. Schneyer, SAE Tech-
nical Paper 2001-01-0154, 2001.
[117] D. Cipolat, in: Proceedings of the Seventh Interna-
tional Conference on Energy for a Clean Environ-
ment, 2003, p. 156.
[118] S. Kajitani, C.L. Chen, M. Konno, SAE Technical
Paper 972973, 1997.
[119] Belvoir, JP-8 Single Fuel Forward Report vol.
20, 1991, U.S. Army R-D & E Centre, 449-
464.
 C. Bae, J. Kim / Proceedings of the Combustion Institute 36 (2017) 3389–3413
[120] R.G. Papagiannakis, P.N. Kotsiopoulos, D.T.
Hountalas, E. Yfantis, SAE Technical Paper
2006-01-1673, 2006.
[121] J. Lee, C. Bae, Fuel 90 (2011) 1762–1770.
[122] D.A. Kouremenos, C.D. Rakopoulos, D.T. Houn-
talas, Int. J. Energy Res. 21 (1997) 1173–1185.
[123] J. Schmitigal, J. Tebbe, JP-8 and Other Military Fu-
els (No. TARDEC-22491), Army Tank Automo-
tive Research Development and Engineering Cen-
ter, Warren, MI, 2011.
[124] Y. Park, C. Bae, C. Mounaïm-Rousselle, F. Foucher,
Int. J. Engine Res. 16 (2015) 92–103.
[125] G. Fernandes, J. Fuschetto, Z. Filipi, D. Assanis,
H. McKee, Proc. Inst. Mech. Eng. Part D: J. Auto-
mobile Eng. 221 (2007) 957–970.
[126] A.K. Pandey, M.R. Nandgaonkar, SAE Int. J. Fu-
els Lubr. 3 (2010) 238–245.
[127] L.M. Pickett, L. Hoogterp, SAE Int. J. Commer.
Veh. 1 (2008) 108–118.
[128] S. Kook, L.M. Pickett, Fuel 93 (2012) 539–548.
[129] J. Lee, H. Oh, C. Bae, Fuel 102 (2012) 264–273.
[130] J. Nargunde, C. Jayakumar, A. Sinha, K. Acharya,
W. Bryzik, N. Henein, SAE Technical Paper 2010-
01-1123, 2010.
[131] S.S. Kee, A. Mohammadi, Y. Kidoguchi, K. Miwa,
SAE Technical Paper 2005-01-2086, 2005.
[132] Y. Kidoguchi, C. Yang, K. Miwa, SAE Technical
Paper 2000-01-1851, 2000.
[133] J.E. Dec, Proc. Combust. Inst. 32 (2009) 2727–2742.
[134] S.A. Kaiser, M. Schild, C. Schulz, Proc. Combust.
Inst. 34 (2013) 2911–2919.
[135] J.E. Dec, W. Hwang, M. Sjoberg, SAE Technical Pa-
per, 2006-01-1518.
3413
[136] M. Sjoberg, J.E. Dec, N.P. Cernansky, SAE Techni-
cal Paper, 2005-01-0113.
[137] F.L. Dryer, Proc. Combust. Inst. 35 (2015) 117–144.
[138] H.J. Curran, P. Gaffuri, W.J. Pitz, C.K. Westbrook,
Combust. Flame 114 (1998) 149–177.
[139] H.J. Curran, P. Gaffuri, W.J. Pitz, C.K. Westbrook,
Combust. Flame 129 (2002) 253–280.
[140] J. Benajes, S. Molina, A. Garcia, J. Monsalve-Ser-
rano, Energy 90 (2015) 1261–1271.
[141] S.L. Kokjohn, R.M. Hanson, D.A. Splitter,
R.D. Reitz, Int. J. Engine Res. 12 (2011) 209–226.
[142] H. Zhao, Advanced Direct Injection Combustion En-
gine Technologies and Development: Gasoline and
Gas Engines, Woodhead Publishing, 2009.
[143] G.T. Kalghatgi, P. Risberg, H. Ångström, SAE
Technical Paper 2006-01-3385, 2006.
[144] V. Manente, B. Johansson, P. Tunestal, W. Cannella,
SAE Int. J. Engines 2 (2010) 71–88.
[145] J. Chang, G.T. Kalghatgi, A. Amer, Y. Viollet, SAE
Technical Paper 2012-01-0677, 2012.
[146] K. Akihama, H. Kosaka, Y. Hotta, et al., SAE Int.
J. Fuels Lubr. 1 (2008) 920–932.
[147] J. Chang, G.T. Kalghatgi, A. Amer, P. Adomeit,
H. Rohs, B. Heuser, SAE Int. J. Engines 6 (2013)
101–119.
[148] P. Borgqvist, P. Tunestal, B. Johansson, SAE Tech-
nical Paper, 2012-01-0687, 2012.
[149] B. Wang, H. Yang, S. Shuai, et al., SAE Technical
Paper 2013-01-2631, 2013.
[150] T. Shudo, H. Yamada, Int. J. Hydrogen Energy 32
(2007) 3066–3072.
[151] R.G. Papagiannakis, D.T. Hountalas, Energy Con-
vers. Manage. 45 (2004) 2971–2987.